AU1050997A - Insecticidal protein toxins from photorhabdus - Google Patents

Description

INSECTICIDAL PROTEIN TOXINS FROM PHOTORHABDUS

Field of the Invention

The present invention relates to toxins isolated from bacteria and the use of said toxins as insecticides.

Background of the Invention

Many insects are widely regarded as pests to homeowners, to picnickers, to gardeners, and to farmers and others whose investments in agricultural products are often destroyed or diminished as a result of insect damage to field crops.

Particularly in areas where the growing season is short,

significant insect damage can mean the loss of all profits to growers and a dramatic decrease in crop yield. Scarce supply of particular agricultural products invariably results in higher costs to food processors and, then, to the ultimate consumers of food plants and products derived from those plants.

Preventing insect damage to crops and flowers and

eliminating the nuisance of insect pests have typically relied on strong organic pesticides and insecticides with broad toxicities. These synthetic products have come under attack by the general population as being too harsh on the environment and on those exposed to such agents. Similarly in non-agricultural settings, homeowners would be satisfied to have insects avoid their homes or outdoor meals without needing to kill the insects. The extensive use of chemical insecticides has raised environmental and health concerns for farmers, companies that produce the insecticides, government agencies, public interest groups, and the public in general. The development of less intrusive pest management strategies has been spurred along both by societal concern for the environment and by the development of biological tools which exploit mechanisms of insect management. Biological control agents present a promising alternative to chemical insecticides.

Organisms at every evolutionary development level have devised means to enhance their own success and survival. The use of biological molecules as tools of defense and aggression is known throughout the animal and plant kingdoms. In addition, the relatively new tools of the genetic engineer allow modifications to biological insecticides to accomplish particular solutions to particular problems.

One such agent, Bacillus thuringiensis (Bt), is an effective insecticidal agent, and is widely commercially used as such. In fact, the insecticidal agent of the Bt bacterium is a protein which has such limited toxicity, it can be used on human food crops on the day of harvest. To non-targeted organisms, the Bt toxin is a digestible non-toxic protein.

Another known class of biological insect control agents are certain genera of nematodes known to be vectors of transmission for insect-killing bacterial symbionts. Nematodes containing insecticidal bacteria invade insect larvae. The bacteria then kill the larvae. The nematodes reproduce in the larval cadaver. The nematode progeny then eat the cadaver from within. The bacteria-containing nematode progeny thus produced can then invade additional larvae.

In the past, insecticidal nematodes in the Steinernema and Heterorhabditis genera were used as insect control agents.

Apparently, each genus of nematode hosts a particular species of bacterium. In nematodes of the Heterorhabditis genus, the symbiotic bacterium is Photorhabdus luminescens.

Although these nematodes are effective insect control agents, it is presently difficult, expensive, and inefficient to produce, maintain, and distribute nematodes for insect control.

It has been known in the art that one may isolate an insecticidal toxin from Photorhabdus luminescens that has activity only when injected into Lepidopteran and Coleopteran insect larvae. This has made it impossible to effectively exploit the insecticidal properties of the nematode or its bacterial symbiont. What would be useful would be a more practical, less labor-intensive wide-area delivery method of an insecticidal toxin which would retain its biological properties after delivery. It would be quite desirous to discover toxins with oral activity produced by the genus Photorhabdus. The isolation and use of these toxins are desirous due to efficacious reasons. Until applicants' discoveries, these toxins had not been isolated or characterized.

Summary of the Invention The native toxins are protein complexes that are produced and secreted by growing bacteria cells of the genus Photorhabdus, of interest are the proteins produced by the species Photorhabdus luminescens. The protein complexes, with a molecular size of approximately 1,000 kDa, can be separated by SDS-PAGE gel analysis into numerous component proteins. The toxins contain no hemolysin, lipase, type C phospholipase, or nuclease activities. The toxins exhibit significant toxicity upon exposure

administration to a number of insects.

The present invention provides an easily administered insecticidal protein as well as the expression of toxin in a heterologous system.

The present invention also provides a method for delivering insecticidal toxins that are functional active and effective against many orders of insects.

Objects, advantages, and features of the present invention will become apparent from the following specification.

Brief Description of the Drawings Fig. 1 is an illustration of a match of cloned DNA isolates used as a part of sequence genes for the toxin of the present invention.

Fig. 2 is a map of three plasmids used in the sequencing process. Fig. 3 is a map illustrating the inter-relationship of several partial DNA fragments.

Fig. 4 is an illustration of a homology analysis between the protein sequences of TcbA_ii and TcaB_ii proteins.

Fig. 5 is a phenogram of Photorhabdus strains. Relationship of Photorhabdus Strains was defined by rep-PCR.

The upper axis of Fig. 5 measures the percentage similarity of strains based on scoring of rep-PCR products (i.e., 0.0 [no similarity] to 1.0 [100% similarityl). At the right axis, the numbers and letters indicate the various strains tested; 14=W-14, Hm=Hm, H9=H9, 7=WX-7, 1=WX-1, 2=WX-2, 88=HP88, NC-1=NC-1, 4=WX-4, 9=WX-9, 8=WX-8, 10=WX-10, WIR=WIR, 3=WX-3, 11=WX-11, 5=WX-5, 6=WX-6, 12=WX-12, x14=WX-14, 15=WX-15, Hb=Hb, B2=B2, 48 through 52=ATCC 43948 through ATCC 43952. Vertical lines separating horizontal lines indicate the degree of relatedness (as read from the extrapolated intersection of the vertical line with the upper axis) between strains or groups of strains at the base of the horizontal lines (e.g., strain W-14 is approximately 60% similar to strains H9 and Hm).

Fig. 6 is an illustration of the genomic maps of the W-14 Strain.

Detailed Description of the Invention The present inventions are directed to the discovery of a unique class of insecticidal protein toxins from the genus

Photorhabdus that have oral toxicity against insects. A unique feature of Photorhabdus is its bioluminescence. Photorhabdus may be isolated from a variety of sources. One such source is nematodes, more particularly nematodes of the genus

Heterorhabdi tis . Another such source is from human clinical samples from wounds, see Farmer et al. 1989 J. Clin. Microbiol. 27 pp. 1594-1600. These saprohytic strains are deposited in the American Type Culture Collection (Rockville, MD) ATCC #s 43948, 43949, 43950, 43951, and 43952, and are incorporated herein by reference. It is possible that other sources could harbor

Photorhabdus bacteria that produce insecticidal toxins. Such sources in the environment could be either terrestrial or aquatic based. The genus Photorhabdus is taxonomically defined as a member of the Family Enterobacteriaceae, although it has certain traits atypical of this family. For example, strains of this genus are nitrate reduction negative, yellow and red pigment producing and bioluminescent. This latter trait is otherwise unknown within the Enterobacteriaceae. Photorhabdus has only recently been described as a genus separate from the Xenorhabdus (Boemare et al., 1993 Int. J. Syst. Bacteriol. 43, 249-255). This differentiation is based on DNA-DNA hybridization studies, phenotypic differences (e.g., presence (Photorhabdus) or absence (Xenorhabdus) of catalase and bioluminescence) and the Family of the nematode host (Xenorhabdus; Steinernematidae, Photorhabdus; Heterorhabditidae). Comparative, cellular fatty-acid analyses (Janse et al. 1990, Lett. Appl. Microbiol 10, 131-135; Suzuki et al. 1990, J. Gen. Appl. Microbiol., 36, 393-401) support the separation of Photorhabdus from Xenorhabdus.

In order to establish that the strain collection disclosed herein was comprised of Photorhabdus strains, the strains were characterized based on recognized traits which define

Photorhabdus and differentiate it from other Enterobacteriaceae and Xenorhabdus species. (Farmer, 1984 Bergey's Manual of

Systemic Bacteriology Vol.1 pp.510-511; Akhurst and Boemare 1988, J. Gen. Microbiol. 134 pp.1835-1845; Boemare et al. 1993 Int. J. Syst. Bacteriol. 43 pp.249-255, which are incorporated herein by reference). The traits studied were the following: gram stain negative rods, organism size, colony pigmentation, inclusion bodies, presence of catalase, ability to reduce nitrate,

bioluminescence, dye uptake, gelatin hydrolysis, growth on selective media, growth temperature, survival under anerobic conditions and motility. Fatty acid analysis was used to confirm that the strains herein all belong to the single genus

Photorhabdus.

Currently, the bacterial genus Photorhabdus is comprised of a single defined species, Photorhabdus luminescens (ATCC Type strain #29999, Poinar et al., 1977, Nematologica 23, 97-102). A variety of related strains have been described in the literature (e.g. Akhurst et al. 1988 J. Gen. Microbiol., 134, 1835-1845;

Boemare et al. 1993 Int. J. Syst. Bacteriol. 43 pp. 249-255; Putz et al. 1990, Appl. Environ. Microbiol., 56, 181-186). Numerous Photorhabdus strains have been characterized herein. Such strains are listed in Table 18 in the Examples. Because there is currently only one species (luminescens) defined within the genus Photorhabdus , the luminescens species traits were used to characterize the strains herein. As can be seen in Fig. 5, these strains are quite diverse. It is not unforeseen that in the future there may be other Photorhabdus species that will have some of the attributes of the luminescens species as well as some different characteristics that are presently not defined as a trait of Photorhabdus luminescens. However, the scope of the invention herein is to any Photorhabdus species or strains which produce proteins that have functional activity as insect control agents, regardless of other traits and characteristics.

Furthermore, as is demonstrated herein, the bacteria of the genus Photorhabdus produce proteins that have functional activity as defined herein. Of particular interest are proteins produced by the species Photorhabdus luminescens. The inventions herein should in no way be limited to the strains which are disclosed herein. These strains illustrate for the first time that proteins produced by diverse isolates of Photorhabdus are toxic upon exposure to insects. Thus, included within the inventions described herein are the strains specified herein and any mutants thereof, as well as any strains or species of the genus

Photorhabdus that have the functional activity described herein.

There are several terms that are used herein that have a particular meaning and are as follows:

By "functional activity" it is meant herein that the protein toxins function as insect control agents in that the proteins are orally active, or have a toxic effect, or are able to disrupt or deter feeding, which may or may not cause death of the insect. When an insect comes into contact with an effective amount of toxin delivered via transgenic plant expression, formulated protein compositions(s), sprayable protein composition(s), a bait matrix or other delivery system, the results are typically death of the insect, or the insects do not feed upon the source which makes the toxins available to the insects. The protein toxins discussed herein are typically referred to as "insecticides". By insecticides it is meant herein that the protein toxins have a "functional activity" as further defined herein and are used as insect control agents.

By the use of the term "oligonucleotides" it is meant a

macromolecule consisting of a short chain of nucleotides of either RNA or DNA. Such length could be at least one nucleotide, but typically are in the range of about 10 to about 12

nucleotides. The determination of the length of the

oligonucleotide is well within the skill of an artisan and should not be a limitation herein. Therefore, oligonucleotides may be less than 10 or greater than 12. By the use of the term "toxic" or "toxicity" as used herein it is meant that the toxins produced by Photorhabdus have "functional activity" as defined herein.

By the use of the term "genetic material" herein, it is meant to include all genes, nucleic acid, DNA and RNA.

Fermentation broths from selected strains reported in

Table 18 were used to determine the following: breadth of insecticidal toxin production by the Photorhabdus genus, the insecticidal spectrum of these toxins, and to provide source material to purify the toxin complexes. The strains

characterized herein have been shown to have oral toxicity against a variety of insect orders. Such insect orders include but are not limited to Coleoptera, Homoptera, Lepidoptera,

Diptera, Acarina, Hymenoptera and Dictyoptera.

As with other bacterial toxins, the rate of mutation of the bacteria in a population causes many related toxins slightly different in sequence to exist. Toxins of interest here are those which produce protein complexes toxic to a variety of insects upon exposure, as described herein. Preferably, the toxins are active against Lepidoptera, Coleoptera, Homopotera, Diptera, Hymenoptera, Dictyoptera and Acarina. The inventions herein are intended to capture the protein toxins homologous to protein toxins produced by the strains herein and any derivative By the use of the term "Photorhabdus toxin" it is meant any protein produced by a Photorhabdus microorganism strain which has functional activity against insects, where the Photorhabdus toxin could be formulated as a sprayable composition, expressed by a transgenic plant, formulated as a bait matrix, delivered via a Baculovirus, or delivered by any other applicable host or delivery system.

strains thereof, as well as any protein toxins produced by

Photorhabdus . These homologous proteins may differ in sequence, but do not differ in function from those toxins described herein. Homologous toxins are meant to include protein complexes of between 300 kDa to 2,000 kDa and are comprised of at least two (2) subunits, where a subunit is a peptide which may or may not be the same as the other subunit. Various protein subunits have been identified and are taught in the Examples herein.

Typically, the protein subunits are between about 18 kDa to about 230 kDa; between about 160 kDa to about 230 kDa; 100 kDa to 160 kDa; about 80 kDa to about 100 kDa; and about 50 kDa to about 80 kDa.

As discussed above, some Photorhabdus strains can be isolated from nematodes. Some nematodes, elongated cylindrical parasitic worms of the phylum Nematoda , have evolved an ability to exploit insect larvae as a favored growth environment. The insect larvae provide a source of food for growing nematodes and an environment in which to reproduce. One dramatic effect that follows invasion of larvae by certain nematodes is larval death. Larval death results from the presence of, in certain nematodes, bacteria that produce an insecticidal toxin which arrests larval growth and inhibits feeding activity.

Interestingly, it appears that each genus of insect

parasitic nematode hosts a particular species of bacterium, uniquely adapted for symbiotic growth with that nematode. In the interim since this research was initiated, the name of the bacterial genus Xenorhabdus was reclassified into the Xenorhabdus and the Photorhabdus . Bacteria of the genus Photorhabdus are characterized as being symbionts of Heterorhabditus nematodes while Xenorhabdus species are symbionts of the Steinernema species. This change in nomenclature is reflected in this specification, but in no way should a change in nomenclature alter the scope of the inventions described herein.

The peptides and genes that are disclosed herein are named according to the guidelines recently published in the Journal of Bacteriology "Instructions to Authors" p. i-xii (Jan. 1996), which is incorporated herein by reference. The following peptides and genes were isolated from Photorhabdus strain W-14. The sequences listed above are grouped by genomic region. The tcbA gene was expressed in E. coli as two protein fragments TcbA and TcbA_iii as illustrated in the Examples. It may be beneficial to have proteolytic clippage of some sequences to obtain the higher activity of the toxins for commercial

transgenic applications.

The toxins described herein are quite unique in that the toxins have functional activity, which is key to developing an insect management strategy. In developing an insect management strategy, it is possible to delay or circumvent the protein degradation process by injecting a protein directly into an organism, avoiding its digestive tract. In such cases, the protein administered to the organism will retain its function until it is denatured, non-specifically degraded, or eliminated by the immune system in higher organisms. Injection into insects of an insecticidal toxin has potential application only in the laboratory, and then only on large insects which are easily injected. The observation that the insecticidal protein toxins herein described exhibits their toxic activity after oral ingestion or contact with the toxins permits the development of an insect management plan based solely on the ability to

incorporate the protein toxins into the insect diet. Such a plan could result in the production of insect baits.

The Photorhabdus toxins may be administered to insects in a purified form. The toxins may also be delivered in amounts from about 1 to about 100 mg / liter of broth. This may vary upon formulation condition, conditions of the inoculum source, techniques for isolation of the toxin, and the like. The toxins may be administered as an exudate secretion or cellular protein originally expressed in a heterologous prokaryotic or eukaryotic host. Bacteria are typically the hosts in which proteins are expressed. Eukaryotic hosts could include but are not limited to plants, insects and yeast. Alternatively, the toxins may be produced in bacteria or transgenic plants in the field or in the insect by a baculovirus vector. Typically the toxins will be introduced to the insect by incorporating one or more of the toxins into the insects' feed.

Complete lethality to feeding insects is useful but is not required to achieve useful toxicity. If the insects avoid the toxin or cease feeding, that avoidance will be useful in some applications, even if the effects are sublethal. For example, if insect resistant transgenic crop plants are desired, a reluctance of insects to feed on the plants is as useful as lethal toxicity to the insects since the ultimate objective is protection of the plants rather than killing the insect.

There are many other ways in which toxins can be

incorporated into an insect's diet. As an example, it is possible to adulterate the larval food source with the toxic protein by spraying the food with a protein solution, as

disclosed herein. Alternatively, the purified protein could be genetically engineered into an otherwise harmless bacterium, which could then be grown in culture, and either applied to the food source or allowed to reside in the soil in an area in which insect eradication was desirable. Also, the protein could be genetically engineered directly into an insect food source. For instance, the major food source of many insect larvae is plant material.

By incorporating genetic material that encodes the

insecticidal properties of the Photorhabdus toxins into the genome of a plant eaten by a particular insect pest, the adult or larvae would die after consuming the food plant. Numerous members of the monocotyledonous and dictyledenous genera have been transformed. Transgenic agronmonic crops as well as fruits and vegetables are of commercial interest. Such crops include but are not limited to maize, rice, soybeans, canola, sunflower, alfalfa, sorghum, wheat, cotton, peanuts, tomatoes, potatoes, and the like. Several techniques exist for introducing foreign genetic material into plant cells, and for obtaining plants that stably maintain and express the introduced gene. Such techniques include acceleration of genetic material coated onto

microparticles directly into cells (U.S. Patents 4,945,050 to Cornell and 5,141,131 to DowElanco). Plants may be transformed using Agrobacterium technology, see U.S. Patent 5,177,010 to University of Toledo, 5,104,310 to Texas A&M, European Patent Application 0131624B1, European Patent Applications 120516, 159418B1 and 176,112 to Schilperoot, U.S. Patents 5,149,645, 5,469,976, 5,464,763 and 4,940,838 and 4,693,976 to Schilperoot, European Patent Applications 116718, 290799, 320500 all to

MaxPlanck, European Patent Applications 604662 and 627752 to Japan Tobacco, European Patent Applications 0267159, and 0292435 and U.S. Patent 5,231,019 all to Ciba Geigy, U.S. Patents

5,463,174 and 4,762,785 both to Calgene, and U.S. Patents

5,004,863 and 5,159,135 both to Agracetus. Other transformation technology includes whiskers technology, see U.S. Patents

5,302,523 and 5,464,765 both to Zeneca. Electroporation

technology has also been used to transform plants, see WO

87/06614 to Boyce Thompson Institute, 5,472,869 and 5,384,253 both to Dekalb, WO9209696 and WO9321335 both to PGS. All of these transformation patents and publications are incorporated by reference. In addition to numerous technologies for transforming plants, the type of tissue which is contacted with the foreign genes may vary as well. Such tissue would include but would not be limited to embryogenic tissue, callus tissue type I and II, hypocotyl, meristem, and the like. Almost all plant tissues may be transformed during dedifferentiation using appropriate techniques within the skill of an artisan.

Another variable is the choice of a selectable marker. The preference for a particular marker is at the discretion of the artisan, but any of the following selectable markers may be used along with any other gene not listed herein which could function as a selectable marker. Such selectable markers include but are not limited to aminoglycoside phosphotransferase gene of

transposon Tn5 (Aph II) which encodes resistance to the

antibiotics kanamycin, neomycin and G418, as well as those genes which code for resistance or tolerance to glyphosate; hygromycin; methotrexate; phosphinothricin (bialophos); imidazolinones, sulfonylureas and triazolopyrimidine herbicides, such as

chlorosulfuron; bromoxynil, dalapon and the like.

In addition to a selectable marker, it may be desirous to use a reporter gene. In some instances a reporter gene may be used without a selectable marker. Reporter genes are genes which are typically not present or expressed in the recipient organism or tissue. The reporter gene typically encodes for a protein which provides for some phenotypic change or enzymatic property. Examples of such genes are provided in K. Weising et al. Ann. Rev. Genetics, 22, 421 (1988), which is incorporated herein by reference. A preferred reporter gene is the glucuronidase (GUS) gene.

Regardless of transformation technique, the gene is preferably incorporated into a gene transfer vector adapted to express the Photorhabdus toxins in the plant cell by including in the vector a plant promoter. In addition to plant promoters, promoters from a variety of sources can be used efficiently in plant cells to express foreign genes. For example, promoters of bacterial origin, such as the octopine synthase promoter, the nopaline synthase promoter, the mannopine synthase promoter;

promoters of viral origin, such as the cauliflower mosaic virus (35S and 19S)and the like may be used. Plant promoters include, but are not limited to ribulose-1,6-bisphosphate (RUBP)

carboxylase small subunit (ssu), beta-conglycinin promoter, phaseolin promoter, ADH promoter, heat-shock promoters and tissue specific promoters. Promoters may also contain certain enhancer sequence elements that may improve the transcription efficiency. Typical enhancers include but are not limited to Adh-intron 1 and Adh-intron 6. Constitutive promoters may be used. Constitutive promoters direct continuous gene expression in all cells types and at all times (e.g., actin, ubiquitin, CaMV 35S). Tissue specific promoters are responsible for gene expression in specific cell or tissue types, such as the leaves or seeds (e.g., zein, oleosin, napin, ACP) and these promoters may also be used. Promoters may also be are active during a certain stage of the plants' development as well as active in plant tissues and organs. Examples of such promoters include but are not limited to pollen-specific, embryo specific, corn silk specific, cotton fiber specific, root specific, seed endosperm specific promoters and the like.

Under certain circumstances it may be desirable to use an inducible promoter. An inducible promoter is responsible for expression of genes in response to a specific signal, such as: physical stimulus (heat shock genes); light (RUBP carboxylase); hormone (Em); metabolites; and stress. Other desirable

transcription and translation elements that function in plants may be used. Numerous plant-specific gene transfer vectors are known to the art.

In addition, it is known that to obtain high expression of bacterial genes in plants it is preferred to reengineer the bacterial genes so that they are more efficiently expressed in the cytoplasm of plants. Maize is one such plant where it is preferred to reengineer the bacterial gene(s) prior to

transformation to increase the expression level of the toxin in the plant. One reason for the reengineering is the very low G+C content of the native bacterial gene(s) (and consequent skewing towards high A+T content). This results in the generation of sequences mimicking or duplicating plant gene control sequences that are known to be highly A+T rich. The presence of some A+T-rich sequences within the DNA of the gene(s) introduced into plants (e.g., TATA box regions normally found in gene promoters) may result in aberrant transcription of the gene(s). On the other hand, the presence of other regulatory sequences residing in the transcribed mRNA (e.g., polyadenylation signal sequences (AAUAAA), or sequences complementary to small nuclear RNAs involved in pre-mRNA splicing) may lead to RNA instability.

Therefore, one goal in the design of reengineered bacterial gene(s), more preferably referred to as plant optimized gene(s), is to generate a DNA sequence having a higher G+C content, and preferably one close to that of plant genes coding for metabolic enzymes. Another goal in the design of the plant optimized gene(s) is to generate a DNA sequence that not only has a higher G+C content, but by modifying the sequence changes, should be made so as to not hinder translation.

An example of a plant that has a high G+C content is maize.

The table below illustrates how high the G+C content is in maize. As in maize, it is thought that G+C content in other plants is also high.

For the data in Table 1, coding regions of the genes were extracted from GenBank (Release 71) entries, and base

compositions were calculated using the MacVector™ program (IBI, New Haven, CT). Intron sequences were ignored in the calculations. Group I and II storage protein gene sequences were distinguished by their marked difference in base composition.

Due to the plasticity afforded by the redundancy of the genetic code (i.e., some amino acids are specified by more than one codon), evolution of the genomes of different organisms or classes or organisms has resulted in differential usage of redundant codons. This "codon bias" is reflected in the mean base composition of protein coding regions. For example, organisms with relatively low G+C contents utilize codons having A or T in the third position of redundant codons, whereas those having higher G+C contents utilize codons having G or C in the third position. It is thought that the presence of "minor" codons within a gene's mRNA may reduce the absolute translation rate of that mRNA, especially when the relative abundance of the charged tRNA corresponding to the minor codon is low. An extension of this is that the diminution of translation rate by individual minor codons would be at least additive for mult iple minor codons. Therefore, mRNAs having high relative contents of minor codons would have correspondingly low translation rates. This rate would be reflected by the synthesis of low levels of the encoded protein.

In order to reengineer the bacterial gene(s), the codon bias of the plant is determined. The codon bias is the statistical codon distribution that the plant uses for coding its proteins. After determining the bias, the percent frequency of the codons in the gene(s) of interest is determined. The primary codons preferred by the plant should be determined as well as the second and third choice of preferred codons. The amino acid sequence of the protein of interest is reverse translated so that the resulting nucleic acid sequence codes for the same protein as the native bacterial gene, but the resulting nucleic acid sequence corresponds to the first preferred codons of the desired plant. The new sequence is analyzed for restriction enzyme sites that might have been created by the modification. The identified sites are further modified by replacing the codons with second or third choice preferred codons. Other sites in the sequence which could affect the transcription or translation of the gene of interest are the exon:intron 5' or 3' junctions, poly A addition signals, or RNA polymerase termination signals. The sequence is further analyzed and modified to reduce the frequency of TA or GC doublets. In addition to the doublets, G or C sequence blocks that have more than about four residues that are the same can affect transcription of the sequence. Therefore, these blocks are also modified by replacing the codons of first or second choice, etc. with the next preferred codon of choice. It is preferred that the plant optimized gene(s) contains about 63% of first choice codons, between about 22% to about 37% second choice codons, and between 15% and 0% third choice codons, wherein the total percentage is 100%. Most preferred the plant optimized gene(s) contain about 63% of first choice codons, at least about 22% second choice codons, about 7.5% third choice codons, and about 7.5% fourth choice codons, wherein the total percentage is 100%. The method described above enables one skilled in the art to modify gene(s) that are foreign to a particular plant so that the genes are optimally expressed in plants. The method is further illustrated in pending provisional application U.S.

60/005,405 filed on October 13, 1995, which is incorporated herein by reference. Thus, in order to design plant optimized gene(s) the amino acid sequence of the toxins are reverse translated into a DNA sequence, utilizing a nonredundant genetic code established from a codon bias table compiled for the gene DNA sequence for the particular plant being transformed. The resulting DNA sequence, which is completely homogeneous in codon usage, is further modified to establish a DNA sequence that, besides having a higher degree of codon diversity, also contains strategically placed restriction enzyme recognition sites, desirable base composition, and a lack of sequences that might interfere with transcription of the gene, or translation of the product mRNA.

It is theorized that bacterial genes may be more easily expressed in plants if the bacterial genes are expressed in the plastids. Thus, it may be possible to express bacterial genes in plants, without optimizing the genes for plant expression, and obtain high express of the protein. See U.S. Patent Nos.

4,762,785; 5,451,513 and 5,545,817, which are incorporated herein by reference. One of the issues regarding commercial exploiting transgenic plants is resistance management. This is of particular concern with Bacillus thuringiensis toxins. There are numerous companies commerically exploiting Bacillus thuringiensis and there has been much concern about Bt toxins becoming resistant. One strataegy for insect resistant management would be to combine the toxins produced by Photorhabdus with toxins such as Bt, vegetative insect proteins (Ciba Geigy) or other toxins. The combinations could be formulated for a sprayable application or could be molecular combinations. Plants could be transformed with

Photorhabdus genes that produce insect toxins and other insect toxin genes such as Bt as with other insect toxin genes such as Bt.

European Patent Application 0400246A1 describes

transformation of 2 Bt in a plant, which could be any 2 genes.

Another way to produce a transgenic plant that contains more than one insect resistant gene would be to produce two plants, with each plant containing an insect resistant gene. These plants would be backcrossed using traditional plant breeding techniques to produce a plant containing more than one insect resistant gene.

In addition to producing a transformed plant containing plant optimized gene(s), there are other delivery systems where it may be desirable to reengineer the bacterial gene(s). Along the same lines, a genetically engineered, easily isolated protein toxin fusing together both a molecule attractive to insects as a food source and the insecticidal activity of the toxin may be engineered and expressed in bacteria or in eukaryotic cells using standard, well-known techniques. After purification in the laboratory such a toxic agent with "built-in" bait could be packaged inside standard insect trap housings.

Another delivery scheme is the incorporation of the genetic material of toxins into a baculovirus vector. Baculoviruses infect particular insect hosts, including those desirably targeted with the Photorhabdus toxins. Infectious baculovirus harboring an expression construct for the Photorhabdus toxins could be introduced into areas of insect infestation to thereby intoxicate or poison infected insects. Transfer of the insecticidal properties requires nucleic acid sequences encoding the coding the amino acid sequences for the Photorhabdus toxins integrated into a protein expression vector appropriate to the host in which the vector will reside. One way to obtain a nucleic acid sequence encoding a protein with insecticidal properties is to isolate the native genetic material which produces the toxins from Photorhabdus, using information deduced from the toxin's amino acid sequence, large portions of which are set forth below. As described below, methods of purifying the proteins responsible for toxin activity are also disclosed.

Using N-terminal amino acid sequence data, such as set forth below, one can construct oligonucleotides complementary to all, or a section of, the DNA bases that encode the first amino acids of the toxin. These oligonucleotides can be radiolabeled and used as molecular probes to isolate the genetic material from a genomic genetic library built from genetic material isolated from strains of Photorhabdus. The genetic library can be cloned in plasmid, cosmid, phage or phagemid vectors. The library could be transformed into Escherichia coli and screened for toxin

production by the transformed cells using antibodies raised against the toxin or direct assays for insect toxicity.

This approach requires the production of a battery of oligonucleotides, since the degenerate genetic code allows an amino acid to be encoded in the DNA by any of several three-nucleotide combinations. For example, the amino acid arginine can be encoded by nucleic acid triplets CGA, CGC, CGG, CGT, AGA, and AGG. Since one cannot predict which triplet is used at those positions in the toxin gene, one must prepare oligonucleotides with each potential triplet represented. More than one DNA molecule corresponding to a protein subunit may be necessary to construct a sufficient number of oligonucleotide probes to recover all of the protein subunits necessary to achieve oral toxicity.

From the amino acid sequence of the purified protein, genetic materials responsible for the production of toxins can readily be isolated and cloned, in whole or in part, into an expression vector using any of several techniques well-known to one skilled in the art of molecular biology. A typical

expression vector is a DNA plasmid, though other transfer means including, but not limited to, cosmids, phagemids and phage are also envisioned. In addition to features required or desired for plasmid replication, such as an origin of replication and antibiotic resistance or other form of a selectable marker such as the bar gene of Streptomyces hygroscopicus or

viridochromogenes, protein expression vectors normally

additionally require an expression cassette which incorporates the cis-acting sequences necessary for transcription and

translation of the gene of interest. The cis-acting sequences required for expression in prokaryotes differ from those required in eukaryotes and plants.

A eukaryotic expression cassette requires a transcriptional promoter upstream (5') to the gene of interest, a transcriptional termination region such as a poly-A addition site, and a ribosome binding site upstream of the gene of interest's first codon. In bacterial cells, a useful transcriptional promoter that could be included in the vector is the T7 RNA Polymerase-binding promoter. Promoters, as previously described herein, are known to

efficiently promote transcription of mRNA. Also upstream from the gene of interest the vector may include a nucleotide sequence encoding a signal sequence known to direct a covalently linked protein to a particular compartment of the host cells such as the cell surface.

Insect viruses, or baculoviruses, are known to infect and adversely affect certain insects. The affect of the viruses on insects is slow, and viruses do not stop the feeding of insects. Thus viruses are not viewed as being useful as insect pest control agents. Combining the Photorhabdus toxins genes into a baculovirus vector could provide an efficient way of transmitting the toxins while increasing the lethality of the virus. In addition, since different baculoviruses are specific to different insects, it may be possible to use a particular toxin to

selectively target particularly damaging insect pests. A particularly useful vector for the toxins genes is the nuclear polyhedrosis virus. Transfer vectors using this virus have been described and are now the vectors of choice for transferring foreign genes into insects. The virus-toxin gene recombinant may be constructed in an orally transmissible form. Baculoviruses normally infect insect victims through the mid-gut intestinal mucosa. The toxin gene inserted behind a strong viral coat protein promoter would be expressed and should rapidly kill the infected insect.

In addition to an insect virus or baculovirus or transgenic plant delivery system for the protein toxins of the present invention, the proteins may be encapsulated using Baci ll us churingi ensi s encapsulation technology such as but not limited to U.S. Patent Nos. 4,695,455; 4,695,462; 4,861,595 which are all incorporated herein by reference. Another delivery system for the protein toxins of the present invention is formulation of the protein into a bait matrix, which could then be used in above and below ground insect bait stations. Examples of such technology include but are not limited to PCT Patent Application WO

93/23998, which is incorporated herein by reference.

As is described above, it might become necessary to modify the sequence encoding the protein when expressing it in a non-native host, since the codon preferences of other hosts may differ from that of Photorhabdus . In such a case, translation may be quite inefficient in a new host unless compensating modifications to the coding sequence are made. Additionally, modifications to the amino acid sequence might be desirable to avoid inhibitory cross-reactivity with proteins of the new host, or to refine the insecticidal properties of the protein in the new host. A genetically modified toxin gene might encode a toxin exhibiting, for example, enhanced or reduced toxicity, altered insect resistance development, altered stability, or modified target species specificity.

In addition to the Photorhabdus genes encoding the toxins, the scope of the present invention is intended to include related nucleic acid sequences which encode amino acid biopolymers homologous to the toxin proteins and which retain the toxic effect of the Photorhabdus proteins in insect species after oral ingestion.

For instance, the toxins used in the present invention seem to first inhibit larval feeding before death ensues. By

manipulating the nucleic acid sequence of Photorhabdus toxins or its controlling sequences, genetic engineers placing the toxin gene into plants could modulate its potency or its mode of action to, for example, keep the eating-inhibitory activity while eliminating the absolute toxicity to the larvae. This change could permit the transformed plant to survive until harvest without having the unnecessarily dramatic effect on the ecosystem of wiping out all target insects. All such modifications of the gene encoding the toxin, or of the protein encoded by the gene, are envisioned to fall within the scope of the present invention.

Other envisioned modifications of the nucleic acid include the addition of targeting sequences to direct the toxin to particular parts of the insect larvae for improving its

efficiency.

Strains ATCC 55397, 43948, 43949, 43950, 43951, 43952 have been deposited in the American Type Culture Collection, 12301 Parklawn Drive, Rockville, MD 20852 USA. Amino acid and

nucleotide sequence data for the W-14 native toxin (ATCC 55397) is presented below. Isolation of the genomic DNA for the toxins from the bacterial hosts is also exemplified herein.

Standard and molecular biology techniques were followed and taught in the specification herein. Additional information may be found in Sambrook, J., Fritsch, E. F., and Maniatis, T.

(1989), Molecular Cloning, A Laboratory Manual, Cold Spring Harbor Press, which is incorporated herein by reference.

The following abbreviations are used throughout the Examples: Tris = tris (hydroxymethyl) amino methane; SDS = sodium dodecyl sulfate; EDTA = ethylenediaminetetraacetic acid, IPTG =

isopropylthio-B-galactoside, X-gal = 5-bromo-4-chloro-3-indoyl-B-D-galactoside, CTAB = cetyltrimethylammonium bromide; kbp = kilobase pairs; dATP, dCTP, dGTP, dTTP, I = 2'-deoxynucleoside 5'-triphosphates of adenine, cytosine, guanine, thymine, and inosine, respectively; ATP = adenosine 5' triphosphate. Example 1

Purification of toxin from P. luminescens and Demonstration of toxicity after oral delivery of purified toxin

The insecticidal protein toxin of the present invention was purified from P. luminescens strain W-14, ATCC Accession Number

55397. Stock cultures of P. luminescens were maintained on petri dishes containing 2% Proteose Peptone No. 3 (i.e., PP3, Difco Laboratories, Detroit MI) in 1.5% agar, incubated at 25°C and transferred weekly. Colonies of the primary form of the bacteria were inoculated into 200 ml of PP3 broth supplemented with 0.5% polyoxyethylene sorbitan mono-stearate (Tween 60, Sigma Chemical Company, St. Louis MO) in a one liter flask. The broth cultures were grown for 72 hours at 30°C on a rotary shaker. The toxin proteins can be recovered from cultures grown in the presence or absence of Tween; however, the absence of Tween can affect the form of the bacteria grown and the profile of proteins produced by the bacteria. In the absence of Tween, a variant shift occurs insofar as the molecular weight of at least one identified toxin subunit shifts from about 200 kDa to about 185 kDa.

The 72 hour cultures were centrifuged at 10,000 x g for 30 minutes to remove cells and debris. The supernatant fraction that contained the insecticidal activity was decanted and brought to 50 mM K₂HPO₄ by adding an appropriate volume of 1.0 M K₂HPO₄. The pH was adjusted to 8.6 by adding potassium hydroxide. This supernatant fraction was then mixed with DEAE-Sephacel (Pharmacia LKB Biotechnology) which had been equilibrated with 50 mM K₂HPO₄. The toxic activity was adsorbed to the DEAE resin. This mixture was then poured into a 2.6 x 40 cm column and washed with 50 mM K₂HPO₄ at room temperature at a flow rate of 30 ml/hr until the effluent reached a steady baseline UV absorbance at 280 nm. The column was then washed with 150 mM KCl until the effluent again reached a steady 280 nm baseline. Finally the column was washed with 300 mM KCl and fractions were collected.

Fractions containing the toxin were pooled and filter sterilized using a 0.2 micron pore membrane filter. The toxin was then concentrated and equilibrated to 100 mM KPO₄, pH 6.9, using an ultrafiltration membrane with a molecular weight cutoff of 100 kDa at 4°C (Centriprep 100, Amicon Division-W.R. Grace and Company). A 3 ml sample of the toxin concentrate was applied to the top of a 2.6 × 95 cm Sephacryl S-400 HR gel filtration column (Pharmacia LKB Biotechnology). The eluent buffer was 100 mM KPO₄, pH 6.9, which was run at a flow rate of 17 ml/hr, at 4°C. The effluent was monitored at 280 nm.

Fractions were collected and tested for toxic activity.

Toxicity of chromatographic fractions was examined in a

biological assay using Manduca sexta larvae. Fractions were either applied directly onto the insect diet (Gypsy moth wheat germ diet, ICN Biochemicals Division - ICN Biomedicals, Inc.) or administered by intrahemocelic injection of a 5 ul sample through the first proleg of 4th or 5th instar larva using a 30 gauge needle. The weight of each larva within a treatment group was recorded at 24 hour intervals. Toxicity was presumed if the insect ceased feeding and died within several days of consuming treated insect diet or if death occurred within 24 hours after injection of a fraction.

The toxic fractions were pooled and concentrated using the Centriprep-100 and were then analyzed by HPLC using a 7.5 mm x 60 cm TSK-GEL G-4000 SW gel permeation column with 100 mM potassium phosphate, pH 6.9 eluent buffer running at 0.4 ml/min. This analysis revealed the toxin protein to be contained within a single sharp peak that eluted from the column with a retention time of approximately 33.6 minutes. This retention time

corresponded to an estimated molecular weight of 1,000 kDa. Peak fractions were collected for further purification while fractions not containing this protein were discarded. The peak eluted from the HPLC absorbs UV light at 218 and 280 nm but did not absorb at 405 nm. Absorbance at 405 nm was shown to be an attribute of xenorhabdin antibiotic compounds.

Electrophoresis of the pooled peak fractions in a non-denaturing agarose gel (Metaphor Agarose, FMC BioProducts) showed that two protein complexes are present in the peak. The peak material, buffered in 50 mM Tris-HCl, pH 7.0, was separated on a 1.5% agarose stacking gel buffered with 100 mM Tris-HCl at pH 7.0 and 1.9% agarose resolving gel buffered with 200 mM Tris-borate at pH 8.3 under standard buffer conditions (anode buffer 1M Tris-HCl, pH 8.3; cathode buffer 0.025 M Tris, 0.192 M glycine). The gels were run at 13 mA constant current at 15°C until the phenol red tracking dye reached the end of the gel. Two protein bands were visualized in the agarose gels using Coomassie brilliant blue staining.

The slower migrating band was referred to as "protein band 1" and faster migrating band was referred to as "protein band 2." The two protein bands were present in approximately equal amounts. The Coomassie stained agarose gels were used as a guide to precisely excise the two protein bands from unstained portions of the gels. The excised pieces containing the protein bands were macerated and a small amount of sterile water was added. As a control, a portion of the gel that contained no protein was also excised and treated in the same manner as the gel pieces containing the protein. Protein was recovered from the gel pieces by electroelution into 100 mM Tris-borate pH 3.3, at 100 volts (constant voltage) for two hours. Alternatively, protein was passively eluted from the gel pieces by adding an equal volume of 50 mM Tris-HCl, pH 7.0, to the gel pieces, then incubating at 30°C for 16 hours. This allowed the protein to diffuse from the gel into the buffer, which was then collected.

Results of insect toxicity tests using HPLC-purified toxin (33.6 min. peak) and agarose gel purified toxin demonstrated toxicity of the extracts. Injection of 1.5 ug of the HPLC purified protein kills within 24 hours. Both protein bands 1 and 2, recovered from agarose gels by passive elution or

electroelution, were lethal upon injection. The protein

concentration estimated for these samples was less than 50 ng/ larva. A comparison of the weight gain and the mortality between the groups of larvae injected with protein bands 1 or 2 indicate that protein band 1 was more toxic by injection

delivery.

When HPLC-purified toxin was applied to larval diet at a concentration of 7.5 μg/larva, it caused a halt in larval weight gain (24 larvae tested). The larvae begin to feed, but after consuming only a very small portion of the toxin treated diet they began to show pathological symptoms induced by the toxin and the larvae cease feeding. The insect frass became discolored and most larva showed signs of diarrhea. Significant insect

mortality resulted when several 5 μg toxin doses were applied to the diet over a 7-10 day period.

Agarose-separated protein band 1 significantly inhibited larval weight gain at a dose of 200 ng/larva. Larvae fed similar concentrations of protein band 2 were not inhibited and gained weight at the same rate as the control larvae. Twelve larvae were fed eluted protein and 45 larvae were fed protein-containing agarose pieces. These two sets of data indicate that protein band 1 was orally toxic to Manduca sexta. In this experiment it appeared that protein band 2 was not toxic to Manduca sexta.

Further analysis of protein bands 1 and 2 by SDS-PAGE under denaturing conditions showed that each band was composed of several smaller protein subunits. Proteins were visualized by Coomassie brilliant blue staining followed by silver staining to achieve maximum sensitivity. The protein subunits in the two bands were very similar. Protein band 1 contains 8 protein subunits of 25.1, 56.2, 60.3, 65.6, 166, 171, 184 and 208 kDa. Protein band 2 had an identical profile except that the 25.1, 60.8, and 65.6 kDa proteins were not present. The 56.2, 60.8, 65.6, and 184 kDa proteins were present in the complex of protein band 1 at approximately equal concentrations and represent 80% or more of the total protein content of that complex.

The native HPLC-purified toxin was further characterized as follows. The toxin was heat labile in that after being heated to 60°C for 15 minutes it lost its ability to kill or to inhibit weight gain when injected or fed to M. sexta larvae. Assays were designed to detect lipase, type C phospholipase, nuclease or red blood cell hemolysis activities and were performed with purified toxin. None of these activities were present. Antibiotic zone inhibition assays were also done and the purified toxin failed to inhibit growth of Gram-negative or positive bacteria, yeast or filamentous fungi, indicating that the toxic is not a xenorhabdin antibiotic.

The native HPLC-purified toxin was tested for ability to kill insects other than Manduca sexta. Table 2 lists insects killed by the HPLC-purified P. luminescens toxin in this study.

Example 2

Insecticide Utility

The Photorhabdus luminescens utility and toxicity were further characterized. Photorhabdus luminescens (strain W-14) culture broth was produced as follows. The production medium was 2% Bacto Proteose Peptone^® Number 3 (PP3, Difco Laboratories, Detroit, Michigan) in Milli-Q^® deionized water. Seed culture flasks consisted of 175 ml medium placed in a 500 ml tribaffled flask with a Delong neck, covered with a Kaput and autoclaved for 20 minutes, T=250°F. Production flasks consisted of 500 mls in a 2.8 liter 500 ml tribaffled flask with a Delong neck, covered by a Shin-etsu silicon foam closure. These were

autoclaved for 45 minutes, T=250ºF. The seed culture was incubated at 28°C at 150 rpm in a gyrotory shaking incubator with a 2 inch throw. After 16 hours of growth, 1% of the seed culture was placed in the production flask which was allowed to grow for 24 hours before harvest. Production of the toxin appears to be during log phase growth. The microbial broth was transferred to a 1L centrifuge bottle and the cellular biomass was pelleted 130 minutes at 2500 RPM at 4°C, [R.C.F. = ~1600] HG-4L Rotor RC3 Sorval centrifuge, Dupont, Wilmington, Delaware). The primary broth was chilled at 4°C for 8 - 16 hours and recentrifuged at least 2 hours (conditions above) to further clarify the broth by removal of a putative mucopolysaccharide which precipitated upon standing. (An alternative processing method combined both steps and involved the use of a 16 hour clarification centrifugation, same conditions as above.) This broth was then stored at 4ºC prior to bioassay or filtration.

Photorhabdus culture broth and protein toxin(s) purified from this broth showed activity (mortality and/or growth

inhibition, reduced adult emergence) against a number of insects. More specifically, the activity is seen against corn rootworm (larvae and adult), Colorado potato beetle, and turf grubs, which are members of the insect order Coleoptera. Other members of the Coleoptera include wireworms, pollen beetles, flea beetles, seed beetles and weevils. Activity has also been observed against aster leafhopper, which is a member of the order, Homoptera.

Other members of the Homoptera include planthoppers, pear pyslla, apple sucker, scale insects, whiteflies, and spittle bugs, as well as numerous host specific aphid species. The broth and purified fractions are also active against beet armyworm, cabbage looper, black cutworm, tobacco budworm, European corn borer, corn earworm, and codling moth, which are members of the order

Lepidoptera. Other typical members of this order are clothes moth, Indian mealmoth, leaf rollers, cabbage worm, cotton bollworm, bagworm, Eastern tent caterpillar, sod webworm, and fall armyworm. Activity is also seen against fruitfly and mosquito larvae, which are members of the order Diptera. Other members of the order Diptera are pea midge, carrot fly, cabbage root fly, turnip root fly, onion fly, crane fly, house fly, and various mosquito species. Activity is seen against carpenter ant and Argentine ant, which are members of the order that also includes fire ants, oderous house ants, and little black ants.

The broth/fraction is useful for reducing populations of insects and were used in a method of inhibiting an insect population. The method may comprise applying to a locus of the insect an effective insect inactivating amount of the active described. Results are reported in Table 3.

Activity against corn rootworm larvae was tested as follows. Photorhabdus culture broth (filter sterilized, cell-free) or purified HPLC fractions were applied directly to the surface (~1.5 cm²) of 0.25 ml of artificial diet in 30 μl aliquots following dilution in control medium or 10 mM sodium phosphate buffer, pH 7.0, respectively. The diet plates were allowed to air-dry in a sterile flow-hood and the wells were infested with single, neonate Diabrotica undecimpunctata howardi (Southern corn rootworm, SCR) hatched from sterilized eggs, with second instar SCR grown on artificial diet or with second instar Diabrotica virgifera virgifera (Western corn rootworm, WCR) reared on corn seedlings grown in Metromix^®. Second instar larvae were weighed prior to addition to the diet. The plates were sealed, placed in a humidified growth chamber and maintained at 27°C for the appropriate period (4 days for neonate and adult SCR, 2-5 days for WCR larvae, 7-14 days for second instar SCR). Mortality and weight determinations were scored as indicated. Generally, 16 insects per treatment were used in all studies. Control

mortalities were as follows: neonate larvae, <5%, adult beetles, 5%. Activity against Colorado potato beetle was tested as follows. Photorhabdus culture broth or control medium was applied to the surface (~2.0 cm²) of 1.5 ml of standard artificial diet held in the wells of a 24-well tissue culture plate. Each well received 50 μl of treatment and was allowed to air dry.

Individual second instar Colorado potato beetle (Leptmotarsa deceml ineata, CPB) larvae were then placed onto the diet and mortality was scored after 4 days. Ten larvae per treatment were used in all studies. Control mortality was 3.3%.

Activity against Japanese beetle grubs and beetles was tested as follows. Turf grubs (Popillia japonica, 2-3rd instar) were collected from infested lawns and maintained in the

laboratory in soil/peat mixture with carrot slices added as additional diet. Turf beetles were pheromone-trapped locally and maintained in the laboratory in plastic containers with maple leaves as food. Following application of undiluted Photorhabdus culture broth or control medium to corn rootworm artificial diet (30 μl/1.54 cm², beetles) or carrot slices (larvae), both stages were placed singly in a diet well and observed for any mortality and feeding. In both cases there was a clear reduction in the amount of feeding (and feces production) observed.

Activity against mosquito larvae was tested as follows. The assay was conducted in a 96-well microtiter plate. Each well contained 200 μl of aqueous solution (Photorhabdus culture broth, control medium or H₂0) and approximately 20, 1-day old larvae (Aedes aegypti). There were 6 wells per treatment. The results were read at 2 hours after infestation and did not change over the three day observation period. No control mortality was seen.

Activity against fruitflies was tested as follows.

Purchased Drosophila melanogaster medium was prepared using 50% dry medium and a 50% liquid of either water, control medium or Photorhabdus culture broth. This was accomplished by placing 8.0 ml of dry medium in each of 3 rearing vials per treatment and adding 8.0 ml of the appropriate liquid. Ten late instar

Drosophila melanogaster maggots were then added to each vial. The vials were held on a laboratory bench, at room temperature, under fluorescent ceiling lights. Pupal or adult counts were made after 3, 7 and 10 days of exposure. Incorporation of

Photorhabdus culture broth into the diet media for fruitfly maggots caused a slight (17%) but significant reduction in day-10 adult emergence as compared to water and control medium (3% reduction),

Activity against aster leafhopper was tested as follows. The ingestion assay for aster leafhopper (Macrosteles severini) is designed to allow ingestion of the active without other external contact. The reservoir for the active/" food" solution is made by making 2 holes in the center of the bottom portion of a 35 × 10 mm Petri dish. A 2 inch Parafilm M^® square is placed across the top of the dish and secured with an "O" ring. A 1 oz. plastic cup is then infested with approximately 7 leafhoppers and the reservoir is placed on top of the cup, Parafilm down. The test solution is then added to the reservoir through the holes. In tests using undiluted Photorhabdus culture broth, the broth and control medium were dialyzed against water to reduce control mortality. Mortality is reported at day 2 where 26.5% control mortality was seen. In the tests using purified fractions (200 mg protein/ml ) a final concentration of 5% sucrose was used in all treatments to improve survivability of the aster leafhoppers. The assay was held in an incubator at 28°C, 70% RH with a 16/8 photoperiod. The assay was graded for mortality at 72 hours. Control mortality was 5.5%.

Activity against Argentine ants was tested as follows. A 1.5 ml aliquot of 100% Photorhabdus culture broth, control medium or water was pipetted into 2.0 ml clear glass vials. The vials were plugged with a piece of cotton dental wick that was

moistened with the appropriate treatment. Each vial was placed into a separate 60×16mm Petri dish with 8 to 12 adult Argentine ants (Linepithema humile). There were three replicates per treatment. Bioassay plates were held on a laboratory bench, at room temperature under fluorescent ceiling lights. Mortality readings were made after 5 days of exposure. Control mortality was 24%.

Activity against carpenter ant was tested as follows. Black carpenter ant workers (Camponotus pennsylvanicus) were collected from trees on DowElanco property in Indianapolis, IN. Tests with Photorhabdus culture broth were performed as follows. Each plastic bioassay container (7 1/8" × 3") held fifteen workers, a paper harborage and 10 ml of broth or control media in a plastic shot glass. A cotton wick delivered the treatment to the ants through a hole in the shot glass lid. All treatments contained 5% sucrose. Bioassays were held in the dark at room temperature and graded at 19 days. Control mortality was 9%. Assays delivering purified fractions utilized artificial ant diet mixed with the treatment (purified fraction or control solution) at a rate of 0.2 ml treatment/2.0 g diet in a plastic test tube. The final protein concentration of the purified fraction was less than 10 μg/g diet. Ten ants per treatment, a water source, harborage and the treated diet were placed in sealed plastic containers and maintained in the dark at 27°C in a humidified incubator. Mortality was scored at day 10. No control mortality was seen.

Activity against various lepidopteran larvae was tested as follows. Photorhabdus culture broth or purified fractions were

2

applied directly to the surface (~1.5 cm ) of 0.25 ml of standard artificial diet in 30 μl aliquots following dilution in control medium or 10 mM sodium phosphate buffer, pH 7.0, respectively. The diet plates were allowed to air-dry in a sterile flow-hood and the wells were infested with single, neonate larva. European corn borer (Ostrinia nubilalis) and corn earworm (Helicoverpa zea) eggs were supplied from commercial sources and hatched in-house, whereas beet armyworm (Spodoptera exigua), cabbage looper (Trichoplusia ni), tobacco budworm (Heliothis virescens), codling moth (Laspeyresia pomonella) and black cutworm (Agrotis ipsilon) larvae were supplied internally. Following infestation with larvae, the diet plates were sealed, placed in a humidified growth chamber and maintained in the dark at 27°C for the appropriate period. Mortality and weight determinations were scored at days 5-7 for Photorhabdus culture broth and days 4-7 for the purified fraction. Generally, 16 insects per treatment were used in all studies. Control mortality ranged from 4-12.5% for control medium and was less than 10% for phosphate buffer. Example 3

Insecticide Utility Upon Soil Application

Photorhabdus l uminescens (strain W-14) culture broth was shown to be active against corn rootworm when applied directly to soil or a soil-mix (Metromix^®). Activity against neonate SCR and WCR in Metromix^® was tested as follows (Table 4). The test was run using corn seedlings (United Agriseeds brand CL614) that were germinated in the light on moist filter paper for 6 days. After roots were approximately 3-6 cm long, a single kernel/seedling was planted in a 591 ml clear plastic cup with 50 gm of dry

Metromix^®. Twenty neonate SCR or WCR were then placed directly on the roots of the seedling and covered with Metromix^®. Upon infestation, the seedlings were then drenched with 50 ml total volume of a diluted broth solution. After drenching, the cups were sealed and left at room temperature in the light for 7 days. Afterwards, the seedlings were washed to remove all Metromix^® and the roots were excised and weighed. Activity was rated as the percentage of corn root remaining relative to the control plants and as leaf damage induced by feeding. Leaf damage was scored visually and rated as either -, +, ++, or +++, with - representing no damage and +++ representing severe damage.

Activity against neonate SCR in soil was tested as follows (Table 5). The test was run using corn seedlings (United

Agriseeds brand CL614) that were germinated in the light on moist filter paper for 6 days. After the roots were approximately 3-6 cm long, a single kernel/seedling was planted in a 591 ml clear plastic cup with 150 gm of soil from a field in Lebanon, IN planted the previous year with corn. This soil had not been previously treated with insecticides. Twenty neonate SCR were then placed directly on the roots of the seedling and covered with soil. After infestation, the seedlings were drenched with 50 ml total volume of a diluted broth solution. After drenching, the unsealed cups were incubated in a high relative humidity chamber (80%) at 78°F. Afterwards, the seedlings were washed to remove all soil and the roots were excised and weighed. Activity was rated as the percentage of corn root remaining relative to the control plant s and as leaf damage induced by f eeding . Leaf damage was scored visually and rated as either -, +, ++, or +++, with - representing no damage and +++ representing severe damage.

Activity of Photorhabdus luminescens (strain W-14) culture broth against second instar turf grubs in Metromix^® was observed in tests conducted as follows (Table 6). Approximately 50 gm of dry Metromix^® was added to a 591 ml clear plastic cup. The

Metromix^® was then drenched with 50 ml total volume of a 50% (v/v) diluted Photorhabdus broth solution. The dilution of crude broth was made with water, with 50% broth being prepared by adding 25 ml of crude broth to 25 ml of water for 50 ml total volume. A 1% (w/v) solution of proteose peptone #3 (PP3), which is a 50% dilution of the normal media concentration, was used as a broth control. After drenching, five second instar turf grubs were placed on the top of the moistened Metromix^®. Healthy turf grub larvae burrowed rapidly into the Metromix^®. Those larvae that did not burrow within 1h were removed and replaced with fresh larvae. The cups were sealed and placed in a 28°C incubator, in the dark. After seven days, larvae were removed from the Metromix^® and scored for mortality. Activity was rated the percentage of mortality relative to control.

Example 4

Insecticide Utility Upon Leaf Application Activity of Photorhabdus broth against European corn borer was seen when the broth was applied directly to the surface of maize leaves (Table 7). In these assays Photorhabdus broth was diluted 100-fold with culture medium and applied manually to the

2

surface of excised maize leaves at a rate of ~6.0 μl/cm of leaf surface. The leaves were air dried and cut into equal sized strips approximately 2 × 2 inches. The leaves were rolled, secured with paper clips and placed in 1 oz plastic shot glasses with 0.25 inch of 2% agar on the bottom surface to provide moisture. Twelve neonate European corn borers were then placed onto the rolled leaf and the cup was sealed. After incubation for 5 days at 27°C in the dark, the samples were scored for feeding damage and recovered larvae.

Activity of the culture broth against neonate tobacco budworm (Heliothis virescens) was demonstrated using a leaf dip methodology. Fresh cotton leaves were excised from the plant and leaf disks were cut with an 18.5 mm cork-borer. The disks were individually emersed in control medium (PP3) or Photorhabdus luminescens (strain W-14) culture broth which had been

concentrated approximately 10-fold using an Amicon (Beverly, MA), Proflux M12 tangential filtration system with a 10 kDa filter. Excess liquid was removed and a straightened paper clip was placed through the center of the disk. The paper clip was then wedged into a plastic, 1.0 oz shot glass containing approximately 2.0 ml of 1% Agar. This served to suspend the leaf disk above the agar. Following drying of the leaf disk, a single neonate tobacco budworm larva was placed on the disk and the cup was capped. The cups were then sealed in a plastic bag and placed in a darkened, 27°C incubator for 5 days. At this time the

remaining larvae and leaf material were weighed to establish a measure of leaf damage (Table 8).

Example 5, Part A

Characterization of Toxin Peptide Components

In a subsequent analysis, the toxin protein subunits of the bands isolated as in Example 1 were resolved on a 7% SDS

polyacrylamide electrophoresis gel with a ratio of 30:0.8

(acrylamide:BIS-acrylamide). This gel matrix facilitates better resolution of the larger proteins. The gel system used to estimate the Band 1 and Band 2 subunit molecular weights in Example 1 was an 18% gel with a ratio of 38:0.18 (acrylamide:BIS-acrylamide), which allowed for a broader range of size

separation, but less resolution of higher molecular weight components.

In this analysis, 10, rather than 8, protein bands were resolved. Table 9 reports the calculated molecular weights of the 10 resolved bands, and directly compares the molecular weights estimated under these conditions to those of the prior example. It is not surprising that additional bands were detected under the different separation conditions used in this example. Variations between the prior and new estimates of molecular weight are also to be expected given the differences in analytical conditions. In the analysis of this example, it is thought that the higher molecular weight estimates are more accurate than in Example 1, as a result of improved resolution. However, these are estimates based on SDS PAGE analysis, which are typically not analytically precise and result in estimates of peptides and which may have been further altered due to post- and co-translational modifications.

Amino acid sequences were determined for the N-terminal portions of five of the 10 resolved peptides. Table 9 correlates the molecular weight of the proteins and the identified

sequences. In SEQ ID NO: 2, certain analyses suggest that the proline at residue 5 may be an asparagine (asn). In SEQ ID NO: 3, certain analyses suggest that the amino acid residues at

positions 13 and 14 are both arginine (arg). In SEQ ID NO:4, certain analyses suggest that the amino acid residue at position 6 may be either alanine (ala) or serine (ser). In SEQ ID NO:5, certain analyses suggest that the amino acid residue at position 3 may be aspartic acid (asp).

Example 5, Part B

Characterization of Toxin Peptide Components

New N-terminal sequence, SEQ ID NO: 15, Ala Gln Asp Gly Asn Gln Asp Thr Phe Phe Ser Gly Asn Thr, was obtained by further N-terminal sequencing of peptides isolated from Native HPLC-purif ied toxin as described in Example 5, Part A, above. This peptide comes from the tcaA gene. The peptide labeled TcaA_ii, starts at position 254 and goes to position 491, where the

TcaA_iii peptide starts, SEQ ID NO: 4. The estimated size of the peptide based on the gene sequence is 25,240 Da.

Example 6

Characterization of Toxin Peptide Components

In yet another analysis, the toxin protein complex was re-isolated from the Photorhabdus luminescens growth medium (after culture without Tween) by performing a 10% - 80% ammonium sulfate precipitation followed by an ion exchange chromatography step (Mono Q) and two molecular sizing chromatography steps. These conditions were like those used in Example 1. During the first molecular sizing step, a second biologically active peak was found at about 100 + 10 kDa. Based upon protein measurements, this fraction was 20 - 50 fold less active than the larger, or primary, active peak of about 860 ± 100 kDa (native). During this isolation experiment, a smaller active peak of about 325 + 50 kDa that retained a considerable portion of the starting biological activity was also resolved. It is thought that the 325 kDa peak is related to or derived from the 860 kDa peak. A 56 kDa protein was resolved in this analysis The N- terminal sequence of this protein is presented in SEQ ID NO:6. It is noteworthy that this protein shares significant identity and conservation with SEQ ID NO:5 at the N-terminus, suggesting that the two may be encoded by separate members of a gene family and that the proteins produced by each gene are sufficiently similar to both be operable in the insecticidal toxin complex

A second, prominent 185 kDa protein was consistently present in amounts comparable to that of protein 3 from Table 9, and may be the same protein or protein fragment. The N-terminal sequence of this 185 kDa protein is shown at SEQ ID NO: 7.

Additional N-terminal amino acid sequence data were also obtained from isolated proteins. None of the determined N-terminal sequences appear identical to a protein identified in Table 9. Other proteins were present in isolated preparation.

One such protein has an estimated molecular weight of 108 kDa and an N-terminal sequence as shown in SEQ ID NO: 8. A second such protein has an estimated molecular weight of 80 kDa and an N-terminal sequence as shown in SEQ ID NO: 9.

When the protein material in the approximately 325 kDa active peak was analyzed by size, bands of approximately 51, 31, 28, and 22 kDa were observed. As in all cases in which a molecular weight was determined by analysis of electrophoretic mobility, these molecular weights were subject to error effects introduced by buffer ionic strength differences, electrophoresis power differences, and the like. One of ordinary skill would understand that definitive molecular weight values cannot be determined using these standard methods and that each was subject to variation. It was hypothesized that proteins of these sizes are degradation products of the larger protein species (of approximately 200 kDa size) that were observed in the larger primary toxin complex.

Finally, several preparations included a protein having the N-terminal sequence shown in SEQ ID NO: 10. This sequence was strongly homologous to known chaperonin proteins, accessory proteins known to function in the assembly of large protein complexes. Although the applicants could not ascribe such an assembly function to the protein identified in SEQ ID NO: 10, it was consistent with the existence of the described toxin protein complex that such a chaperonin protein could be involved in its assembly. Moreover, although such proteins have not directly been suggested to have toxic activity, this protein may be important to determining the overall structural nature of the protein toxin, and thus, may contribute to the toxic activity or durability of the complex in vivo after oral delivery.

Subsequent analysis of the stability of the protein toxin complex to proteinase K was undertaken. It was determined that after 24 hour incubation of the complex in the presence of a 10-fold molar excess of proteinase K, activity was virtually eliminated (mortality on oral application dropped to about 5%). These data confirm the proteinaceous nature of the toxin.

The toxic activity was also retained by a dialysis membrane, again confirming the large size of the native toxin complex. Example 7

Isolation, Characterization and Partial Amino Acid

Sequencing of Photorhabdus Toxins

Isolation and N-Terminal Amino Acid Sequencing: In a set of experiments conducted in parallel to Examples 5 and 6, ammonium sulfate precipitation of Photorhabdus proteins was performed by adjusting Photorhabdus broth, typically 2-3 liters, to a final concentration of either 10% or 20% by the slow addition of ammonium sulfate crystals. After stirring for 1 hour at 4°C, the material was centrifuged at 12,000 x g for 30 minutes. The supernatant was adjusted to 80% ammonium sulfate, stirred at 4°C for 1 hour, and centrifuged at 12,000 x g for 60 minutes. The pellet was resuspended in one-tenth the volume of 10 mM Na₂·PO₄, pH 7.0 and dialyzed against the same phosphate buffer overnight at 4°C. The dialyzed material was centrifuged at 12,000 x g for 1 hour prior to ion exchange chromatography.

A HR 16/50 Q Sepharose (Pharmacia) anion exchange column was equilibrated with 10 mM Na₂·PO₄, pH 7.0. Centrifuged, dialyzed ammonium sulfate pellet was applied to the Q Sepharose column at a rate of 1.5 ml/min and washed extensively at 3.0 ml/min with equilibration buffer until the optical density (O.D. 280) reached less than 0.100. Next, either a 60 minute NaCl gradient ranging from 0 to 0.5 M at 3 ml/min, or a series of step elutions using 0.1 M, 0.4 M and finally 1.0 NaCl for 60 minutes each was applied to the column. Fractions were pooled and concentrated using a Centriprep 100. Alternatively, proteins could be eluted by a single 0.4 M NaCl wash without prior elution with 0.1 M NaCl.

Two milliliter aliquots of concentrated Q Sepharose samples were loaded at 0.5 ml/min onto a HR 16/50 Superose 12 (Pharmacia) gel filtration column equilibrated with 10 mM Na₂·PO₄, pH 7.0. The column was washed with the same buffer for 240 min at 0.5 ml/min and 2 min samples were collected. The void volume material was collected and concentrated using a Centriprep 100. Two milliliter aliquots of concentrated Superose 12 samples were loaded at 0.5 ml/min onto a HR 16/50 Sepharose 4B-CL (Pharmacia) gel filtration column equilibrated with 10 mM Na₂·PO₄, pH 7.0. The column was washed with the same buffer for 240 min at 0.5 ml/min and 2 min samples were collected.

The excluded protein peak was subjected to a second

fractionation by application to a gel filtration column that used a Sepharose CL-4B resin, which separates proteins ranging from ~30 kDa to 1000 kDa. This fraction was resolved into two peaks; a minor peak at the void volume (>1000 kDa) and a major peak which eluted at an apparent molecular weight of about 860 kDa. Over a one week period subsequent samples subjected to gel filtration showed the gradual appearance of a third peak

(approximately 325 kDa) that seemed to arise from the major peak, perhaps by limited proteolysis. Bioassays performed on the three peaks showed that the void peak had no activity, while the 860 kDa toxin complex fraction was highly active, and the 325 kDa peak was less active, although quite potent. SDS PAGE analysis of Sepharose CL-4B toxin complex peaks from different

fermentation productions revealed two distinct peptide patterns, denoted "P" and "S". The two patterns had marked differences in the molecular weights and concentrations of peptide components in their fractions. The "S" pattern, produced most frequently, had 4 high molecular weight peptides (> 150 kDa) while the "P" pattern had 3 high molecular weight peptides. In addition, the "S" peptide fraction was found to have 2-3 fold more activity against European Corn Borer. This shift may be related to variations in protein expression due to age of inoculum and/or other factors based on growth parameters of aged cultures.

Milligram quantities of peak toxin complex fractions determined to be "P" or "S" peptide patterns were subjected to preparative SDS PAGE, and transblotted with TRIS-glycine (Seprabuff™ to PVDF membranes (ProBlott™, Applied Biosystems) for 3-4 hours. Blots were sent for amino acid analysis and N-terminal amino acid sequencing at Harvard MicroChem and Cambridge ProChem, respectively. Three peptides in the "S" pattern had unique N-terminal amino acid sequences compared to the sequences identified in the previous example. A 201 kDa (TcdA_ii) peptide set forth as SEQ ID NO:.13 below shared between 33% amino acid identity and 50% similarity with SEQ ID NO:1 (TcbA_ii) (Table 10, in Table 10 vertical lines denote amino acid identities and colons indicate conservative amino acid substitutions). A second peptide of 197 kDa, SEQ ID NO:14 (TcdB), had 42% identity and 58% homology with SEQ ID NO:2 (TcaC). Yet a third peptide of 205 kDa was denoted TcdA_ii. In addition, a limited N-terminal amino acid sequence, SEQ ID NO: 16 (TcbA), of a peptide of at least 235 kDa was identical in homology with the amino acid sequence, SEQ ID NO: 12, deduced from a cloned gene (tcbA), SEQ ID NO: 11,

containing a deduced amino acid sequence corresponding to SEQ ID NO:1 (TcbA_ii). This indicates that the larger 235+ kDa peptide was proteolytically processed to the 201 kDa peptide, (TcbA_ii), (SEQ ID NO:1) during fermentation, possibly resulting in

activation of the molecule. In yet another sequence, the sequence originally reported as SEQ ID NO: 5 (TcaB_ii) reported in Example 5 above, was found to contain an aspartic acid residue (Asp) at the third position rather than glycine (Gly) and two additional amino acids Gly and Asp at the eighth and ninth positions, respectively. In yet two other sequences, SEQ ID NO: 2 (TcaC) and SEQ ID NO:3 (TcaB_i), additional amino acid sequence was obtained. Densitometric quantitation was performed using a sample that was identical to the "S" preparation sent for N-terminal analysis. This analysis showed that the 201 kDa and 197 kDa peptides represent 7.0% and 7.2%, respectively, of the total Coomassie brillant blue stained protein in the "S" pattern and are present in amounts similar to the other abundant peptides. It is speculated that these peptides may represent protein homologs, analogous to the situation found with other bacterial toxins, such as various CryI Bt toxins. These proteins vary from 40-90% homology at their N-terminal amino acid sequence, which encompasses the toxic fragment. Internal Amino Acid Sequencing: To facilitate cloning of toxin peptide genes, internal amino acid sequences of selected peptides were obtained as followed. Milligram quantities of peak 2A fractions determined to be "P" or "S" peptide patterns were subjected to preparative SDS PAGE, and transblotted with TRIS-glycine (Seprabuff™ to PVDF membranes (ProBlott™, Applied

Biosystems) for 3-4 hours. Blots were sent for amino acid analysis and N-terminal amino acid sequencing at Harvard

MicroChem and Cambridge ProChem, respectively. Three peptides, referred to as TcbA_ii (containing SEQ ID NO:1), TcdA_ii, and TcaB, (containing SEQ ID NO: 3) were subjected to trypsin digestion by Harvard MicroChem followed by HPLC chromatography to separate individual peptides. N-terminal amino acid analysis was

performed on selected tryptic peptide fragments. Two internal peptides were sequenced for the peptide TcaB_i (205 kDa peptide) referred to as TcaB_i-PT111 (SEQ ID NO: 17) and TcaB_i-PT79 (SEQ ID NO: 18). Two internal peptides were sequenced for the peptide TcaB_i (68 kDa peptide) referred to as TcaB_i-PT158 (SEQ ID NO: 19) and TcaB_i-PT108 (SEQ ID NO:20). Four internal peptides were sequenced for the peptide TcbA_ii (201 kDa peptide) referred to as TCBAII-PT103 (SEQ ID NO.21), TcbA_ii-PT56 (SEQ ID NO:22), TcbA_ii-PT81(a) (SEQ ID NO:23), and TcbA_ii-PT81 (b) (SEQ ID NO:24).

Example 8

Construction of a cosmid library of Photorhabdus luminescens W-14 genomic DNA and its screening to isolate genes encoding peptides comprising the toxic protein preparation

As a prerequisite for the production of Photorhabdus insect toxic proteins in heterologous hosts, and for other uses, it is necessary to isolate and characterize the genes that encode those peptides. This objective was pursued in parallel. One approach, described later, was based on the use of monoclonal and

polyclonal antibodies raised against the purified toxin which were then used to isolate clones from an expression library. The other approach, described in this example, is based on the use of the N-terminal and internal amino acid sequence data to design degenerate oligonucleotides for use in PCR amplication. Either method can be used to identify DNA clones that contain the peptide-encoding genes so as to permit the isolation of the respective genes, and the determination of their DNA base

sequence.

GENOMIC DNA ISOLATION: Photorhabdus luminescens strain W-14 (ATCC accession number 55397) was grown on 2% proteose peptone #3 agar (Difco Laboratories, Detroit, MI) and insecticidal toxin competence was maintained by repeated bioassay after passage, using the method described in Example 1 above. A 50 ml shake culture was produced in a 175 ml baffled flask in 2% proteose peptone #3 medium, grown at 28°C and 150 rpm for approximately 24 hours. 15 ml of this culture was pelleted and frozen in its medium at -20°C until it was thawed for DNA isolation. The thawed culture was centrifuged, (700 x g, 30 min) and the floating orange mucopolysaccharide material was removed. The remaining cell material was centrifuged (25,000 x g, 15 min) to pellet the bacterial cells, and the medium was removed and discarded.

Genomic DNA was isolated by an adaptation of the CTAB method described in section 2.4.1 of Current Protocols in Molecular Biology (Ausubel et al. eds, John Wiley & Sons, 1994) [modified to include a salt shock and with all volumes increased 10-fold]. The pelleted bacterial cells were resuspended in TE buffer (10 mM Tris-HCl, 1 mM EDTA, pH 8.0) to a final volume of 10 ml, then 12 ml of 5 M NaCl was added; this mixture was centrifuged 20 min at 15,000 x g. The pellet was resuspended in 5.7 ml TE and 300 ml of 10% SDS and 60 ml of 20 mg/ml proteinase K (Gibco BRL

Products, Grand Island, NY; in sterile distilled water) were added to the suspension. This mixture was incubated at 37°C for 1 hr; then approximately 10 mg lysozyme (Worthington Biochemical Corp., Freehold, NJ) was added. After an additional 45 min, 1 ml of 5 M NaCl and 800 ml of CTAB/NaCl solution (10% w/v CTAB, 0.7 M NaCl) were added. This preparation was incubated 10 min at 65ºC. then gently agitated and further incubated and agitated for approximately 20 min to assist clearing of the cellular material. An equal volume of chloroform/isoamyl alcohol solution (24:1, v/v) was added, mixed gently and centrifuged. After two

extractions with an equal volume of PCI

(phenol/chloroform/isoamyl alcohol; 50:49:1, v/v/v; equilibrated with 1 M Tris-HCl, pH 8.0; Intermountain Scientific Corporation, Kaysville, UT), the DNA was precipitated with 0.6 volume of isopropanol. The DNA precipitate was gently removed with a glass rod, washed twice with 70% ethanol, dried, and dissolved in 2 ml STE (10 mM Tris-HCl pH 8.0, 10 mM NaCl, 1 mM EDTA). This

preparation contained 2.5 mg/ml DNA, as determined by optical density at 260 nm (i.e., OD₂₆₀).

The molecular size range of the isolated genomic DNA was evaluated for suitability for library construction. CHEF gel analysis was performed in 1.5% agarose (Seakem^® LE, FMC

BioProducts, Rockland, ME) gels with 0.5 X TBE buffer (44.5 mM Tris-HCl pH 8.0, 44.5 mM H₃BO₃, 1 mM EDTA) on a BioRad CHEF-DR II apparatus with a Pulsewave 760 Switcher (Bio-Rad Laboratories, Inc., Richmond, CA). The running parameters were: initial A time, 3 sec; final A time, 12 sec; 200 volts; running

temperature, 4-18°C; run time, 16.5 hr. Ethidium bromide staining and examination of the gel under ultraviolet light indicated the DNA ranged from 30-250 kbp in size.

CONSTRUCTION OF LIBRARY: A partial Sau3A 1 digest was made of this Photorhabdus genomic DNA preparation. The method was based on section 3.1.3 of Ausubel (supra.). Adaptions included running smaller scale reactions under various conditions until nearly optimal results were achieved. Several scaled-up large reactions with varied conditions were run, the results analyzed on CHEF gels, and only the best large scale preparation was carried forward. In the optimal case, 200 μg of Photorhabdus genomic DNA was incubated with 1.5 units of Sau3A 1 (New England Biolabs, "NEB", Beverly, MA) for 15 min at 37°C in 2 ml total volume of 1X NEB 4 buffer (supplied as 10X by the manufacturer). The reaction was stopped by adding 2 ml of PCI and centrifuging at 8000 x g for 10 min. To the supernatant were added 200 μl of 5 M NaCl plus 6 ml of ice-cold ethanol. This preparation was chilled for 30 min at -20°C, then centrifuged at 12,000 x g for 15 min. The supernatant was removed and the precipitate was dried in a vacuum oven at 40°C, then resuspended in 400 μl STE. Spectrophotometric assay indicated about 40% recovery of the input DNA. The digested DNA was size fractionated on a sucrose gradient according to section 5.3.2 of CPMB ( op . cit . ) . A 10% to 40% (w/v) linear sucrose gradient was prepared with a gradient maker in Ultra-Clear™ tubes (Beckman Instruments, Inc., Palo Alto, CA) and the DNA sample was layered on top. After

centrifugation, (26,000 rpm, 17 hr, Beckman SW41 rotor, 20°C), fractions (about 750 μl) were drawn from the top of the gradient and analyzed by CHEF gel electrophoresis (as described earlier). Fractions containing Sau3A 1 fragments in the size range 20-40 kbp were selected and DNA was precipitated by a modification (amounts of all solutions increased approximately 6.3-fold) of the method in section 5.3.3 of Ausubel ( supra . ) . After overnight precipitation, the DNA was collected by centrifugation (17,000 x g, 15 min), dried, redissolved in TE, pooled into a final volume of 80 μl, and reprecipitated with the addition of 8 μl 3 M sodium acetate and 220 μl ethanol. The pellet collected by

centrifugation as above was resuspended in 12 μl TE.

Concentration of the DNA was determined by Hoechst 33258 dye (Polysciences, Inc., Warrington, PA) fluorometry in a Hoefer TKO100 fluorimeter (Hoefer Scientific Instruments, San Francisco, CA). Approximately 2.5 μg of the size-fractionated DNA was recovered.

Thirty μg of cosmid pWE15 DNA (Stratagene, La Jolla, CA) was digested to completion with 100 units of restriction enzyme BamH 1 (NEB) in the manufacturer's buffer (final volume of 200 μl, 37°C, 1 hr). The reaction was extracted with 100 μl of PCI and DNA was precipitated from the aqueous phase by addition of 20 μl 3M sodium acetate and 550 μl -20°C absolute ethanol. After 20 min at -70°C, the DNA was collected by centrifugation (17,000 x g, 15 min), dried under vacuum, and dissolved in 180 μl of 10 mM Tris-HCl, pH 8.0. To this were added 20 μl of 10X CIP buffer

(100 mM Tris-HCl, pH 8.3; 10 mM ZnCl₂; 10 mM MgCl₂), and 1 μl (0.25 units) of 1:4 diluted calf intestinal alkaline phosphatase (Boehringer Mannheim Corporation, Indianapolis, IN). After 30 min at 37°C, the following additions were made: 2 μl 0.5 M EDTA, pH 8.0; 10 μl 10% SDS; 0.5 μl of 20 mg/ml proteinase K (as above), followed by incubation at 55°C for 30 min. Following sequential extractions with 100 μl of PCI and 100 μl phenol

(Intermountain Scientific Corporation, equilibrated with 1 M Tris-HCl, pH 8.0), the dephosphorylated DNA was precipitated by addition of 72 μl of 7.5 M ammonium acetate and 550 μl -20°C ethanol, incubation on ice for 30 min, and centrifugation as above. The pelleted DNA was washed once with 500 μl -20°C 70% ethanol, dried under vacuum, and dissolved in 20 μl of TE buffer.

Ligation of the size-fractionated Sau3A 1 fragments to the BamH 1-digested and phosphatased pWE15 vector was accomplished using T4 ligase (NEB) by a modification (i.e., use of premixed 10X ligation buffer supplied by the manufacturer) of the protocol in section 3.33 of Ausubel. Ligation was carried out overnight in a total volume of 20 μl at 15°C, followed by storage at -20°C.

Four μl of the cosmid DNA ligation reaction, containing about 1 μg of DNA, was packaged into bacteriophage lambda using a commercial packaging extract (Gigapack' III Gold Packaging

Extract, Stratagene), following the manufacturer's directions. The packaged preparation was stored at 4°C until use. The packaged cosmid preparation was used to infect Escherichia col i XLl Blue MR cells (Stratagene) according to the Gigapack^® III Gold protocols ("Titering the Cosmid Library"), as follows. XL1 Elue MR cells were grown in LB medium (g/L: Bacto-tryptone, 10; Bacto-yeast extract, 5; Bacto-agar, 15; NaCl, 5; [Difco Laboratories, Detroit, MI]) containing 0.2% (w/v) maltose plus 10 mM MgSO₄, at 37°C. After 5 hr growth, cells were pelleted at 700 x g (15 min) and resuspended in 6 ml of 10 mM MgSO₄. The culture density was adjusted with 10 mM MgSO₄ to OD₆₀₀ = 0.5. The packaged cosmid library was diluted 1:10 or 1:20 with sterile SM medium (0.1 M NaCl, 10 mM MgSO₄, 50 mM Tris-HCl pH 7.5, 0.01% w/v gelatin), and 25 μl of the diluted preparation was mixed with 25 μl of the diluted XL1 Blue MR cells. The mixture was incubated at 25°C for 30 min (without shaking), then 200 μl of LB broth was added, and incubation was continued for approximately 1 hr with occasional gentle shaking. Aliquots (20-40 μl) of this culture were spread on LB agar plates containing 100 mg/l ampicillin (i.e., LB-Amp₁₀₀) and incubated overnight at 37°C. To store the library without amplification, single colonies were picked and inoculated into individual wells of sterile 96-well microwell plates; each well containing 75 μl of Terrific Broth (TB media: 12 g/l Bacto-tryptone, 24 g/l Bacto-yeast extract, 0.4% v/v glycerol, 17 mM KH₂PO₄, 72 mM K₂HPO₄) plus 100 mg/l ampicillin (i.e., TB-Amp₁₀₀) and incubated (without shaking) overnight at 37°C. After replicating the 96-well plate into a copy plate, 75 μl/well of filter-sterilized TB:glycerol (1:1, v/v; with, or without, 100 mg/l ampicillin) was added to the plate, it was shaken briefly at 100 rpm, 37°C, and then closed with Parafilm^® (American National Can, Greenwich, CT) and placed in a -70°C freezer for storage. Copy plates were grown and processed identically to the master plates. A total of 40 such master plates (and their copies) were

prepared.

SCREENING OF THE LIBRARY WITH RADIOLABELED DNA PROBES: To prepare colony filters for probing with radioactively labeled probes, ten 96-well plates of the library were thawed at 25°C (bench top at room temperature). A replica plating tool with 96 prongs was used to inoculate a fresh 96-well copy plate

containing 75 μl/well of TB-Amp₁₀₀. The copy plate was grown overnight (stationary) at 37°C, then shaken about 30 min at 100 rpm at 37°C. A total of 800 colonies was represented in these copy plates, due to nongrowth of some isolates. The replica tool was used to inoculate duplicate impressions of the 96-well at rays onto Magna NT (MSI, Westboro, MA) nylon membranes (0.45 micron, 220 × 250 mm) which had been placed on solid LB-Amp₁₀₀ (100 ml/dish) in Bio-assay plastic dishes (Nunc, 243 × 243 × 18 mm; Curtin Mathison Scientific, Inc., Wood Dale, IL). The colonies were grown on the membranes at 37°C for about 3 hr.

A positive control colony (a bacterial clone containing a GZ4 sequence insert, see below) was grown on a separate Magna NT membrane (Nunc, 0.45 micron, 82 mm circle) on LB medium

supplemented with 35 mg/l chloramphenicol (i.e., LB-Cam₃₅). and processed alongside the library colony membranes. Bacterial colonies on the membranes were lysed, and the DNA was denatured and neutralized according to a protocol taken from the Genius™ System User's Guide version 2.0 (Boehringer Mannheim,

Indianapolis, IN). Membranes were placed colony side up on filter paper soaked with 0.5 N NaOH plus 1.5 M NaCl for 15 min to denature, and neutralized on filter paper soaked with 1 M Tris-HCl pH 8.0, 1.5 M NaCl for 15 min. After UV-crosslinking using a Stratagene UV Stratalinker set on auto crosslink, the membranes were stored dry at 25°C until use. Membranes were trimmed into strips containing the duplicate impressions of a single 96-well plate, then washed extensively by the method of section 6.4.1 in CPMB ( op. ci t . ) : 3 hr at 25°C in 3X SSC, 0.1% (w/v) SDS, followed by 1 hr at 65°C in the same solution, then rinsed in 2X SSC in preparation for the hybridization step (20X SSC = 3 M NaCl, 0.3 M sodium citrate, pH 7.0).

Amplification of a specific genomic fragment of a tcaC gene. Based on the N-terminal amino acid sequence determined for the purified TcaC peptide fraction [disclosed herein as SEQ ID NO:2], a pool of degenerate oligonucleotides (pool S4Psh) was

synthesized by standard β-cyanoethyl chemistry on an Applied

BioSystem ABI394 DNA/RNA Synthesizer (Perkin Elmer, Foster City, CA). The oligonucleotides were deprotected 8 hours at 55°C, dissolved in water, quantitated by spectrophotometric

measurement, and diluted for use. This pool corresponds to the determined N-terminal amino acid sequence of the TcaC peptide.

The determined amino acid sequence and the corresponding

degenerate DNA sequence are given below, where A, C, G, and T are the standard DNA bases, and I represents inosine:

Amino Met Gln Asp Ser Pro Glu Val Acid

S4Psh 5' ATG CA(A/G) GA(T/C) (T/A) (C/G) (T/A) CCI GA(A/G) GT 3'

Another set of degenerate oligonucleotides was synthesized (pool P2.3.5R), representing the complement of the coding strand for the determined amino acid sequence of the SEQ ID NO: 17:

Amino

Acid Ala Phe Asn Ile Asp Asp Val Codons 5' GCN TT(T/C) AA(T/C) AT(A/T/C) GA(T/C) GA(T/C) GT 3' P2.3.5R 3'CG(A/C/G/T) AA(A/G) TT(A/G) TA(T/A/G) CT(A/G) CT(A/G) CA 5'

These oligonucleotides were used as primers in Polymerase Chain Reactions (PCR^®, Roche Molecular Systems, Branchburg, NJ) to amplify a specific UNA tragment from genomic DNA prepared from Photorhabdus strain W-14 (see above). A typical reaction (50 μl) contained 125 pmol of each primer pool P2Psh and P2.3.5R, 253 ng of genomic template DNA, 10 nmol each of dATP, dCTP, dGTP, and dTTP, IX GeneAmp^® PCR buffer, and 2.5 units of AmpliTaq^® DNA polymerase (both from Roche Molecular Systems; 10X GeneAmp^® buffer is 100 mM Tris-HCl pH 8.3, 500 mM KCl, 0.01% w/v gelatin).

Amplifications were performed in a Perkin Elmer Cetus DNA Thermal Cycler (Perkin Elmer, Foster City, CA) using 35 cycles of 94°c (1.0 min), 55°C (2.0 min), 72°C (3.0 min), followed by an extension period of 7.0 min at 72°C. Amplification products were analyzed by electrophoresis through 2% w/v NuSieve^® 3:1 agarose (FMC BioProducts) in TEA buffer (40 mM Tris-acetate, 2 mM EDTA, pH 8.0). A specific product of estimated size 250 bp was observed amongst numerous other amplification products by ethidium bromide (0.5 μg/ml) staining of the gel and examination under ultraviolet light.

The region of the gel containing an approximately 250 bp product was excised, and a small plug (0.5 mm dia.) was removed and used to supply template for PCR amplification (40 cycles). The reaction (50 μl) contained the same components as above, minus genomic template DNA. Following amplification, the ends of the fragments were made blunt and were phosphorylated by

incubation at 25°C for 20 min with 1 unit of T4 DNA polymerase (NEB), 1 nmol ATP, and 2.15 units of T4 kinase (Pharmacia Biotech Inc., Piscataway, NJ).

DNA fragments were separated from residual primers by electrophoresis through 1% w/v GTG^® agarose (FMC) in TEA. A gel slice containing fragments of apparent size 250 bp was excised, and the DNA was extracted using a Qiaex kit (Qiagen Inc.,

Chatsworth, CA).

The extracted DNA fragments were ligated to plasmid vector pBC KS(+) (Stratagene) that had been digested to completion with restriction enzyme Sma 1 and extracted in a manner similar to that described for pWE15 DNA above. A typical ligation reaction (16.3 μl) contained 100 ng of digested pBC KS(+) DNA, 70 ng of 250 bp fragment DNA, 1 nmol [Co(NH₃)₆]Cl₃, and 3.9 Weiss units of T4 DNA ligase (Collaborative Biomedical Products, Bedford, MA), in 1X ligation buffer (50 mM Tris-HCl, pH 7.4; 10 mM MgCl₂; 10 mM dithiothreitol; 1 mM spermidine, 1 mM ATP, 100 mg/ml bovine serum albumin). Following overnight incubation at 14°C, the ligated products were transformed into frozen, competent Escherichia coli DH5α cells (Gibco BRL) according to the suppliers'

recommendations, and plated on LB-Cam₃₅ plates, containing IPTG (119 μg/ml) and X-gal (50 μg/ml). Independent white colonies were picked, and plasmid DNA was prepared by a modified alkaline-lysis/PEG precipitation method (PRISM™ Ready Reaction DyeDeoxy™ Terminator Cycle Sequencing Kit Protocols; ABI/Perkin Elmer). The nucleotide sequence of both strands of the insert DNA was determined, using T7 primers [pBC KS(+) bases 601-623:

TAAAACGACGGCCAGTGAGCGCG) and LacZ primers [pBC KS(+) bases 792-816: ATGACCATGATTACGCCAAGCGCGC) and protocols supplied with the PRISM™ sequencing kit (ABI/Perkin Elmer). Nonincorporated dye-terminator dideoxyribonucleotides were removed by passage through Centri-Sep 100 columns (Princeton Separations, Inc., Adelphia, NJ) according to the manufacturer's instructions. The DNA sequence was obtained by analysis of the samples on an ABI Model 373A DNA Sequencer (ABI/Perkin Elmer). The DNA sequences of two isolates, GZ4 and HB14, were found to be as illustrated in Figure 1.

This sequence illustrates the following features: 1) bases 1-20 represent one of the 64 possible sequences of the S4Psh degenerate oligonucleotides, ii) the sequence of amino acids 1-3 and 6-12 correspond exactly to that determined for the N-terminus of TcaC (disclosed as SEQ ID NO:2), iii) the fourth amino acid encoded is a cysteine residue rather than serine. This difference is encoded within the degeneracy for the serine codons (see above), iv) the fifth amino acid encoded is proline,

corresponding to the TcaC N-terminal sequence given as SEQ ID NO: 2, v) bases 257-276 encode one of the 192 possible sequences designed into the degenerate pool, vi) the TGA termination codon introduced at bases 268-270 is the result of complementarity to the degeneracy built into the oligonucleotide pool at the corresponding position, and does not indicate a shortened reading frame for the corresponding gene.

Labeling of a TcaC peptide gene-specific probe. DNA fragments corresponding to the above 276 bases were amplified (35 cycles) by PCR^® in a 100 μl reaction volume, using 100 pmol each of P2Psh and P2.3.5R primers, 10 ng of plasmids GZ4 or HB14 as templates, 20 nmol each of dATP, dCTP, dGTP, and dTTP, 5 units of AmpliTAq^® DNA polymerase, and IX concentration of GeneAmp^® buffer, under the same temperature regimes as described above. The amplification products were extracted from a 1% GTG^® agarose gel by Qiaex kit and quantitated by fluorometry.

The extracted amplification products from plasmid HB14 template (approximately 400 ng) were split into five aliquots and labeled with ³²P-dCTP using the High Prime Labeling Mix

(Boehringer Mannheim) according to the manufacturer's

instructions. Nonincorporated radioisotope was removed by passage through NucTrap^® Probe Purification Columns (Stratagene), according to the supplier's instructions. The specific activity of the labeled DNA product was determined by scintillation counting to be 3.11 × 10⁸ dpm/μg. This labeled DNA was used to probe membranes prepared from 800 members of the genomic library.

Screening with a TcaC-peptide gene specific probe. The radiolabeled HB14 probe was boiled approximately 10 min, then added to "minimal hyb" solution. [Note: The "minimal hyb" method is taken from a CERES protocol; "Restriction Fragment Length Polymorphism Laboratory Manual version 4.0", sections 4-40 and 4-47; CERES/NPI, Salt Lake City, UT. NPI is now defunct, with its successors operating as Linkage Genetics]. "Minimal hyb" solution contains 10% w/v PEG (polyethylene glycol, M.W. approx. 8000), 7% w/v SDS; 0.6X SSC, 10 mM sodium phosphate buffer (from a 1M stock containing 95 g/l NaH₂PO₁·1H₂O and 84.5 g/l

Na₂HPO₄·7H₂O), 5 mM EDTA, and 100 mg/ml denatured salmon sperm DNA. Membranes were blotted dry briefly then, without

prehybridization, 5 strips of membrane were placed in each of 2 plastic boxes containing 75 ml of "minimal hyb" and 2.6 ng/ml of radiolabeled HB14 probe. These were incubated overnight with slow shaking (50 rpm) at 60°C. The filters were washed three times for approximately 10 min each at 25°C in "minimal hyb wash solution" (0.25X SSC, 0.2% SDS), followed by two 30-min washes with slow shaking at 60°C in the same solution. The filters were placed on paper covered with Saran Wrap^® (Dow Brands,

Indianapolis, IN) in a light-tight autoradiographic cassette and exposed to X-Omat X-ray film (Kodak, Rochester, NY) with two DuPont Cronex Lightning-Plus Cl enhancers (Sigma Chemical Co., it. Louis, MO), for 4 hr at -70°C. Upon development (standard photographic procedures), significant signals were evident in both replicates amongst a high background of weaker, more irregular signals. The filters were again washed for about 4 hr at 68°C in "minimal hyb wash solution" and then placed again in the cassettes and film was exposed overnight at -70°C. Twelve possible positives were identified due to strong signals on both of the duplicate 96-well colony impressions. No signal was seen with negative control membranes (colonies of XL1 Blue MR cells containing pWE15), and a very strong signal was seen with positive control membranes (DH5α cells containing the GZ4 isolate of the PCR product) that had been processed concurrently with the experimental samples.

The twelve putative hybridization-positive colonies were retrieved from the frozen 96-well library plates and grown overnight at 37°C on solid LB-Amp₁₀₀ medium. They were then patched (3/plate, plus three negative controls: XL1 Blue MR cells containing the pWE15 vector) onto solid LB-Amp₁₀₀. Two sets of membranes (Magna NT nylon, 0.45 micron) were prepared for hybridization. The first set was prepared by placing a filter directly onto the colonies on a patch plate, then removing it with adherent bacterial cells, and processing as below. Filters of the second set were placed on plates containing LB-Amp₁₀₀ medium, then inoculated by transferring cells from the patch plates onto the filters. After overnight growth at 37°C, the filters were removed from the plates and processed.

Bacterial cells on the filters were lysed and DNA denatured by placing each filter colony-side-up on a pool (1.0 ml) of 0.5 N NaOH in a plastic plate for 3 min. The filters were blotted dry on a paper towel, then the process was repeated with fresh 0.5 N NaOH. After blotting dry, the filters were neutralized by placing each on a 1.0 ml pool of 1 M Tris-HCl, pH 7.5 for 3 min, blotted dry, and reneutralised with fresh buffer. This was followed by two similar soakings (5 min each) on pools of 0.5 M Tris-HCl pH 7.5 plus 1.5 M NaCl. After blotting dry, the DNA was UV crosslinked to the filter (as above), and the filters were washed (25°C, 100 rpm) in about 100 ml of 3X SSC plus 0.1% (w/v) SDS (4 times, 30 min each with fresh solution for each wash). They were then placed in a minimal volume of prehybridizat ion solution [6X SSC plus 1% w/v each of Ficoll 400 (Pharmacia), polyvinylpyrrolidone (av. M.W. 360,000; Sigma ) and bovine serum albumin Fraction V; (Sigma)] for 2 hr at 65°C, 50 rpm. The prehybridization solution was removed, and replaced with the HB14 3²P-labeled probe that had been saved from the previous

hybridization of the library membranes and which had been denatured at 95°C for 5 min. Hybridization was performed at 60°C for 16 hr with shaking at 50 rpm.

Following removal of the labeled probe solution, the membranes were washed 3 times at 25°C (50 rpm, 15 min) in 3X SSC (about 150 ml each wash). They were then washed for 3 hr at 68°C (50 rpm) in 0.25X SSC plus 0.2% SDS (minimal hyb wash solution), and exposed to X-ray film as described above for 1.5 hr at 25°C (no enhancer screens). This exposure revealed very strong hybridization signals to cosmid isolates 22G12, 25A10, 26A5, and 26B10, and a very weak signal with cosmid isolate 8B10. No signal was seen with the negative control (pWE15) colonies, and a very strong signal was seen with positive control membranes (DH5α cells containing the GZ4 isolate of the PCR product) that had been processed concurrently with the experimental samples.

Amplification of a specific genomic fragment of a tcaB gene. Based on the N-terminal amino acid sequence determined for the purified TcaB, peptide fraction (disclosed here as SEQ ID NO: 3) a pool of degenerate oligonucleotides (pool P8F) was synthesized as described for peptide TcaC. The determined amino acid sequence and the corresponding degenerate DNA sequence are given below, where A, C, G, and T are the standard DNA bases, and I represents inosine:

Another set of degenerate oligonucleotides was synthesized (pool P8.108.3R), representing the complement of the coding strand for the determined amino acid sequence of the TcaB_i-PT108 internal peptide (disclosed herein as SEQ ID NO:20):

These oligonucleotides were used as primers for PCR^® using

HotStart 50 Tubes™ (Molecular Bio-Products, Inc., San Diego, CA) to amplify a specific DNA fragment from genomic DNA prepared from

Photorhabdus strain W-14 (see above). A typical reaction (50 μl) contained (bottom layer) 25 pmol of each primer pool P8F and

P8.108.3R, with 2 nmol each of dATP, dCTP, dGTP, and dTTP, in IX GeneAmp^® PCR buffer, and (top layer) 230 ng of genomic template DNA, 8 nmol each of dATP, dCTP, dGTP, and dTTP, and 2.5 units of AmpliTaq^® DNA polymerase, in IX GeneAmp^® PCR buffer.

Amplifications were performed by 35 cycles as described for the TcaC peptide. Amplification products were analyzed by

electrophoresis through 0.7% w/v SeaKem^® LE agarose (FMC) in TEA buffer. A specific product of estimated size 1600 bp was observed.

Four such reactions were pooled, and the amplified DNA was extracted from a 1.0% SeaKem^® LE gel by Qiaex kit as described for the TcaC peptide. The extracted DNA was used directly as the template for sequence determination (PRISM Sequencing Kit) using, the P8F and P8.108.3R primer pools. Each reaction contained about 100 ng template DNA and 25 pmol of one primer pool, and was processed according to standard protocols as described for the TcaC peptide. An analysis of the sequence derived from extension of the P8F primers revealed the short DNA sequence (and encoded amino acid sequence):

GAT GCA TTG NTT GCT

Asp Ala Leu (Val) Ala

which corresponds to a portion of the N-terminal peptide sequence disclosed as SEQ ID NO:3 (TcaB_i). Labeling of a TcaB_i-peptide gene-specific probe.

Approximately 50 ng of gel-purified TcaB_i DNA fragment was labeled with ³²P-dCTP as described above, and nonincorporated radioisotopes were removed by passage through a NICK Column^® (Pharmacia). The specific activity of the labelled DNA was determined to be 6 × 10⁹ dpm/μg. This labeled DNA was used to probe colony membranes prepared from members of the genomic library that had hybridized to the TcaC-peptide specific probe.

The membranes containing the 12 colonies identified in the TcaC-probe library screen (see above) were stripped of

radioactive TcaC-specific label by boiling twice for

approximately 30 min each time in 1 liter of 0.1X SSC plus 0.1 % SDS. Removal of radiolabel was checked with a 6 hr film

exposure. The stripped membranes were then incubated with the TcaB_i peptide-specific probe prepared above. The labeled DNA was denatured by boiling for 10 min, and then added to the filters that had been incubated for 1 hr in 100 ml of "minimal hyb" solution at 60°C. After overnight hybridization at this temperature, the probe solution was removed, and the filters were washed as follows (all in 0.3X SSC plus 0.1% SDS): once for 5 min at 25°C, once for 1 hr at 60°C in fresh solution, and once for 1 hr at 63°C in fresh solution. After 1.5 hr exposure to X-ray film by standard procedures, 4 strongly-hybridizing colonies were observed. These were, as with the TcaC-specific probe, isolates 22G12, 25A10, 26A5, and 26B10.

The same TcaB_i probe solution was diluted with an equal volume (about 100 ml) of "minimal hyb" solution, and then used to screen the membranes containing the 800 members of the genomic library. After hybridization, washing, and exposure to X-ray film as described above, only the four cosmid clones 22G12, 25A10, 26A5, and 26B10, were found to hybridize strongly to this probe.

ISOLATION OF SUBCLONES CONTAINING GENES ENCODING TcaC AND TcaBj PEPTIDES, AND DETERMINATION OF DNA BASE SEQUENCE THEREOF: Three hybridization-positive cosmids in strain XL1 Blue MR were grown with shaking overnight (200 rpm) at 30°C in 100 ml TB-Amp₁₀₀. After harvesting the cells by centrifugation, cosmid DNA was prepared using a commercially available kit (BIGprep™, 5 Prime 3 Prime, Inc., Boulder, CO), following the manufacturer's protocols. Only one cosmid, 26A5, was successfully isolated by this procedure. When digested with restriction enzyme EcoR 1 (NEB) and analyzed by gel electrophoresis, fragments of

approximate sizes 14, 10, 8 (vector), 5, 3.3, 2.9, and 1.5 kbp were detected. A second attempt to isolate cosmid DNA from the same three strains (8 ml cultures; TB-Amp₁₀₀, 30°C) utilized a boiling miniprep method (Evans G. and G Wahl., 1987, "Cosmid vectors for genomic walking and rapid restriction mapping." in Guide to Molecular Cloning Techniques. Meth. Enzymology, vol. 152, S. Berger and A. Kimmel, eds., pgs. 604-610). Only one cosmid, 25A10, was successfully isolated by this method. When digested with restriction enzyme EcoR 1 (NEB) and analyzed by gel electrophoresis, this cosmid showed a fragmentation pattern identical to that previously seen with cosmid 26A5.

A 0.15 μg sample of 26A5 cosmid DNA was used to transform 50 ml of E. col i DH5α cells (Gibco BRL), by the supplier's

protocols. A single colony isolate of that strain was inoculated into 4 ml of TB-Amp₁₀₀, and grown for 8 hr at 37°C.

Chloramphenicol was added to a final concentration of 225 μg/ml, incubation was continued for another 24 hr, then cells were harvested by centrifugation and frozen at -20°C. Isolation of the 26A5 cosmid DNA was by a standard alkaline lysis miniprep (Maniatis et al . , op . ci t . , p. 382), modified by increasing all volumes by 50% and with stirring or gentle mixing, rather than vortexing, at every step. After washing the DNA pellet in 70% ethanol, it was dissolved in TE containing 25 μg/ml ribonuclease A (Boehringer Mannheim).

Identification of EcoR 1 fragments hybridizing to GZ4-derived and TcaB_i- probes. Approximately 0.4 μg of cosmid 25A10 (from XL1 Blue MR cells) and about 0.5 μg of cosmid 26A5 (from chloramphenicol-amplified DH5α cells) were each digested with about 15 units of EcoR 1 (NEB) for 85 min, frozen overnight, then heated at 65°C for five min, and electrophoresed in a 0.7% agarose gel (Seakem^® LE, 1X TEA, 80 volts, 90 min). The DNA was stained with ethidium bromide as described above, and

photographed under ultraviolet light. The EcoR 1 digest of cosmid 25A10 was a complete digestion, but the sample of cosmid 26A5 was only partially digested under these conditions. The agarose gel containing the DNA fragments was subjected to depurination, denaturation and neutralization, followed by

Southern blotting onto a Magna NT nylon membrane, using a high salt (20X SSC) protocol, all as described in section 2.9 of Ausubel et al . (CPMB, op . ci t . ) . The transferred DNA was then UV-crosslinked to the nylon membrane as before. An TcaC-peptide specific DNA fragment corresponding to the insert of plasmid isolate GZ4 was ampl i f ied by PCR^® in a 100 ml reaction volume as described previously above. The amplification products from three such reactions were pooled and were extracted from a 1% GTG^® agarose gel by Qiaex kit, as described above, and quantitated by fluorometry. The gel-purified DNA (100 ng) was labeled with ³²P-dCTP using the High Prime Labeling Mix

(Boehringer Mannheim) as described above, to a specific activity of 6.34 × 10⁸ dpm/μg.

The ³²P-labeled GZ4 probe was boiled 10 min, then added to

"minimal hyb" buffer (at 1 ng/ml), and the Southern blot membrane containing the digested cosmid DNA fragments was added, and incubated for 4 hr at 60°C with gentle shaking at 50 rpm. The membrane was then washed 3 times at 25°C for about 5 min each (minimal hyb wash solution), followed by two washes for 30 min each at 60°C. The blot was exposed to film (with enhancer screens) for about 30 min at -70°C. The GZ4 probe hybridized strongly to the 5.0 kbp (apparent size) EcoR 1 fragment of both these two cosmids, 26A5 and 25A10.

The membrane was stripped of radioactivity by boiling for about 30 min in 0.1X SSC plus 0.1 % SDS, and absence of

radiolabel was checked by exposure to film. It was then

hybridized at 60°C for 3.5 hours with the (denatured! TcaB_i probe in "minimal hyb" buffer previously used for screening the colony membranes (above), washed as described previously, and exposed to film for 40 min at -70°C with two enhancer screens. With both cosmids, the TcaB_i probe hybridized lightly with the about 5.0 kbp EcoR 1 fragment, and strongly with a fragment of

approximately 2.9 kbp.

The sample of cosmid 26A5 DNA previously described, (from

DH5α cells) was used as the source of DNA from which to subclone the bands of interest. This DNA (2.5 μg) was digested with about 3 units of EcoR 1 (NEB) in a total volume of 30 μl for 1.5 hr, to give a partial digest, as confirmed by gel electrophoresis. Ten μg of pBC KS (+) DNA (Stratagene) were digested for 1.5 hr with 20 units of EcoR 1 in a total volume of 20 μl, leading to total digestion as confirmed by electrophoresis. Both EcoR 1-cut DNA preparations were diluted to 50 μl with water, to each an equal volume of PCI was added, the suspension was gently mixed, spun in a microcentrif uge and the aqueous supernatant was collected DNA was precipitated by 150 μl ethanol, and the mixture was placed at -20°C overnight. Following centrifugation and drying, the EcoR 1-digested pBC KS (+) was dissolved in 100 μl TE; the partially digested 26A5 was dissolved in 20 μl TE. DNA recovery was checked by fluorometry

In separate reactions, approximately 60 ng of EcoR 1-digested pBC KS(+) DNA was ligated with approximately 180 ng or 270 ng of partially digested cosmid 26A5 DNA. Ligations were carried out in a volume of 20 μl at 15°C for 5 hr, using T4 ligase and buffer from New England BioLabs. The ligation mixture, diluted to 100 μl with sterile TE, was used to transform frozen, competent DH5α cells (Gibco BRL) according to the supplier's instructions. Varying amounts (25-200 μl) of the transformed cells were plated on freshly prepared solid LB-Cam₃₅ medium with 1 mM IPTG and 50 mg/l X-gal. Plates were incubated at 37°C about 20 hr, then chilled in the dark for approximately 3 hr to intensify color for insert selection. White colonies were picked onto patch plates of the same composition and incubated overnight at 37°C.

Two colony lifts of each of the selected patch plates were prepared as follows. After picking white colonies to fresh plates, round Magna NT nylon membranes were pressed onto the patch plates, the membrane was lifted off, and subjected to denaturation, neutralization and UV crosslinking as described above for the library colony membranes. The crosslinked colony lifts were vigorously washed, including gently wiping off the excess cell debris with a tissue. One set was hybridized with the GZ4(TcaC) probe solution described earlier, and the other set was hybridized with the TcaB_i probe solution described earlier, according to the 'minimal hyb' protocol, followed by washing and film exposure as described for the library colony membranes.

Colonies showing hybridization signals either only with the GZ4 probe, with both GZ4 and TcaB_i probes, or only with the TcaB_i probe, were selected for further work and cells were streaked for single colony isolation onto LB-Cam₃₅ media with IPTG and X-gal as before. Approximately 35 single colonies, from 16 different isolates, were picked into liquid LB-Cam₃₅ media and grown overnight at 37°C; the cells were collected by centrifugation and plasmid DNA was isolated by a standard alkaline lysis miniprep according to Maniatis et al. (op. cit. p. 368). DNA pellets were dissolved in TE + 25 μg/ml ribonuclease A and DNA concentration was determined by fluorometry. The EcoR 1 digestion pattern was analyzed by gel electrophoresis. The following isolates were picked as useful. Isolate A17.2 contains religated pBC KS (+) only and was used for a (negative) control. Isolates D38.3 and C44.1 each contain only the 2.9 kbp, TcaB_i -hybridizing EcoR 1 fragment inserted into pBC KS(+). These plasmids, named pDAB2000 and pDAB2001, respectively, are illustrated in Fig. 2.

Isolate A35.3 contains only the approximately 5 kbp, GZ4)-hybridizing EcoR 1 fragment, inserted into pBC KS(+). This plasmid was named pDAB2002 (also Fig. 2). These isolates provided templates for DNA sequencing.

Plasmids pDAB2000 and pDAB2001 were prepared using the BIGprep™ kit as before. Cultures (30 ml) were grown overnight in TB-Cam₃₅ to an OD₆₀₀ of 2, then plasmid was isolated according to the manufacturer's directions. DNA pellets were redissolved in 100 μl TE each, and sample integrity was checked by EcoR 1 digestion and gel electrophoretic analysis.

Sequencing reactions were run in duplicate, with one replicate using as template pDAB2000 DNA, and the other replicate using as template pDAB2001 DNA. The reactions were carried out using the dideoxy dye terminator cycle sequencing method, as described above for the sequencing of the GZ4/HB14 DNAs. Initial sequencing runs utilized as primers the LacZ and T7 primers described above, plus primers based on the determined sequence of the TcaB_i PCR amplification product (TH1 =

ATTGCAGACTGCCAATCGCTTCGG, TH12 = GAGAGTATCCAGACCGCGGATGATCTG).

After alignment and editing of each sequencing output, each was truncated to between 250 to 350 bases, depending on the integrity of the chromatographic data as interpreted by the Perkin Elmer Applied Biosystems Division SeqEd 675 software.

Subsequent sequencing "steps" were made by selecting appropriate sequence for new primers. With a few exceptions, primers

(synthesized as described above) were 24 bases in length with a 50% G+C composition. Sequencing by this method was carried out on both strands of the approximately 2.9 kbp EcoR 1 fragment. To further serve as template tor DNA sequencing, plasmid DNA from isolate pDAB2002 was prepared by BIGprep™ kit. Sequencing reactions were performed and analyzed as described above.

Initially, a T3 primer (pBS SK (+) bases 774-796:

CGCGCAATTAACCCTCACTAAAG) and a T7 primer (pBS KS (+) bases 621-643: GCGCGTAATACGACTCACTATAG) were used to prime the sequencing reactions from the flanking vector sequences, reading into the insert DNA. Another set of primers, (GZ4F:

GTATCGATTACAACGCTGTCACTTCCC; TH13: GGGAAGTGACAGCGTTGTAATCGATAC, TH14: ATGTTGGGTGCGTCGGCTAATGGACATAAC; and LW1-204:

GGGAAGTGACAGCGTTGTAATCGATAC) was made to prime from internal sequences, which were determined previously by degenerate oligonucleotide-mediated sequencing of subcloned TcaC-peptide PCR products. From the data generated during the initial rounds of sequencing, new sets of primers were designed and used to walk the entire length of the ~5 kbp fragment. A total of 55 oligo primers was used, enabling the identification of 4832 total bp or contiguous sequence.

When the DNA sequence of the EcoR 1 fragment insert of pDAB2002 is combined with part of the determined sequence of the pDAB2000/pDAB2001 isolates, a total contiguous sequence of 6005 bp was generated (disclosed herein as SEQ ID NO:25). When long open reading frames were translated into the corresponding amino acids, the sequence clearly shows the TcaB_i N-terminal peptide (disclosed as SEQ ID NO:3), encoded by bases 19-75, immediately following a methionine residue (start of translation). Upstream lies a potential ribosome binding site (bases 1-9), and

downstream, at bases 166-228 is encoded the TcaB_i-PT158 internal peptide (disclosed herein as SEQ ID NO:19). Further downstream, in the same reading frame, at bases 1738-1773, exists a sequence encoding the TcaB_i-PT108 internal peptide (disclosed herein as SEQ ID NO: 20). Also in the same reading frame, at bases 1897-1923, is encoded the TcaB_ii N-terminal peptide (disclosed herein as SEQ ID NO:5), and the reading frame continues uninterrupted to a translation termination codon at nucleotides 3586-3588.

The lack of an in-frame stop codon between the end of the sequence encoding TcaB_i-PT108 and the start of the TcaB_ii encoding region, and the lack of a discernible ribosome binding site immediately upstream of the TcaB_ii coding region, indicate that peptides TcaB_ii and TcaB_i are encoded by a single open reading frame of 3567 bp beginning at base pair 16 in SEQ ID NO:25), and are most likely derived from a single primary gene product of 1189 amino acids (131,586 Daltons; disclosed herein as SEQ ID NO:26) by post-translational cleavage. If the amino acid immediately preceding the TcaB_ii N-terminal peptide represents the C-terminal amino acid of peptide TcaB_i, then the predicted mass of TcaB_ii (627 amino acids) is 70,814 Daltons (disclosed herein as SEQ ID NO:28), somewhat higher than the size observed by SDS-PAGE (68 kDa). This peptide would be encoded by a contiguous stretch of 1881 base pairs (disclosed herein as SEQ ID NO:27). it is thought that the native C-terminus of TcaB_i lies somewhat closer to the C-terminus of TcaB_i-PT108. The molecular mass of PT108 [3.438 kDa; determined during N-terminal amino acid sequence analysis of this peptide] predicts a size of 30 amino acids. Using the size of this peptide to designate the C-terminus of the TcaB_i coding region [Glu at position 604 of SEQ ID NO:28], the derived size of TcaB_i is determined to be 604 amino acids or 68,463 Daltons, more in agreement with

experimental observations.

Translation of the TcaB_ii peptide coding region of 1686 base pairs (disclosed herein as SEQ ID NO:29) yields a protein of 562 amino acids (disclosed herein as SEQ ID NO: 30) with predicted mass of 60,789 Daltons, which corresponds well with the observed 61 kDa.

A potential ribosome binding site (bases 3633-3638) is found 48 bp downstream of the stop codon for the tcaB open reading frame. At bases 3645-3677 is found a sequence encoding the N-terminus of peptide TcaC, (disclosed as SEQ ID NO.2). The open reading frame initiated by this N-terminal peptide continues uninterrupted to base 6005 (2361 base pairs, disclosed herein as the first 2361 base pairs of SEQ ID NO.31). A gene ( tcaC) encoding the entire TcaC peptide, (apparent size ~165 kDa; ~1500 amino acids), would comprise about 4500 bp.

Another isolate containing cloned EcoR 1 fragments of cosmid 26A5, E20.6, was also identified by its homology to the

previously mentioned GZ4 and TcaB_i probes. Agarose gel analysis of EcoR 1 digests of the DNA of the plasmid harbored by this strain (pDAB2004, Fig. 2), revealed insert fragments of estimated sizes 2.9, 5, and 3.3 kbp. DNA sequence analysis initiated from primers designed from the sequence of plasmid pDAB2002 revealed that the 3.3 kbp EcoR 1 fragment of pDAB2004 lies adjacent to the 5 kbp EcoR 1 fragment represented in pDAB2002. The 2361 base pair open reading frame discovered in pDAB2002 continues

uninterrupted for another 2094 bases in pDAB2004 [disclosed herein as base pairs 2362 to 4458 of SEQ ID NO:31]. DNA sequence analysis using the parent cosmid 26A5 DNA as template confirmed the continuity of the open reading frame. Altogether, the open reading frame ( TcaC SEQ ID NO: 31) comprises 4455 base pairs, and encodes a protein (TcaC) of 1485 amino acids [disclosed herein as SEQ ID NO:32]. The calculated molecular size of 166,214 Daltons is consistent with the estimated size of the TcaC peptide (165 kDa), and the derived amino acid sequence matches exactly that disclosed for the TcaC N-terminal sequence [SEQ ID NO:2].

The lack of an amino acid sequence corresponding to SEQ ID NO: 17; used to design the degenerate oligonucleotide primer pool in the discovered sequence indicates that the generation of the PCR® products found in isolates GZ4 and HB14, which were used as probes in the initial library screen, were fortuitously generated by reverse-strand priming by one of the primers in the degenerate pool. Further, the derived protein sequence does not include the internal fragment disclosed herein as SEQ ID NO: 18. These sequences reveal that plasmid pDAB2004 contains the complete coding region for the TcaC peptide.

Example 9

Screening of the Photorhabdus genomic library for genes encoding the TcbA_ii peptide

This example describes a method used to identify DNA clones that contain the TcbA_ii peptide-encoding genes, the isolation of the gene, and the determination of its partial DNA base sequence Primers and PCR reactions

The TcbA_ii polypeptide of the insect active preparation is ~206 kDa. The amino acid sequence of the N-terminus of this peptide is disclosed as SEQ ID NO:1. Four pools of degenerate

oligonucleotide primers ("Forward primers": TH-4, TH-5, TH-6, and TH-7) were synthesized to encode a portion of this amino acid sequence, as described in Example 8, and are shown below.

In addition, a primary ("a") and a secondary ("b") sequence of an internal peptide preparation (TcbA_ii-PT81) have been determined and are disclosed herein as SEQ ID No:23 and SEQ ID No:24, respectively. Four pools of degenerate oligonucleotides ("Reverse Primers": TH-8, TH-9, TH-10 and TH-11) were similarly designed and synthesized to encode the reverse complement of sequences that encode a portion of the peptide of SEQ ID NO: 23, as shown below.

Sets of these primers were used in PCR" reactions to amplity TcbA_ii- encoding gene fragments from the genomic Photorhabdus luminescens W-14 DNA prepared in Example 6. All PCR^® reactions were run with the "Hot Start" technique using AmpliWax™ gems and other Perkin Elmer reagents and protocols. Typically, a mixture (total volume 11 μl) of MgCl₂, dNTP's, 10X GeneAmp^® PCR Buffer II, and the primers were added to tubes containing a single wax bead.

[10X GeneAmp^® PCR Buffer II is composed of 100 mM Tris-HCl, pH 8.3; and 500 mM KCl.] The tubes were heated to 80°C for 2 minutes and allowed to cool. To the top of the wax seals, a solution containing 10X GeneAmp^® PCR Buffer II, DNA template, and AmpliTaq^® DNA polymerase were added. Following melting of the wax seal and mixing of components by thermal cycling, final reaction conditions (volume of 50 μl) were: 10 mM Tris-HCl, pH 8.3; 50 mM KCl; 2.5 mM MgCl₂; 200 μM each in dATP, dCTP, dGTP, dTTP; 1.25 mM in a single Forward primer pool; 1.25 μM in a single Reverse primer pool, 1.25 units of AmpliTaq^® DNA polymerase, and 170 ng of template DNA.

The reactions were placed in a thermocycler (as in

Example 8) and run with the following program:

A series of amplifications was run at three different annealing temperatures (55°, 60°, 65° C) using the degenerate primer pools. Reactions with annealing at 65°C had no

amplification products visible following agarose gel

electrophoresis. Reactions having a 60°C annealing regime and containing primers TH-5+TH-10 produced an amplification product that had a mobility corresponding to 2.9 kbp. A lesser amount of the 2.9 kbp product was produced under these conditions with primers TH-7+TH-10. When reactions were annealed at 55°C, these primer pairs produced more of the 2.9 kbp product, and this product was also produced by primer pairs TH-5+TH-8 and TH-5+TH- 11. Additional very faint 2.9 kbp bands were seen in lanes containing amplification products from primer pairs TH-7 plus TH- 8, TH-9, TH-10, or TH-11.

To obtain sufficient PCR amplification product for cloning and DNA sequence determination, 10 separate PCR reactions were set up using the primers TH-5+TH-10, and were run using the above conditions with a 55°C annealing temperature. All reactions were pooled and the 2.9 kbp product was purified by Qiaex extraction from an agarose gel as described above.

Additional sequences determined for TcbA_ii internal peptides are disclosed herein as SEQ ID NO: 21 and SEQ ID NO: 22. As before, degenerate oligonucleotides (Reverse primers TH-17 and TH-18) were made corresponding to the reverse complement of sequences that encode a portion of the amino acid sequence of these peptides.

Degenerate oligonucleotides TH-18 and TH-17 were used in an amplification experiment with Photorhabdus l uminescens W-14 DNA as template and primers TH-4, TH-5, TH-6, or TH-7 as the 5'- (Forward) primers. These reactions amplified products of approximately 4 kbp and 4.5 kbp, respectively. These DNAs were transferred from agarose gels to nylon membranes and hybridized with a ³²P-labeled probe (as described above) prepared from the 2.9 kbp product amplified by the TH-5+TH10 primer pair. Both the 4 kbp and the 4.5 kbp amplification products hybridized strongly to the 2.9 kbp probe. These results were used to construct a map ordering the TcbA_ii internal peptide sequences as shown in

Fig. 3. Approximate distances between the primers are shown in nucleotides in Fig. 3. DNA Sequence of the 2.9 kbp TcbA_ii-encoding fragment

Approximately 200 ng of the purified 2.9 kbp fragment

(prepared above) was precipitated with ethanol and dissolved in 17 ml of water. One-half of this was used as sequencing template with 25 pmol of the TH-5 pool as primers, the other half was used as template for TH-10 priming. Sequencing reactions were as given in Example 8. No reliable sequence was produced using the TH-10 primer pool; however, reactions with TH-5 primer pool produced the sequence disclosed below:.

Based on this sequence, a sequencing primer (TH-21, 5'-CCGGGCGACGTTTATCTAGG-3') was designed to reverse complement bases 120-139, and initiate polymerization towards the 5' end (i.e., TH-5 end) of the gel-purified 2.9 kbp TcbA_ii-encoding PCR fragment. The determined sequence is shown below, and is compared to the biochemically determined N-terminal peptide sequence of TcbAii SEQ ID NO:1.

From the homology of the derived amino acid sequence to the biochemically determined one, it is clear that the 2.9 kbp PCR fragment represents the TcbA coding region. This 2.9 kbp fragment was then used as a hybridization probe to screen the Photorhabdus W-14 genomic library prepared in Example 8 for cosmids containing the TcbA_ii-encoding gene.

Screening the Photorhabdus cosmid library

The 2.9 kb gel-purified PCR fragment was labeled with "p using the Boehringer Mannheim High Prime labeling kit as described in Example 8. Filters containing remnants of

approximately 800 colonies from the cosmid library were screened as described previously (Example 8), and positive clones were streaked for isolated colonies and rescreened. Three clones (8A11, 25G8, and 26D1) gave positive results through several screening and characterization steps. No hybridization of the TcbA_ii-specific probe was ever observed with any of the four cosmids identified in Example 8, and which contain the tcaB and tcaC genes. DNA from cosmids 8A11, 25G8, and 26D1 was digested with restriction enzymes Bgl 2, EcoR 1 or Hind 3 (either alone or in combination with one another), and the fragments were

separated on an agarose gel and transferred to a nylon membrane as described in Example 8. The membrane was hybridized with ³²P- labeled probe prepared from the 4.5 kbp fragment (generated by amplification of Photorhabdus genomic DNA with primers TH-5+TH- 17). The patterns generated from cosmid DNAs 8A11 and 26D1 were identical to those generated with similarly-cut genomic DNA on the same membrane. It is concluded that cosmids 8A11 and 26D1 are accurate representations of the genomic TcbA_ii encoding locus. However, cosmid 25G8 has a single Bgl 2 fragment which is slightly larger than the genomic DNA. This may result from positioning of the insert within the vector. DNA sequence of the tcbA-encoding gene

The membrane hybridization analysis of cosmid 26D1 revealed that the 4.5 kbp probe hybridized to a single large EcoR 1

fragment (greater than 9 kbp) This fragment was gel purified and ligated into the EcoR 1 site of pBC KS (+) as described in

Example 8, to generate plasmid pBC-Sl/Rl. The partial DNA

sequence of the insert DNA of this plasmid was determined by

"primer walking" from the flanking vector sequence, using

procedures described in Example 8 Further sequence was

generated by extension from new oligonucleotides designed from the previously determined sequence. When compared to the

determined DNA sequence for the tcbA gene identified by other

methods (disclosed herein as SEQ ID NO: 11 as described in Example 12 below), complete homology was found to nucleotides 1-272, 319-826, 2578-3036, and 3068-3540 (total bases = 1712). It was

concluded that both approaches can be used to identify DNA

fragments encoding the TcbA_ii peptide.

Analysis of the derived amino acid sequence of the tcbA gene.

The sequence of the DNA fragment identified as SEQ ID NO: 11 encodes a protein whose derived amino acid sequence is disclosed herein as SEQ ID NO: 12. Several features verify the identity of the gene as that encoding the TcbA_ii protein. The TcbA_ii N-terminal peptide (SEQ ID NO:1; Phe He Gln Gly Tyr Ser Asp Leu Phe Gly Asn Arg Ala) is encoded as amino acids 88-100. The TcbA_ii internal peptide TcbA_ii- PT81(a) (SEQ ID NO:23) is encoded as amino acids 1065-1077, and TcbA_ii-PT81(b) (SEQ ID NO:24) is encoded as amino acids 1571-1592. Further, the internal peptide TcbA_ii-PT56 (SEQ ID NO:22) is encoded as amino acids 1474-1488, and the internal peptide TcbA_ii-PT103 (SEQ ID NO:24) is encoded as amino acids 1614-1639. It is obvious that this gene is an authentic clone encoding the TcbA_ii peptide as isolated from insecticidal protein preparations of Photorhabdus l uminescens strain W-14.

The protein isolated as peptide TcbA_ii is derived from cleavage of a longer peptide. Evidence for this is provided by the fact that the nucleotides encoding the TcbA_ii N-terminal peptide SEQ ID NO: 1 are preceded by 261 bases (encoding 87 N-terminal-proximal amino acids) of a longer open reading frame (SEQ ID NO:11). This reading frame begins with nucleotides that encode the amino acid sequence Met Gln Asn Ser Leu which correspond to the N-terminal sequence of the large peptide TcbA, and is disclosed herein as SEQ ID 11O:16. It is thought that TcbA is the precursor protein for TcbA_ii Relationship of tcbA , tcaB and tcaC genes.

The tcaB and ccaC genes are closely linked and may be

transcribed as a single mRNA (Example 8). The tcbA gene is borne on cosmids that apparently do not overlap the ones harboring the t caB and t caC cluster, since the respective genomic library

screens identified different cosmids. However, comparison of the amino sequences encoded by the tcaB and tcaC genes with the tcbA gene reveals a substantial degree of homology. The amino acid conservation (Protein Alignment Mode of MacVector™ Sequence

Analysis Software, scoring matrix pam250, hash value = 2; Kodak

Scientific Imaging Systems, Rochester, NY) is shown in Fig. 4.

On the score line of each panel in Fig. 4, up carats (^) indicate homology or conservative amino acid changes, and down carats (v) indicate nonhomology.

This analysis shows that the amino acid sequence of the TcbA peptide from residues 1739 to 1894 is highly homologous to amino acids 441 to 603 of the TcaB_i peptide (162 of the total 627 amino acids of P8; SEQ ID NO:28). In addition, the sequence of TcbA amino acids 1932 to 2459 is highly homologous to amino acids 12 to 531 of peptide TcaB_ii (520 of the total 562 amino acids; SEQ

ID NO:30). Considering that the TcbA peptide (SEQ ID NO:12)

comprises 2505 amino acids, a total of 684 amino acids (27%) at the C-proximal end of it is homologous to the TcaB_i or TcaB_ii

peptides, and the homologies are arranged colinear to the

arrangement of the putative TcaB preprotein (SEQ ID NO:26). A sizeable gap in the TcbA homology coincides with the junction

between the TcaB_i and TcaB_ii portions of the TcaB preprotein.

Clearly the TcbA and TcaB gene products are evolutionarily

related, and it is proposed that they share some common

function(s) in Photorhabdus . Example 10

Characterization of zinc-metalloproteases in Photorhabdus Broth:

Protease Inhibition, Classification, and Purification Protease Inhibition and Classification Assays: Protease assays were performed using FITC-casein dissolved in water as substrate (0.08% final assay concentration). Proteolysis reactions were performed at 25°C for 1 h in the appropriate buffer with 25 μl of Photorhabdus broth (150 μl total reaction volume). Samples were also assayed in the presence and absence of dithiothreitol. After incubation, an equal volume of 12% trichloroacetic acid was added to precipitate undigested protein. Following precipitation for 0.5 h and subsequent centrifugation, 100 μl of the supernatant was placed into a 96-well microtiter plate and the pH of the solution was adjusted by addition of an equal volume of 4N NaOH. Proteolysis was then quantitated using a Fluoroskan II fluorometric plate reader at excitation and emission wavelengths of 485 and 538 nm, respectively. Protease activity was tested over a range from pH 5.0-10.0 in 0.5 units increments. The following buffers were used at 50 mM final concentration: sodium acetate (pH 5.0 - 6.5); Tris-HCL (pH 7.0 -8.0); and bis-Tris propane (pH 8.5-10.0). To identify the class of protease(s) observed, crude broth was treated with a variety of protease inhibitors (0.5 μg/μl final concentration) and then examined for protease activity at pH 8.0 using the substrate described above. The protease inhibitors used included E-64 (L-trans-expoxysaccinylleucylamido[4-, -guaindino]-butane), 3,4 dichloroisocoumarin, Leupeptin, pepstatin, amastatin,

ethylenediaminetetraacetic acid (EDTA) and 1,10 phenanthroline.

Protease assays performed over a pH range revealed that indeed protease(s) were present which exhibited maximal activity at - pH 8.0 (Table 16). Addition of DTT did not have any effect on protease activity. Crude broth was then treated with a variety of protease inhibitors (Table 17). Treatment of crude broth with the inhibitors described above revealed that 1,10 phenanthroline caused complete inhibition of all protease activity when added at a final concentration of 50 μg, with the IC₅₀ = 5 μg in 100 μl of a 2 mg/ml crude broth solution. These data indicate that the most abundant protease(s) found in the Phot orhabdus broth are from the zinc-metalloprotease class of enzymes.

. .

The isolation of a zinc-metalloprotease was performed by applying dialyzed 10-80% ammonium sulfate pellet to a Q Sepharose column equilibrated at 50 mM Na₂PO₄, pH 7.0 as described in Example 5 for Photorhabdus toxin. After extensive washing, a 0 to 0.5 M NaCl gradient was used to elute toxin protein. The majority of biological activity and protein was eluted from 0.15 - 0.45 M NaCl. However, it was observed that the majority of proteolytic activity was present in the 0.25-0.35 M NaCl fraction with some activity in the 0.15-0.25 M NaCl fraction. SDS PAGE analysis of the 0.25-0.35 M NaCl fraction showed a major peptide band of approximately 60 kDa. The 0.15-0.25 M NaCl fraction contained a similar 60 kDa band but at lower relative protein concentration. Subsequent gel filtration of this fraction using a Superose 12 HR 16/50 column resulted in a major peak migrating at 57.5 kDa that contained a predominant (> 90% of total stained protein) 58.5 kDa band by SDS PAGE analysis. Additional analysis of this fraction using various protease inhibitors as described above determined that the protease was a zinc-metalloprotease. Nearly all of the protease activity present in Photorhabdus broth at day 1 of fermentation corresponded to the ~58 kDa zinc- metalloprotease.

In yet a second isolation of zinc-metalloprotease(s), W-14 Photorhabdus broth grown for three days was taken and protease activity was visualized using sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) laced with gelatin as described in Schmidt, T.M., Bleakley, B. and Nealson, K.M. 1988. SDS running gels (5.5 × 8 cm) were made with 12.5 % polyacrylamide (40% stock solution of acrylamide/bis-acrylamide; Sigma Chemical Co., St. Louis, MO) into which 0.1% gelatin final concentration (Biorad EIA grade reagent; Richmond CA) was incorporated upon dissolving in water. SDS-stacking gels (1.0 × 8 cm) were made with 5% polyacrylamide, also laced with 0.1% gelatin. Typically, 2.5 μg of protein to be tested was diluted in 0.03 ml of SDS-PAGE loading buffer without dithiothreitol (DTT) and loaded onto the gel. Proteins were electrophoresed in SDS running buffer (Laemmli, U.K. 1970. Nature 227, 680) at 0° C and at 8 mA. After electrophoresis was complete, the gel was washed for 2 h in 2.5% (v/v) Triton X-100. Gels were then incubated for 1 h at 37 °C in 0.1 M glycine (pH 8.0). After incubation, gels were fixed and stained overnight with 0.1% amido black in methanol-acetic acid water (30:10:60, vol./vol./vol.; Sigma Chemical Co.). Protease activity was visualized as light areas against a dark, amido black stained background due to proteolysis and subsequent diffusion of incorporated gelatin. At least three distinct bands produced by proteolytic activity at 58-, 41-, and 38 kDa were observed.

Activity assays of the different proteases in W-14 day three culture broth were performed using FITC-casein dissolved in water as substrate (0.02% final assay concentration). Proteolysis experiments were performed at 37 °C for 0-0.5 h in 0.1M Tris-HCl (pH 8.0) with different protein fractions in a total volume of 0.15 ml. Reactions were terminated by addition of an equal volume of 12% trichloroacetic acid (TCA) dissolved in water.

After incubation at room temperature for 0.25 h, samples were centrifuged at 10,000 x g for 0.25 h and 0.10 ml aliquots were removed and placed into 96-well microtiter plates. The solution was then neutralized by the addition of an equal volume of 2 II sodium hydroxide, followed by quantitation using a Fluoroskan II fluorometric plate reader with excitation and emission

wavelengths of 485 and 538 nm, respectively. Activity

measurements were performed using FITC-Casein with different protease concentrations at 37° C for 0-10 min. A unit of activity was arbitrarily defined as the amount of enzyme needed to produce 1000 fluorescent units/min and specific activity was defined as units/mg of protease.

Inhibition studies were performed using two zinc-metalloprotease inhibitors; 1,10 phenanthroline and N-(a-rhamnopyranosyloxyhydroxyphosphinyl)-Leu-Trp (phosphoramidon) with stock solutions of the inhibitors dissolved in 100% ethanol and water, respectively. Stock concentrations were typically 10 mg/ml and 5 mg/ml for 1,10 phenanthroline and phosphoramidon, respectively, with final concentrations of inhibitor at 0.5-1.0 mg/ml per reaction. Treatment of three day W-14 crude broth with 1,10 phenanthroline, an inhibitor of all zinc metalloproteases, resulted in complete elimination of all protease activity while treatment with phosphoramidon, an inhibitor of thermolysin-like proteases (Weaver, L.H., Kester, W.R., and Matthews, B.W. 1977. J. Mol. Biol. 114, 119-132), resulted in ~56% reduction of protease activity. The residual proteolytic activity could not be further reduced with additional phosphoramidon.

The proteases of three day W-14 Photorhabdus broth were purified as follows: 4.0 liters of broth were concentrated using an Amicon spiral ultra filtration cartridge Type S1Y100 attached to an Amicon M-12 filtration device. The flow-through material having native proteins less than 100 kDa in size (3.8 L) was concentrated to 0.375 L using an Amicon spiral ultra filtration cartridge Type S1Y10 attached to an Amicon M-12 filtration device. The retentate material contained proteins ranging in size from 10-100 kDa. This material was loaded onto a Pharmacia HR16/10 column which had been packed with PerSeptive Biosystem (Framington, MA) Poros^® 50 HQ strong anion exchange packing that had been equilibrated in 10 mM sodium phosphate buffer (pH 7.0). Proteins were loaded on the column at a flow rate of 5 ml/min, followed by washing unbound protein with buffer until A₂₈₀ =

0.00. Afterwards, proteins were eluted using a NaCl gradient of 0-1.0 M NaCl in 40 min at a flow rate of 7.5 ml/min. Fractions were assayed for protease activity, supra., and active fractions were pooled. Proteolytically active fractions were diluted with 50% (v/v) 10 mM sodium phosphate buffer (pH 7.0) and loaded onto a Pharmacia HR 10/10 Mono Q column equilibrated in 10 mM sodium phosphate. After washing the column with buffer until A₂₈₀ = 0.00, proteins were eluted using a NaCl gradient of 0-0.5 M NaCl for 1 h at a flow rate of 2.0 ml/min. Fractions were assayed for protease activity. Those fractions having the greatest amount of phosphoramidon-sensitive protease activity, the phosphoramidon sensitive activity being due to the 41/38 kDa protease, infra . , were pooled. These fractions were found to elute at a range of 0.15-0.25 M NaCl. Fractions containing a predominance of phosphoramidon-insensitive protease activity, the 58 kDa

protease, were also pooled. These fractions were found to elute at a range of 0.25-0.35 M NaCl. The phosphoramidon-sensitive protease fractions were then concentrated to a final volume of 0.75 ml using a Millipore Ultrafree^®-15 centrifugal filter device Biomax-5K NMWL membrane. This material was applied at a flow rate of 0.5 ml/min to a Pharmacia HR 10/30 column that had been packed with Pharmacia Sephadex G-50 equilibrated in 10 mM sodium phosphate buffer (pH 7.0)/ 0.1 M NaCl. Fractions having the maximal phosphoramidon-sensitive protease activity were then pooled and centrifuged over a Millipore Ultrafree®-15 centrifugal filter device Biomax-50K NMWL membrane. Proteolytic activity analysis, supra . , indicated this material to have only

phosphoramidon-sensitive protease activity. Pooling of the phosphoramidon-insensitive protease, the 58 kDa protein, was followed by concentrating in a Millipore Ultrafree®-15

centrifugal filter device Biomax-50K NMWL membrane and further separation on a Pharmacia Superdex-75 column. Fractions

containing the protease were pooled.

Analysis of purified 58- and 41/38 kDa purified proteases revealed that, while both types of protease were completely inhibited with 1,10 phenanthroline, only the 41/38 kDa protease was inhibited with phosphoramidon. Further analysis of crude broth indicated that protease activity of day 1 W-14 broth has 23% of the total protease activity due to the 41/38 kDa protease, increasing to 44% in day three W-14 broth.

Standard SDS-PAGE analysis for examining protein purity and obtaining amino terminal sequence was performed using 4-20% gradient MiniPlus SepraGels purchased from Integrated Separation Systems (Natick, MA). Proteins to be amino-terminal sequenced were blotted onto PVDF membrane following purification, infra . , (ProBlott™ Membranes; Applied Biosystems, Foster City, CA), visualized with 0.1% amido black, excised, and sent to Cambridge Prochem; Cambridge, MA, for sequencing.

Deduced amino terminal sequence of the 58- (SEQ ID NO: 45) and 41/38 kDa (SEQ ID NO: 44) proteases from three day old W-14 broth were DV-GSEKANEKLK (SEQ ID NO: 45) and DSGDDDKVTNTDIHR (SEQ ID NO:44), respectively.

Sequencing of the 41/38 kDa protease revealed several amino termini, each one having an additional amino acid removed by proteolysis. Examination of the primary, secondary, tertiary and quartenary sequences for the 38 and 41 kDa polypeptides allowed for deduction of the sequence shown above and revealed that these two proteases are homologous.

Example 11, Part A

Screening of Photorhabdus Genomic Library via use of Antibodies for Genes encoding TcbA Peptide

In parallel to the sequencing described above, suitable probing and sequencing was done based on the TcbA_ii peptide (SEQ ID NO:1). This sequencing was performed by preparing bacterial culture broths and purifying the toxin as described in Examples 1 and 2 above.

Genomic DNA was isolated from the Photorhabdus l uminescens strain W-14 grown in Grace's insect tissue culture medium. The bacteria were grown in 5 ml of culture medium in a 250 ml

Erlenmeyer flask at 28°C and 250 rpm for approximately 24 hours. Bacterial cells from 100 ml of culture medium were pelleted at 5000 x g for 10 minutes. The supernatant was discarded, and the cell pellets then were used for the genomic DNA isolation

The genomic DNA was isolated using a modification of the

CTAB method described in Section 2.4.3 of Ausubel (supra.) The section entitled "Large Scale CsCl prep of bacterial genomic DNA" was followed through step 6. At this point, an additional chloroform/isoamyl alcohol (24:1) extraction was performed followed by a phenol/chloroform/isoamyl (25:24:1) extraction step and a final chloroform/isoamyl/alcohol (24:1) extraction. The DNA was precipitated by the addition of a 0.6 volume of

isopropanol. The precipitated DNA was hooked and wound around the end of a bent glass rod, dipped briefly into 70% ethanol as a final wash, and dissolved in 3 ml of TE buffer. The DNA concentration, estimated by optical density at 230/260 nm, was approximately 2 mg/ml.

Using this genomic DNA, a library was prepared

Approximately 50 μg of genomic DNA was partly digested with Sau3 Al. Then NaCl density gradient centrifugation was used to size fractionate the partially digested DNA fragments. Fractions containing DNA fragments with an average size of 12 kb, or larger, as determined by agarose gel electrophoresis, were ligated into the plasmid BluScript, Stratagene, La Jolla,

California, and transformed into an E. coli DH5α or DHB10 strain.

Separately, purified aliquots of the protein were sent to the biotechnology hybridoma center at the University of

Wisconsin, Madison for production of monoclonal antibodies to the proteins. The material that was sent was the HPLC purified fraction containing native bands 1 and 2 which had been denatured at 65°C, and 20 μg of which was injected into each of four mice. Stable monoclonal antibody-producing hybridoma cell lines were recovered after spleen cells from unimmunized mouse were fused with a stable myeloma cell line. Monoclonal antibodies were recovered from the hybridomas.

Separately, polyclonal antibodies were created by taking native agarose gel purified band 1 (see Example 1) protein which was then used to immunize a New Zealand white rabbit. The protein was prepared by excising the band from the native agarose gels, briefly heating the gel pieces to 65°C to melt the agarose, and immediately emulsifying with adjuvant. Freund's complete adjuvant was used for the primary immunizations and Freund's incomplete was used for 3 additional injections at monthly intervals. For each injection, approximately 0.2 ml of

emulsified band 1, containing 50 to 100 micrograms of protein, was delivered by multiple subcontaneous injections into the back of the rabbit. Serum was obtained 10 days after the final injection and additional bleeds were performed at weekly

intervals for 3 weeks. The serum complement was inactivated by heating to 56°C for 15 minutes and then stored at -20°C.

The monoclonal and polyclonal antibodies were then used to screen the genomic library for the expression of antigens which could be detected by the epitope. Positive clones were detected on nitrocellulose filter colony lifts. An immunoblot analysis of the positive clones was undertaken. An analysis of the clones as defined by both immunoblot and Southern analysis resulted in the tentative identification of five classes of clones.

In the first class of clone was a gene encoding the peptide designated here as TcbA_ii. Full DNA sequence of this gene ( TcbA) was obtained. It is set forth as SEQ ID NO: 11. Confirmation that the sequence encodes the internal sequence of SEQ ID NO:1 is demonstrated by the presence of SEQ ID NO:1 at amino acid number 38 from the deduced amino acid sequence created by the open reading frame of SEQ ID NO: 11. This can be confirmed by

referring to SEQ ID NO: 12, which is the deduced amino acid sequence created by SEQ ID NO: 11.

The second class of toxin peptides contains the segments referred to above as TcaB_i, TcaB_ii and TcaC. Following the screening of the library with the polyclonal antisera, this second class of toxin genes was identified by several clones which produced different size proteins, all of which cross-reacted with the polyclonal antibody on an immunoblot and were also found to share DNA homology on a Southern Blot. Sequence comparison revealed that they belonged to the gene complex designated TcaB and TcaC above.

Three other classes of antibody toxin clones were also isolated in the polyclonal screen. These classes produced proteins that cross-react with a polyclonal antibody and also shared DNA homology with the classes as determined by Southern blotting. The classes have been designated Class III, Class IV and Class V. It was also possible to identify monoclonals that cross-reacted with Class I, II, III, and IV. This suggests that all have regions of high protein homology. Thus, it appears that the P. luminescens extracellular protein genes represent a family of genes which are evolutionarily related.

To further pursue the concept that there might be

evolutionarily related variations in the toxin peptides contained within this organism, two approaches have been undertaken to examine other strains of P. l uminescens tor the presence of related proteins. This was done both by PCR amplification of genomic DNA and by immunoblot analysis using the polyclonal and monoclonal antibodies. The results indicate that related proteins are produced by P . l uminescens strains WX-2, WX-3, WX-4, WX-5, WX-6, WX-7, WX-8, WX-11, WX-12, WX-15 and W-14. Example 11, Part B

Sequence and anaylsis of Class III toxin clones - t cc

Further DNA sequencing was performed on plasmids isolated from Class III E. col i clones described in Example 11, Part A. The nucleotide sequence was shown to be three closely linked open reading frames at this genomic locus. This locus was designated tcc with the three open reading frames designated cccA SEQ ID NO:56, tccB SEQ ID NO:58 and tccC SEQ ID NO:60 (Fig. 6B).

The deduced amino acid from the tccA open reading frame indicates the gene encodes a protein of 105,459 Da. This protein was designated TccA. The first 12 amino acids of this protein match the N-terminal sequence obtained from a 108 kDa protein, SEQ ID NO: 7, previously identified as part of the toxin complex.

The deduced amino acid from the tccB open reading frame indicates this gene encodes a protein of 175,716 Da. This protein was designated TccB. The first 11 amino acids of this protein match the N-terminal sequence obtained from a protein with estimated molecular weight of 185 kDa, SEQ ID NO: 8.

The deduced amino acid sequence of tccC indicated that this open reading frame encodes a protein of 111,694 Da and the protein product was designated TccC.

Example 12

Characterization of Photorhabdus Strains

In order to establish that the collection described herein was comprised of Photorhabdus strains, the strains herein were assessed in terms of recognized microbiological traits that are characteristic of Photorhabdus and which differentiate it from other Enterobacteriaceae and Xenorhabdus spp. (Farmer, J.J. 1984. Bergey's Manual of Systemic Bacteriology, vol 1. pp. 510-511. (ed. Kreig N.R. and Holt, J.G.). Williams & Wilkins, Baltimore.; Akhurst and Boemare, 1988, Boemare et al., 1993). These

characteristic traits are as follows: Gram's stain negative rods, organism size of 0.5-2 μm in width and 2-10 μm in length red/yellow colony pigmentation, presence of crystalline inclusion bodies, presence of catalase, inability to reduce nitrate, presence of bioluminescence, ability to take up dye from growth media, positive for protease production, growth-temperature range below 37°C, survival under anaerobic conditions and positively motile. (Table 18). Reference Escherichia col i , Xenorhabdus and Photorhabdus strains were included in all tests for comparison. The overall results are consistent with all strains being part ot the family Enterobacteriaceae and the genus Photorhabdus .

A luminometer was used to establish the bioluminescence of each strain and provide a quantitative and relative measurement of light production. For measurement of relative light emitting units, the broths from each strain (cells and media) were measured at three time intervals after inoculation in liquid culture (6, 12, and 24 hr) and compared to background luminosity (uninoculated media and water). Prior to measuring light emission from the various broths, cell density was established by measuring light absorbance (560 nM) in a Gilford Systems

(Oberlin, OH) spectrophotometer using a sipper cell. Appropriate dilutions were then made (to normalize optical density to 1.0 unit) before measuring luminosity. Aliquots of the diluted broths were then placed into cuvettes (300 μl each) and read in a Bio-Orbit 1251 Luminometer (Bio-Orbit Oy, Twiku, Finland). The integration period for each sample was 45 seconds. The samples were continuously mixed (spun in baffled cuvettes) while being read to provide oxygen availability. A positive test was determined as being ≥ 5- fold background luminescence (-5-10 units). In addition, colony luminosity was detected with photographic film overlays and visually, after adaptation in a darkroom. The Gram's staining characteristics of each strain were established with a commercial Gram's stain kit (BBL,

Cockeysville, MD) used in conjunction with Gram's stain control slides (Fisher Scientific, Pittsburgh, PA). Microscopic evaluation was then performed using a Zeiss microscope (Carl Zeiss, Germany) 100X oil immersion objective lens (with 10X ocular and 2X body magnification). Microscopic examination of individual strains for organism size, cellular description and inclusion bodies (the latter after logarithmic growth) was performed using wet mount slides (10X ocular, 2X body and 40X objective magnification) with oil immersion and phase contrast microscopy with a micrometer (Akhurst, R.J. and Boemare, N.E. 1990. Entomopathogenic Nematodes in Biological Control (ed.

Gaugler, R. and Kaya, H.). pp. 75-90. CRC Press, Boca Raton,

USA.; Baghdiguian S., Boyer-Giglio M.H., Thaler, J.O., Bonnot G., Boemare II. 1993. Biol. Cell 79, 177-185.). Colony pigmentation was observed after inoculation on Bacto nutrient agar, (Difco Laboratories, Detroit, MI) prepared as per label instructions. Incubation occurred at 28°C and descriptions were produced after 5-7 days. To test for the presence of the enzyme catalase, a colony of the test organism was removed on a small plug from a nutrient agar plate and placed into the bottom of a glass test tube. One ml of a household hydrogen peroxide solution was gently added down the side of the tube. A positive reaction was recorded when bubbles of gas (presumptive oxygen) appeared immediately or within 5 seconds. Controls of uninoculated nutrient agar and hydrogen peroxide solution were also examined. To test for nitrate reduction, each culture was inoculated into 10 ml of Bacto Nitrate Broth (Difco Laboratories, Detroit, MI). After 24 hours incubation at 28°C, nitrite production was tested by the addition of two drops of sulfanilic acid reagent and two drops of alpha-naphthylamine reagent (see Difco Manual, 10th edition, Difco Laboratories, Detroit, MI, 1984). The generation of a distinct pink or red color indicates the formation of nitrite from nitrate. The ability of each strain to uptake dye from growth media was tested with Bacto MacConkey agar containing the dye neutral red; Bacto Tergitol-7 agar containing the dye bromothymol blue and Bacto EMB Agar containing the dye eosin-Y (agars from Difco Laboratories, Detroit, MI, all prepared according to label instructions). After inoculation on these media, dye uptake was recorded after incubation at 28°C for 5 days. Growth on these latter media is characteristic for members of the family Enterobacteriaceae. Motility of each strain was tested using a solution of Bacto Motility Test Medium (Difco

Laboratories, Detroit, MI) prepared as per label instructions. A butt-stab inoculation was performed with each strain and motility was judged macroscopically by a diffuse zone of growth spreading from the line of inoculum. In many cases, motility was also reserved microscopically from liquid culture under wet mount slides. Biochemical nutrient evaluation for each strain was performed using BBL Enterotube II (Benton, Dickinson, Germany). Product instructions were followed with the exception that incubation was carried out at 28°C for 5 days. Results were consistent with previously cited reports for Photorhabdus . The production of protease was tested by observing hydrolysis of gelatin using Bacto gelatin (Difco Laboratories, Detroit, MI) plates made as per label instructions. Cultures were inoculated and the plates were incubated at 28°C for 5 days. To assess growth at different temperatures, agar plates [2% proteose peptone #3 with two percent Bacto-Agar (Difco, Detroit, MI) in deionized water] were streaked from a common source of inoculum. Plates were sealed with Nesco® film and incubated at 20, 28 and 37°C for up to three weeks. Plates showing no growth at 37°C snowed no cell viability after transfer to a 28°C incubator for one week. Oxygen requirements for Photorhabdus strains were tested in the following manner. A butt-stab inoculation into fluid thioglycolate broth medium (Difco, Detroit, MI) was made. The tubes were incubated at room temperature for one week and cultures were then examined for type and extent of growth. The indicator resazurin demonstrates the level of medium oxidation or the aerobiosis zone (Difco Manual, 10th edition, Difco

Laboratories, Detroit, MI). Growth zone results obtained for the Photorhabdus strains tested were consistent with those of a facultative anaerobic microorganism.

,

Cellular fatty acid analysis is a recognized tool for bacterial characterization at the genus and species level

(Tornabene, T.G. 1985. Lipid Analysis and the Relationship to Chemotaxonomy in Methods in Microbiology, Vol 18, 209-224.;

Goodfellow, M. and O'Donnell, A.G 1993 Foots of Bacterial

Systematics in Handbook of New Bacterial Systematics (ed.

Goodfellow, M. & O'Donnell, A.G.) pp. 3-54. London: Academic Press Ltd.), these references are incorporated herein by

reference, and were used to confirm that our collection was related at the genus level. Cultures were shipped to an

external, contract laboratory for fatty acid methyl ester analysis (FAME) using a Microbial ID (MIDI, Newark, DE, USA) Microbial Identification System (MIS). The MIS system consists of a Hewlett Packard HP5890A gas chromatograph with a 25mm × 0.2mm 5% methylphenyl silicone fused silica capillary column. Hydrogen is used as the carrier gas and a flame-ionization detector functions in conjunction with an automatic sampler, integrator and computer. The computer compares the sample fatty acid methyl esters to a microbial fatty acid library and against a

calibration mix of known fatty acids. As selected by the contract laboratory, strains were grown for 24 hours at 28ºC on trypticase soy agar prior to analysis. Extraction of samples was performed by the contract lab as per standard FAME methodology. There was no direct identification of the strains to any

luminescent bacterial group other than Photorhabdus . When the cluster analysis was performed, which compares the fatty acid profiles of a group of isolates, the strain fatty acid profiles were related at the genus level.

The evolutionary diversity of the Photorhabdus strains in our collection was measured by analysis of PCR (Polymerase Chain Reaction) mediated genomic fingerprinting using genomic DNA from each strain. This technique is based on families of repetitive DNA sequences present throughout the genome of diverse bacterial species (reviewed by Versalovic, J., Schneider, M., DE Bruijn, F.J. and Lupski, J.R. 1994. Methods Mol. Cell. Biol., 5, 25-40.) . Three of these, repetitive extragenic palindromic sequence (REP), enterobacterial repetitive intergenic consensus (ERIC) and the BOX element are thought to play an important role in the

organization of the bacterial genome. Genomic organization is believed to be shaped by selection and the differential

dispersion of these elements within the genome of closely related bacterial strains can be used to discriminate these strains (e.g Louws, F.J., Fulbright, D.W., Stephens, C.T. and DE Bruijn, F.J. 1994. Appl. Environ. Micro. 60, 2286-2295.). Rep-PCR utilizes oligonucleotide primers complementary to these repetitive sequences to amplify the variably sized DNA fragments lying between them. The resulting products are separated by

electrophoresis to establish the DNA "fingerprint" for each strain.

To isolate genomic DNA from our strains, cell pellets were resuspended in TE buffer (10 mM Tris-HCl, 1 mM EDTA, pH 8.0) to a final volume of 10 ml and 12 ml of 5 M NaCl was then added. This mixture was centrifuged 20 min. at 15,000 x g. The resulting pellet was resuspended in 5.7 ml of TE and 300 μl of 10% SDS and 60 μl 20 mg/ml proteinase K (Gibco BRL Products, Grand Island, NY) were added. This mixture was incubated at 37 °C for 1 hr, approximately 10 mg of lysozyme was then added and the mixture was incubated for an additional 45 min. One milliliter of 5M NaCl and 800 μl of CTAB/NaCl solution (10% w/v CTAB, 0.7 M NaCl) were then added and the mixture was incubated 10 min. at 65°C, gently agitated, then incubated and agitated for an additional 20 min. to aid in clearing of the cellular material. An equal volume of chloroform/isoamyl alcohol solution (24:1, v/v) was added, mixed gently then centrifuged. Two extractions were then performed with an equal volume of phenol/chloroform/isoamyl alcohol (50:49:1). Genomic DNA was precipitated with 0.6 volume of isopropanol.

Precipitated DNA was removed with a glass rod, washed twice with 70% ethanol, dried and dissolved in 2 ml of STE (10 mM Tris-HCl pH8.0, 10 mM NaCl, 1 mM EDTA). The DNA was then quantitated by optical density at 260 nm. To perform rep-PCR analysis of

Photorhabdus genomic DNA the following primers were used, REP1R-I; 5'-IIIICGICGICATCIGGC-3' and REP2-I; 5'-ICGICTTATCIGGCCTAC-3'. PCR was performed using the following 25μl reaction: 7.75 μl H₂O,

2.5 μl 10X LA buffer (PanVera Corp., Madison, WI), 16 ul dNTP mix (2.5 mM each), 1 μl of each primer at 50 pM/μl, 1 μl DMSO, 1.5 μl genomic DNA (concentrations ranged from 0.075-0.480 μg/μl) and 0.25 μl TaKaRa EX Taq (PanVera Corp., Madison, WI). The PCR amplification was performed in a Perkin Elmer DNA Thermal Cycler (Norwalk, CT) using the following conditions: 95°C/7 min. then 35 cycles of; 94°C/1 min.,44°C/1 min., 65°C/8 min., followed by 15 min. at 65°C. After cycling, the 25 μl reaction was added to 5 μl of 6X gel loading buffer (0.25% bromophenol blue, 40% w/v sucrose in H₂O). A 15×20cm 1%-agarose gel was then run in TBE buffer

(0.09 M Tris-borate, 0.002 M EDTA) using 8 μl of each reaction. The gel was run for approximately 16 hours at 45v. Gels were then stained in 20 μg/ml ethidium bromide for 1 hour and destained in TBE buffer for approximately 3 hours. Polaroid® photographs ot the gels were then taken under UV illumination.

The presence or absence of bands at specific sizes for each strain was scored from the photographs and entered as a

similarity matrix in the numerical taxonomy software program, NTSYS-pc (Exeter Software, Setauket, NY). Controls of E. coli strain HB101 and Xanthomonas oryzae pv. oryzae assayed at the same time produced PCR "fingerprints" corresponding to published reports (Versalovic, J., Koeuth, T. and Lupski, J.R. 1991.

Nucleic Acids Res. 19, 6823-6831; Vera Cruz, CM., Halda-Alija, L . , Louws, F., Skinner, D.Z., George, M.L., Nelson, R.J., DE Bruijn, F.J., Rice, C. and Leach, J.E. 1995. Int. Rice Res.

Notes, 20, 23-24.; Vera Cruz, CM., Ardales, E.Y., Skinner, D.Z., Talag, J., Nelson, R. J., Louws, F.J., Leung, H., Mew, T.W. and Leach, J.E. 1996. Phytopathology (in press, respectively). The data from Photorhabdus strains were then analyzed with a series of programs within NTSYS-pc; SIMQUAL (Similarity for Qualitative data) to generate a matrix of similarity coefficients (using the Jaccard coefficient) and SAHN (Sequential, Agglomerative,

Heirarchical and Nested) clustering [using the UPGMA (Unweighted Pair-Group Method with Arithmetic Averages) method) which groups related strains and can be expressed as a phenogram (Figure 5). The COPH (cophenetic values) and MXCOMP (matrix comparison) programs were used to generate a cophenetic value matrix and compare the correlation between this and the original matrix upon which the clustering was based. A resulting normalized Mantel statistic (r) was generated which is a measure of the goodness of fit for a cluster analysis (r=0.8-0.9 represents a very good fit). In our case r = 0.919. Therefore, our collection is comprised of a diverse group of easily distinguishable strains representative of the Photorhabdus genus. Example 13

Insecticidal Utility of Toxin is) Produced by Various Photorhabdus Strains Initial "seed" cultures of the various Photorhabdus strains were produced by inoculating 175 ml of 2% Proteose Peptone #3 (PP3) (Difco Laboratories, Detroit, MI) liquid media with a primary variant subclone in a 500 ml tribaffled flask with a Delong neck, covered with a Kaput. Inoculum for each seed culture was derived from oil-overlay agar slant cultures or plate cultures. After inoculation, these flasks were incubated for 16 hrs at 28°C on a rotary shaker at 150 rpm. These seed cultures were then used as uniform inoculum sources for a given

fermentation of each strain. Additionally, overlaying the post-log seed culture with sterile mineral oil, adding a sterile magnetic stir bar for future resuspension and storing the culture in the dark, at room temperature provided long-term preservation of inoculum in a toxin-competent state. The production broths were inoculated by adding 1% of the actively growing seed culture to fresh 2% PP3 media (e.g. 1.75 ml per 175 ml fresh media).

Production of broths occurred in either 500 ml tribaffled flasks (see above), or 2800 ml baffled, convex bottom flasks (500 ml volume) covered by a silicon foam closure. Production flasks were incubated for 24-48 hrs under the above mentioned

conditions. Following incubation, the broths were dispensed into sterile 1 L polyethylene bottles, spun at 2600 x g for 1 hr at 10°C and decanted from the cell and debris pellet. The liquid broth was then vacuum filtered through Whatman GF/D (2.7 μM retention) and GF/B (1.0 uM retention) glass filters to remove debris. Further broth clarification was achieved with a tangential flow microfiltration device (Pall Filtron,

Northborough, MA) using a 0.5 μM open-channel filter. When necessary, additional clarification could be obtained by chilling the broth (to 4°C) and centrifuging for several hours at 2600 x g. Following these procedures, the broth was filter sterilized using a 0.2 μM nitrocellulose membrane filter. Sterile broths were then used directly for biological assay, biochemical analysis or concentrated (up to 15-fold) using a 10,000 Mw cut-off, M12 ultra-filtration device (Amicon, Beverly MA) or centrifugal concentrators (Millipore, Bedford, MA and Pall

Filtron, Northborough, MA) with a 10,000 MW pore size. In the case of centrifugal concentrators, the broth was spun at 2000 x g for approximately 2 hr. The 10,000 MW permeate was added to the corresponding retentate to achieve the desired concentration of components greater than 10,000 MW. Heat inactivation of

processed broth samples was acheived by heating the samples at 100°C in a sand-filled heat block for 10 minutes.

The broth(s) and toxin complex(es) from different

Photorhabdus strains are useful for reducing populations of insects and were used in a method of inhibiting an insect population which comprises applying to a locus of the insect an effective insect inactivating amount of the active described. A demonstration of the breadth of insecticidal activity observed from broths of a selected group of Photorhabdus strains fermented as described above is shown in Table 19. It is possible that additional insecticidal activities could be detected with these strains through increased concentration of the broth or by employing different fermentation methods. Consistent with the activity being associated with a protein, the insecticidal activity of all strains tested was heat labile (see above).

Culture broth(s) from diverse Photorhabdus strains show differential insecticidal activity (mortality and/or growth inhibition, reduced adult emergence) against a number of insects. More specifically, the activity is seen against corn rootworm larvae and boll weevil larvae which are members of the insect order Coleoptera . Other members of the Coleoptera include wireworms, pollen beetles, flea beetles, seed beetles and

Colorado potato beetle. Activity is also observed against aster leafhopper and corn plant hopper, which are members of the order Homoptera . Other members of the Homoptera include planthoppers, pear psylla, apple sucker, scale insects, whiteflies, spittle bugs as well as numerous host specific aphid species. The broths and purified toxin complex (es) are also active against tobacco budworm, tobacco hornworm and European corn borer which are members of the order Lepidoptera . Other typical members of this order are beet armyworm, cabbage looper, black cutworm, corn earworm, codling moth, clothes moth, Indian mealmoth, leaf rollers, cabbage worm, cotton bollworm, bagworm, Eastern tent caterpillar, soα webworm and fall armyworm. Activity is also seen against fruitfly and mosquito larvae which are members of the order Diptera . Other members of the order Diptera are, pea midge, carrot fly, cabbage root fly, turnip root fly, onion fly, crane fly and house fly and various mosquito species. Activity with broth(s) and toxin complex(es) is also seen against two-spotted spider mite which is a member of the order Acarina which includes strawberry spider mites, broad mites, citrus red mite, European red mite, pear rust mite and tomato russet mite.

Activity against corn rootworm larvae was tested as follows Photorhabdus culture broth(s) (0-15 fold concentrated, filter sterilized), 2% Proteose Peptone #3, purified toxin complex(es) [0.23 mg/ml] or 10 mM sodium phosphate buffer , pH 7.0 were applied directly to the surface (about 1.5 cm²) of artificial diet (Rose, R. I. and McCabe, J. M. (1973). J. Econ. Entomol. 66, (398-400) in 40 μl aliquots. Toxin complex was diluted in 10 mM sodium phosphate buffer, pH 7.0. The diet plates were allowed to air-dry in a sterile flow-hood and the wells were infested with single, neonate Diabrotica undecimpunctata howardi (Southern corn rootworm, SCR) hatched from surface sterilized eggs. The plates were sealed, placed in a humidified growth chamber and maintained at 27°C for the appropriate period (3-5 days). Mortality and larval weight determinations were then scored. Generally, 16 insects per treatment were used in all studies. Control

mortality was generally less than 5%.

Activity against boll weevil (Anthomonas grandis ) was tested as follows. Concentrated (1-10 fold) Photorhabdus broths, control medium (2% Proteose Peptone #3), purified toxin

complex(es) [0.23 mg/ml] or 10 mM sodium phosphate buffer, pH 7 0 were applied in 60 μl aliquots to the surface of 0.35 g of artificial diet (Stoneville Yellow lepidopteran diet) and allowed to dry. A single, 12-24 hr boll weevil larva was placed on the diet, and the wells were sealed and held at 25°C, 50% RH for 5 days. Mortality and larval weights were then assessed. Control mortality ranged between 0-13%.

Activity against mosquito larvae was tested as follows. The assay was conducted in a 96-well microtiter plate. Each well contained 200 μl of aqueous solution (10-fold concentrated

Photorhabdus culture broth(s), control medium (2% Proteose Peptone # 3 ) , 10 mM sodium phosphate buffer, toxin complex(es) @ 0.23 mg/ml or H₂O) and approximately 20, 1-day old larvae (Aedes aegypt i ) . There were 6 wells per treatment The results were read at 3-4 days after infestation. Control mortality was between 0-20%.

Activity against fruitflies was tested as follows.

Purchased Drosophi l a melanogaster medium was prepared using 50% dry medium and a 50% liquid of either water, control medium (2% Proteose Peptone #3), 10-fold concentrated Photorhabdus culture broth(s), purified toxin complex(es) [0.23 mg/ml] or 10 mM sodium phosphate buffer , pH 7.0. This was accomplished by placing 4.0 ml of dry medium in each of 3 rearing vials per treatment and adding 4.0 ml of the appropriate liquid. Ten late instar

Drosophi la melanogaster maggots were then added to each 25 ml vial. The vials were held on a laboratory bench, at room temperature, under fluorescent ceiling lights. Pupal or adult counts were made after 15 days of exposure. Adult emergence as compared to water and control medium (0-16% reduction).

Activity against aster leafhopper adults (Macrosteles severini ) and corn planthopper nymphs (Peregrinus maidis) was tested with an ingestion assay designed to allow ingestion of the active without other external contact. The reservoir for the active/"food" solution is made by making 2 holes in the center of the bottom portion of a 35X10 mm Petri dish. A 2 inch Parafilm M^® square is placed across the top of the dish and secured with an "O" ring. A 1 oz. plastic cup is then infested with

approximately 7 hoppers and the reservoir is placed on top of the cup, Parafilm down. The test solution is then added to the reservoir through the holes. In tests using 10-fold concentrated Photorhabdus culture broth(s), the broth and control medium (2% Proteose Peptone #3) were dialyzed against 10 mM sodium phosphate buffer, pH 7.0 and sucrose (to 5%) was added to the resulting solution to reduce control mortality. Purified toxin complex(es) [0.23 mg/ml] or 10 mM sodium phosphate buffer, pH 7.0 was also tested. Mortality is reported at day 3. The assay was held in an incubator at 28°C, 70% RH with a 16/8 photoperiod. The assays were graded for mortality at 72 hours. Control mortality was less than 6%. Activity against lepidopteran larvae was tested as follows. Concentrated (10-fold) Photorhabdus culture broth(s), control medium (2% Proteose Peptone #3), purified toxin complex(es) [0.23 mg/ml] or 10 mM sodium phosphate buffer, pH 7.0 were applied directly to the surface (~1.5 cm²) of standard artificial lepidopteran diet (Stoneville Yellow diet) in 40 μl aliquots. The diet plates were allowed to air-dry in a sterile flow-hood and each well was infested with a single, neonate larva. European corn borer ( Ostrinia nubilalis) and tobacco hornworm (Man duca sexta ) eggs were obtained from commercial sources and hatched in-house, whereas tobacco budworm (Heliothis virescens ) larvae were supplied internally. Following infestation with larvae, the diet plates were sealed, placed in a humidified growth chamber and maintained in the dark at 27°C for the appropriate period.

Mortality and weight determinations were scored at day 5.

Generally, 16 insects per treatment were used in all studies. Control mortality generally ranged from 4-12.5% for control medium and was less than 10% for phosphate buffer.

Activity against two-spotted spider mite ( Tetranychus urticae) was determined as follows. Young squash plants were trimmed to a single cotyledon and sprayed to run-off with 10- fold concentrated broth(s), control medium (2% Proteose Peptone #3), purified toxin complex(es) [0.23 mg/ml] or 10 mM sodium phosphate buffer, pH 7.0. After drying, the plants were infested with a mixed population of spider mites and held at lab temperature and humidity for 72 hr. Live mites were then counted to determine levels of control.

Example 14

Non W-14 Photorhabdus Strains:

Purification, Characterization and Activity Spectrum Purification

The protocol, as follows, is similar to that developed for the purification of W-14 and was established based on purifying those fractions having the most activity against Southern corn root worm (SCR), as determined in bioassays (see Example 13). Typically, 4-20 L of broth that had been filtered, as described in Example 13, were received and concentrated using an Amicon spiral ultra filtration cartridge Type S1Y100 attached to an Amicon M-12 filtration device. The retentate contained native proteins consisting of molecular sizes greater than 100 kDa, whereas the flow through material contained native proteins less than 100 kDa in size. The majority of the activity against SCR was contained in the 100 kDa retentate. The retentate was then continually diafiltered with 10 mM sodium phosphate (pH = 7.0) until the filtrate reached an A280 < 0.100. Unless otherwise stated, all procedures from this point were performed in buffer as defined by 10 mM sodium phosphate (pH 7.0). The retentate was then concentrated to a final volume of approximately 0.20 L and filtered using a 0.45 mm Nalgene™ Filterware sterile filtration unit. The filtered material was loaded at 7.5 ml/min onto a Pharmacia HR16/10 column which had been packed with PerSeptive Biosystem Poros® 50 HQ strong anion exchange matrix equilibrated in buffer using a PerSeptive Biosystem Sprint® HPLC system.

After loading, the column was washed with buffer until an A₂₈₀

0.100 was achieved. Proteins were then eluted from the column at 2.5 ml/min using buffer with 0.4 M NaCl for 20 min for a total volume of 50 ml. The column was then washed using buffer with 1.0 M NaCl at the same flow rate for an additional 20 min (final volume = 50 ml). Proteins eluted with 0.4 M and 1.0 M NaCl were placed in separate dialysis bags (Spectra/Por® Membrane MWCO: 2,000) and allowed to dialyze overnight at 4° C in 12 L buffer.

The majority of the activity against SCR was contained in the 0 4 M fraction. The 0.4 M fraction was further purified by

application of 20 ml to a Pharmacia XK 26/100 column that had been prepacked with Sepharose CL4B (Pharmacia) using a flow rate of 0.75 ml/min. Fractions were pooled based en A₂₈₀ peak profile and concentrated to a final volume of 0.75 ml using a Millipore Ultrafree®-15 centrifugal filter device Biomax-50K NMWL membrane. Protein concentrations were determined using a Biorad Protein Assay Kit with bovine gamma globulin as a standard.

Characterization

The native molecular weight of the SCR toxin complex was determined using a Pharmacia HR 16/50 that had been prepacked with Sepharose CL4B in buffer. The column was then calibrated using proteins of known molecular size thereby allowing for calculation of the toxin approximate native molecular size. As shown in Table 20, the molecular size of the toxin complex ranged from 777 kDa with strain Hb to 1,900 kDa with strain WX-14. The yield of toxin complex also varied, from strain WX-12 producing 0.8 mg/L to strain Hb, which produced 7.0 mg/L.

Proteins found in the toxin complex were examined for individual polypeptide size using SDS-PAGE analysis. Typically, 20 mg protein of the toxin complex from each strain was loaded onto a 2-15% polyacrylamide gel (Integrated Separation Systems) and electrophoresed at 20 mA in Biorad SDS-PAGE buffer. After completion of electrophoresis, the gels were stained overnight in Biorad Coomassie blue R-250 (0.2% in methanol: acetic acid:

water; 40:10:40 v/v/v). Subsequently, gels were destained in methanol: acetic acid: water; 40:10:40 (v/v/v). The gels were then rinsed with water for 15 min and scanned using a Molecular Dynamics Personal Laser Densitometer®. Lanes were quantitated and molecular sizes were calculated as compared to Biorad high molecular weight standards, which ranged from 200-45 kDa.

Sizes of the individual polypeptides comprising the SCR toxin complex from each strain are listed in Table 21. The sizes of the individual polypeptides ranged from 230 kDa with strain WX-1 to a size of 16 kDa, as seen with strain WX-7. Every strain, with the exception of strain Hb, had polypeptides comprising the toxin complex that were in the 160-230 kDa range, the 100-160 kDa range, and the 50-80 kDa range. These data indicate that the toxin complex may vary in peptide composition and components from strain to strain, however, in all cases the toxin attributes appears to consist of a large, oligomeric protein complex.

Activity Spectrum

As shown in Table 21, the toxin complexes purified from strains Hm and H9 were tested for activity against a variety of insects, with the toxin complex from strain W-14 for comparison. The assays were performed as described in Example 13. The toxin complex from all three strains exhibited activity against tobacco bud worm, European corn borer, Southern corn root worm, and aster leafhopper. Furthermore, the toxin complex from strains Hm and W-14 also exhibited activity against two-spotted spider mite. In addition, the toxin complex from W-14 exhibited activity against mosquito larvae. These data indicate that the toxin complex, while having similarities in activities between certain orders of insects, can also exhibit differential activities against other orders of insects.

Examp1e 15

Sub-Fractionation of Photorhabdus Protein Toxin Complex

The Photorhabdus protein toxin complex was isolated as described in Example 14. Next, about 10 mg toxin was applied to a MonoQ 5/5 column equilibrated with 20 mM Tris-HCl, pH 7.0 at a flow rate of 1ml /min. The column was washed with 20 mM Tris-HCl, pH 7.0 until the optical density at 280 nm returned to baseline absorbance. The proteins bound to the column were eluted with a linear gradient of 0 to 1.0 M NaCl in 20 mM Tris-HCl, pH 7 0 at 1 ml/min for 30 min. One ml fractions were collected and subjected to Southern corn rootworm (SCR) bioassay (see Example 13). Peaks of activity were determined by a series of dilutions of each fraction in SCR bioassays. Two activity peaks against SCR were observed and were named A (eluted at about 0.2-0.3 M NaCl) and B (eluted at 0.3-0.4 M NaCl). Activity peaks A and B were pooled separately and both peaks were further purified using a 3-step procedure described below.

Solid (NH₄)₂SO₄ was added to the above protein fraction to a final concentration of 1.7 M. Proteins were then applied to a phenyl-Superose 5/5 column equilibrated with 1.7 M (NH₄)₂SO₄ in 50 mM potassium phosphate buffer, pH 7 at 1 ml/min. Proteins bound to the column were eluted with a linear gradient of 1.7 M (NH₄)₂SO₄, 0% ethylene glycol, 50 mM potassium phosphate, pH 7 0 to 25% ethylene glycol, 25 mM potassium phosphate, pH 7.0 (no (NH₄)₂SO₄) at 0.5 ml/min. Fractions were dialyzed overnight against 10 mM sodium phosphate buffer, pH 7.0 Activities in each fraction against SCR were determined by bioassay.

The fractions with the highest activity were pooled and applied to a MonoQ 5/5 column which was equilibrated with 20 mM Tris-HCl, pH 7.0 at 1 ml/min. The proteins bound to the column weie eluted at 1 ml /min by a linear gradient of 0 to 1M NaCl in

20 mM Tris-HCl, pH 7.0.

For the final step of purification, the most active

fractions above (determined by SCR bioassay) were pooled and subjected to a second phenyl-Superose 5/5/ column. Solid

(NH₄)₂SO₄ was added to a final concentration of 1.7 M. The solution was then loaded onto the column equilibrated with 1.7 M

(NH₄)₂SO₄ in 50 mM potassium phosphate buffer, pH 7 at 1ml/min.

Proteins bound to the column were eluted with a linear gradient of 1.7 M (NH₄)₂SO₄, 50 mM potassium phosphate, pH 7.0 to 10 mM potassium phosphate, pH 7.0 at 0.5 ml/min. Fractions were dialyzed overnight against 10 mM sodium phosphate buffer, pH 7 0.

Activities in each fraction against SCR were determined by bioassay.

The final purified protein by the above 3 -step procedure from peak A was named toxin A and the final purified protein from peak B was named toxin B.

Characterization and Amino Acid Sequencing of Toxin A and Toxin B In SDS-PAGE, both toxin A and toxin B contained two major (> 90% of total Commassie stained protein) peptides: 192 kDa (named A1 and B1, respectively) and 58 kDa (named A2 and B2,

respectively). Both toxin A and toxin B revealed only one major band in native PAGE, indicating A1 and A2 were subunits of one protein complex, and B1 and B2 were subunits of one protein complex. Further, the native molecular weight of both toxin A and toxin B were determined to be 860 kDa by gel filtration chromatography. The relative molar concentrations of A1 to A2 was judged to be a 1 to 1 equivalence as determined by

densiometric analysis of SDS-PAGE gels. Similarly, B1 and B2 peptides were present at the same molar concentration.

Toxin A and toxin B were electrophoresed in 10% SDS-PAGE and transblotted to PVDF membranes. Blots were sent for amino acid analysis and N-terminal amino acid sequencing at Harvard

MicroChem and Cambridge ProChem, respectively. The N-terminal amino sequence of B1 was determined to be identical to SEQ ID NO:1, the TcbA_ii region of the tcbA gene (SEQ ID NO: 12, position 37 to 99). A unique N-terminal sequence was obtained for peptide B2 (SEQ ID NO:40). The N-terminal amino acid sequence of peptide B2 was identical to the TcbA_iii region of the derived amino acid sequence for the tcbA gene (SEQ ID NO:12, position 1935 to 1945). Therefore, the B toxin contained predominantly two peptides, TcbA_ii and TcbA_iii, that were observed to be derived from the same gene product, TcbA.

The N-terminal sequence of A2 (SEQ ID NO: 41) was unique in comparison to the TcbA_iii peptide and other peptides. The A2 peptide was denoted TcdA_iii (see Example 17). SEQ ID NO:6 was determined to be a mixture of amino acid sequences SEQ ID NO: 40 and 41.

Peptides A1 and A2 were further subjected to internal amino acid sequencing. For internal amino acid sequencing, 10 μg of toxin A was electrophoresized in 10% SDS-PAGE and transblotted to PVDF membrane. After the blot was stained with amido black, peptides A1 and A2, denoted TcdA_ii and TcdA_iii, respectively, were excised from the blot and sent to Harvard MicroChem and

Cambridge ProChem. Peptides were subjected to trypsin digestion followed by HPLC chromatography to separate individual peptides. N-terminal amino acid analysis was performed on selected tryptic peptide fragments. Two internal amino acid sequences of peptide A1 (TcdA_ii-PK71, SEQ ID NO:38 and TcdA_ii-PK44, SEQ ID NO: 39) were found to have significant homologies with deduced amino acid sequences of the TcbA_ii region of the tcbA gene (SEQ ID NO:12). Similarly, the N-terminal sequence (SEQ ID NO: 41) and two internal sequences of peptides A2 (TcdA_iii-PK57, SEQ ID NO: 42 and TcdA_iii-PK20, SEQ ID NO.43) also showed significant homology with deduced amino acid sequences of TcbA_iii region of the tcbA gene (SEQ ID NO: 12).

In summary of above results, the toxin complex has at least two active protein toxin complexes against SCR; toxin A and toxin B. Toxin A and toxin B are similar in their native and subunits molecular weight, however, their peptide compositions are different. Toxin A contained peptides TcdA_ii and TcdA_iii as the major peptides and the toxin B contains TcbA_ii and TcbA_iii as the major peptides. Example 16

Cleavage and Activation of TcbA Peptide

In the toxin B complex, peptide TcbA_ii and TcbA_iii originate from the single gene product TcbA (Example 15). The processing of TcbA peptide to TcbA_ii and TcbA_iii is presumably by the action of

Photorhabdus protease(s), and most likely, the metalloproteases described in Example 10. In some cases, it was noted that when

Photorhabdus W-14 broth was processed, TcbA peptide was present in toxin B complex as a major component, in addition to peptides TcbA_ii and TcbA_iii. Identical procedures, described for the purification of toxin B complex (Example 15), were used to enrich peptide TcbA from toxin complex fraction of W-14 broth. The final purified material was analyzed in a 4-20% gradient SDS-PAGE and major peptides were quantified by densitometry. It was determined that TcbA, TcbA_ii and TcbA_iii comprised 58%, 36%, and 6%, respectively, of total protein. The identities of these peptides were confirmed by their respective molecular sizes in SDS-PAGE and Western blot analysis using monospecific antibodies. The native molecular weight of this fraction was determined to be 860 kDa.

The cleavage of TcbA was evaluated by treating the above purified material with purified 38 kDa and 58 kDa W-14

Photorhabdus metalloproteases (Example 10), and Trypsin as a control enzyme (Sigma, MO). The standard reaction consisted 17.5 μg the above purified fraction, 1.5 unit protease, and 0.1 M Tris buffer, pH 8.0 in a total volume of 100 μl. For the control reaction, protease was omitted. The reaction mixtures were incubated at 37 °C for 90 min. At the end of the reaction, 20 μl was taken and boiled with SDS-PAGE sample buffer immediately for electrophoresis analysis in a 4-20% gradient SDS-PAGE. It was determined from SDS-PAGE that in both 38 kDa and 58 kDa protease treatments, the amount of peptides TcbA_ii and TcbA_iii increased about 3-fold while the amount of TcbA peptide decreased

proportionally (Table 23). The relative reduction and

augmentation of selected peptides was confirmed by Western blot analyses. Furthermore, gel filtration of the cleaved material revealed that the native molecular size of the complex remained the same. Upon trypsin treatment, peptides TcbA and TcbA_ii were nonspecifically digested into small peptides. This indicated that 38 kDa and 58 kDa Photorhabdus proteases can specifically process peptide TcbA into peptides TcbA_ii and TcbA_iii. Protease treated and untreated control of t he remaining 80 ul react ion mixture were seria l di lut ed with 10 mM sodium phosphate buffer, pH 7.0 and analyzed by SCR bioassay. By comparing activity in several dilution, it was determined that the 38 kDa protease treatment increased SCR insecticidal activity approximately 3 to 4 fold. The growth inhibition of remaining insects in the protease treatment was also more severe than control (Table 23).

Example 17

Screening of the library for a gene encoding the TcdA_ii Peptide

The cloning and characterization of a gene encoding the TcdA_ii peptide, described as SEQ ID NO: 17 (internal peptide

TcdA_ii-PT111 N-terminal sequence) and SEQ ID NO: 18 (internal peptide TcdA_ii-PT79 N-terminal sequence) was completed. Two pools of degenerate oligonucleotides, designed to encode the amino acid sequences of SEQ ID NO: 17 (Table 24) and SEQ ID NO: 13 (Table 25), and the reverse complements of those sequences, were synthesized as described in Example 8. The DNA sequence of the oligonucleotides is given below:

Polymerase Chain Reactions (PCR) were performed essentially as described in Example 3, using as forward primers F2.3.6.CB or P2.3.5, and as reverse primers P2.79.R.1 or P2.79R.CB, in all forward/reverse combinations, using Photorhabdus W-14 genomic DNA as template. In another set of reactions, primers P2.79.2 or

P2.79.3 were used as forward primers, and P2.3.5R, P2.3.5RI, and P2.3R.CB were used as reverse primers in all forward/reverse combinations. Only in the reactions containing P2.3.6.CB as the forward primers combined with P2.79.R.1 or P2.79R.CB as the reverse primers was a non-artifactual amplified product seen, of estimated size (mobility on agarose gels) of 2500 base pairs. The order of the primers used to obtain this amplification product indicates that the peptide fragment TcdA_ii-PT111 lies amino-proximal to the peptide fragment TcdA_ii-PT79.

The 2500 bp PCR products were ligated to the plasmid vector pCR™II (Invitrogen, San Diego, CA) according to the supplier's instructions, and tne DNA sequences across the ends of the insert fragments of two isolates (HS24 and HS27) were determined using the supplier's recommended primers and the sequencing methods described previously. The sequence of both isolates was the same. New primers were synthesized based on the determined sequence, and used to prime additional sequencing reactions to obtain a total of 2557 bases of the insert [SEQ ID NO:36].

Translation of the partial peptide encoded by SEQ ID No: 36 yields the 845 amino acid sequence disclosed as SEQ ID NO: 37.

Protein homology analysis of this portion of the TcdA_ii peptide fragment reveals substantial amino acid homology (68% similarity; 53% identity) to residues 542 to 1390 of protein TcbA [SEQ ID NO:12]. It is therefore apparent that the gene represented in part by SEQ ID NO: 36 produces a protein of similar, but not identical, amino acid sequence as the TcbA protein, and which likely has similar, but not identical biological activity as the TcbA protein.

In yet another instance, a gene encoding the peptides

TcdA_ii-PK44 and the TcdA_iii 58 kDa N-terminal peptide, described as SEQ ID NO:9 (internal peptide TcdA_ii-PK44 sequence), and SEQ ID NO:41(TcdA_iii 58 kDa N-terminal peptide sequence) was isolated. Two pools of degenerate oligonucleotides, designed to encode the amino acid sequences described as SEQ ID NO: 39 (Table 271 and SEQ ID NO:41 (Table 26), and the reverse complements of those sequences, were synthesized as described in Example 3, and their DNA sequences.

Polymerase Chain Reactions (PCR) were performed essentially as described in Example 8, using as forward primers A1.44.1 or A1.44.2, and reverse primers A2.3R or A2.4R, in all

forward/ reverse combinations, using Phot orhabdus W-14 genomic DNA as template. In another set of reactions, primers A2.1 or A2.2 were used as forward primers, and A1.44.1R, and A1.44.2R were used as reverse primers in all forward/reverse combinations.

Only in the reactions containing A1.44.1 or A1.44.2 as the forward primers combined with A2.3R as the reverse primer was a non-artifactual amplified product seen, of estimated size

(mobility on agarose gels) of 1400 base pairs. The order of the primers used to obtain this amplification product indicates that the peptide fragment TcdA_ii-PK44 lies amino-proximal to the 58 kDa peptide fragment of TcdAxn.

The 1400 bp PCR products were ligated to the plasmid vector pCR™II according to the supplier's instructions. The DNA sequences across the ends of the insert fragments of four isolates were determined using primers similar in sequence to the supplier's recommended primers and using sequencing methods described previously. The nucleic acid sequence of all isolates differed as expected in the regions corresponding to the

degenerate primer sequences, but the amino acid sequences deduced from these data were the same as the actual amino acid sequences for the peptides determined previously, (SEQ ID NOS:41 and 39).

Screening of the W-14 genomic cosmid library as described in Example 8 with a radiolabeled probe comprised of the DNA prepared above (SEQ ID NO:36) identified five hybridizing cosmid isolates, namely 17D9, 20B10, 21D2, 27B10, and 26D1. These cosmids were distinct from those previously identified with probes corresponding to the genes described as SEQ ID NO: 11 or SEQ ID NO:25. Restriction enzyme analysis and DNA blot

hybridizations identified three EcoR I fragments, of approximate sizes 3.7, 3.7, and 1.1 kbp, that span the region comprising the DNA of SEQ ID NO: 36. Screening of the W-14 genomic cosmid library using as probe the radiolabeled 1.4 kbp DNA fragment prepared in this example identified the same five cosmids (17D9, 20B10, 21D2, 27B10, and 26D1). DNA blot hybridization to EcoR I-digested cosmid DNAs also showed hybridization to the same subset of EcoR I fragments as seen with the 2.5 kbp TcdA_ii gene probe, indicating that both fragments are encoded on the genomic DNA.

DNA sequence determination of the cloned EcoR I fragments revealed an uninterrupted reading frame of 7551 base pairs (SEQ ID NO: 46), encoding a 232.9 kDa protein of 2516 amino acids (SEQ ID NO: 47). Analysis of the amino acid sequence of this protein revealed all expected internal fragments of peptides TcdA_ii (SEQ ID NOS:17, 18, 37, 38 and 39) and the TcdA_iii peptide N-terminus (SEQ ID NO: 41) and all TcdA_iii internal peptides (SEQ ID NOS:42 and 43). The peptides isolated and identified as TcdA_ii and TcdA_iii are each products of the open reading frame, denoted tcdA, disclosed as SEQ ID NO:46. Further, SEQ ID NO:47 shows, starting at position 89, the sequence disclosed as SEQ ID NO: 13, which is the N-terminal sequence of a peptide of size

approximately 201 kDa, indicating that the initial protein produced from SEQ ID No: 46 is processed in a manner similar to that previously disclosed for SEQ ID NO: 12. In addition, the protein is further cleaved to generate a product of size 209.2 kDa, encoded by SEQ ID NO: 48 and disclosed as SEQ ID NO: 49

(TcdA_ii peptide), and a product of size 63.6 kDa, encoded by SEQ ID NO:50 and disclosed as SEQ ID NO:51 (TcdA_iii peptide). Thus, it is thought that the insecticidal activity identified as toxin A (Example 15) derived from the products of SEQ ID MO: 46, as exemplified by the full-length protein of 282.9 kDa disclosed as SEQ ID NO: 47, is processed to produce the peptides disclosed as SEQ ID NOS:49 and 51. It is thought that the insecticidal activity identified as toxin B (Example 15) derives from the products of SEQ ID NO: 11, as exemplified by the 280.6 kDa protein disclosed as SEQ ID NO: 12. This protein is proteolytically processed to yield the 207.6 kDa peptide disclosed as SEQ ID

NO: 53, which is encoded by SEQ ID NO: 52, and the 62.9 kDa peptide having N-terminal sequence disclosed as SEQ ID NO: 40, and further disclosed as SEQ ID NO: 55, which is encoded by SEQ ID NO: 54.

Amino acid sequence comparisons between the proteins disclosed as SEQ ID NO: 12 and SEQ ID NO: 47 reveal that they have 69% similarity and 54% identity. This high degree of

evolutionary relationship is not uniform throughout the entire amino acid sequence of these peptides, but is higher towards the carboxy-terminal end of the proteins, since the peptides disclosed as SEQ ID NO: 51 (derived from SEQ ID NO: 47) and SEQ ID NO: 55 (derived from SEQ ID NO: 12) have 76% similarity and 64% identity.

Example 18

Control of European Cornborer- Induced Leaf Damage on Maize Plants by Spray Application of Photorhabdus (Strain W-14) Broth The ability of Photorhabdus toxin(s) to reduce plant damage caused by insect larvae was demonstrated by measuring leaf damage caused by European corn borer (Ostrinia nubilalis) infested onto maize plants treated with Photorhabdus broth. Fermentation broth from Photorhabdus strain W-14 was produced and concentrated approximately 10-fold using ultrafiltration (10,000 MW pore-size) as described in Example 13. The resulting concentrated broth was then filter sterilized using 0.2 micron nitrocellulose membrane filters. A similarly prepared sample of uinnoculated 2% proteose peptone #3 was used for control purposes. Maize plants (a

DowElanco proprietary inbred line) were grown from seed to vegetative stage 7 or 8 in pots containing a soilless mixture in a greenhouse (27°C day; 22°C night, about 50%RH, 14 hr day-length, watered/fertilized as needed). The test plants were arranged in a randomized complete block design (3 reps/treatment, 6 plants/treatment) in a greenhouse with temperature about 22°C day; 18°C night, no artificial light and with partial shading, about 50%RH and watered/ fertilized as needed. Treatments

(uinnoculated media and concentrated Photorhabdus broth) were applied with a syringe sprayer, 2.0 mls applied from directly (about 6 inches) over the whorl and 2.0 additional mls applied in a circular motion from approximately one foot above the whorl. In addition, one group of plants received no treatment. After the treatments had dried (approximately 30 minutes), twelve neonate European corn borer larvae (eggs obtained from commercial sources and hatched in-house) were applied directly to the whorl. After one week, the plants were scored for damage to the leaves using a modified Guthrie Scale (Koziel, M. G., Beland, G. L., Bowman, C, Carozzi, N. B., Crenshaw, R., Crossland, L., Dawson, J., Desai, N., Hill, M., Kadwell, S., Launis, K., Lewis, K., Maddox, D., McPherson, K., Meghji, M. Z., Merlin, E., Rhodes, R., Warren, G. W., Wright, M. and Evola, S. V. 1993).

Bio/Technology, 11, 194-195.) and the scores were compared statistically [T-test (LSD) p<0.05 and Tukey's Studentized Range (HSD) Test p<0.1]. The results are shown in Table 23. For reference, a score of 1 represents no damage, a score of 2 represents fine "window pane" damage on the unfurled leaf with no pinhole penetration and a score of 5 represents leaf penetration with elongated lesions and/or mid rib feeding evident on more than three leaves (lesions < 1 inch). These data indicate that broth or other protein containing fractions may confer protection against specific insect pests when delivered in a sprayable formulation or when the gene or derivative thereof, encoding the protein or part thereof, is delivered via a transgenic plant or microbe.

Example 19

Genetic Engineering of Genes for Expression in E. col i Summary of constructions

A series of plasmids were constructed to express the tcbA gene of Photorhabdus W-14 in Escherichia col i . A list of the plasmids is shown in Table 29. A brief description of each construction follows as well as a summary of the E. col i

expression data obtained.

Construction of pDAB634

In Example 9, a large EcoR I fragment which hybridizes to the TcbA_ii probe is described. This fragment was subcloned into pBC (Stratagene, La Jolla CA). Sequence analysis indicates that this fragment is 8816 base pairs. The fragment encodes the tcbA gene with the initiating ATG at position 571 and the terminating TAA at position 8086. The fragment therefore carries 570 base pairs of Photorhabdus DNA upstream of the ATG and 730 base pairs downstream of the TAA.

Construction of Plasmid pAcGP67B/tcbA

The tcbA gene was PCR amplified using the following primers; 5' primer (S1Ac51) 5' TTT AAA CCA TGG GAA ACT CAT TAT CAA GCA CTA TC 3' and 3' primer (S1Ac31) 5' TTT AAA GCG GCC GCT TAA CGG ATG GTA TAA CGA ATA TG 3'. PCR was performed using a TaKaRa LA PCR kit from PanVera (Madison, Wisconsin) in the following reaction: 57.5 ml water, 10 ml 10X LA buffer, 16 ml dNTPs (2.5 mM each stock solution), 20 ml each primer at 10 pmoles/ml, 300 ng of the plasmid pDAB634 containing the W-14 tcbA gene and one ml of TaKaRa LA Taq polymerase. The cycling conditions were 98°C/20 sec, 68°C/5 min, 72°C/10 min for 30 cycles. A PCR product of the expected about 7526bp was isolated in a 0.8% agarose gel in TBE (100 mM Tris, 90 mM boric acid, 1 mM EDTA) buffer and purified using a Qiaex II kit from Qiagen (Chatsworth, California). The purified tcbA gene was digested with Nco I and Not I and ligated into the baculovirus transfer vector pAcGP67B (PharMingen (San Diego, California)) and transformed into DH5α E. col i . The tcbA gene was then cut from pAcGP67B and transferred to pET27b to create plasmid pDAB635. A missense mutation in the tcbA gene was repaired in pDAB635. The repaired tcbA gene contains two changes from the sequence shown in Sequence ID NO: 11; an A>G at 212 changing an asparagine 71 to serine 71 and a G>A at 229 changing an alanine 77 to threonine 77. These changes are both upstream of the proposed TcbA_ii N-terminus.

Construction of pET15-tcbA

The tcbA coding region of pDAB635 was transferred to vector pET15b. This was accomplished using shotgun ligations, the DNAs were cut with restriction enzymes Nco I and Xho I. The resulting recombinant is called pET15-tcbA .

Expression of TcbA in E. col i from plasmid pET15-tcbA

Expression of tcbA in E. col i was obtained by modification of the methods previously described by Studier et al . (Studier,

F.W., Rosenberg, A., Dunn, J., and Dubendorff, J., (1990) Use of T7 RNA polymerase to direct expression of cloned genes. Methods Enzymol., 185: 60-89.). Competent E. col i cells strain BL21(DE3) were transformed with plasmid pET15-tcbA and plated on LB agar containing 100 μg/ml ampicillin and 40 mM glucose. The

transformed cells were plated to a density of several hundred isolated colonies/plate. Following overnight incubation at 37°C the cells were scraped from the plates and suspended in LB broth containing 100 μg /ml ampicillin. Typical culture volumes were from 200-500 ml. At time zero, culture densities (OD600) were from 0.05-0.15 depending on the experiment. Cultures were shaken at one of three temperatures (22°C, 30°C or 37°C) until a density of 0.15-0.5 was obtained at which time they were induced with 1 mM isopropylthio-β-galactoside (IPTG). Cultures were incubated at the designated temperature for 4-5 hours and then were transferred to 4°C until processing (12-72 hours).

Purification and characterization of TcbA expressed in E. col i from Plasmid pET15-tcbA.

E. coli cultures expressing TcbA peptides were processed as follows. Cells were harvested by centrifugation at 17,000 x G and the media was decanted and saved in a separate container.

The media was concentrated about 8x using the M12 (Amicon,

Beverly MA) filtration system and a 100 kD molecular mass cut-off filter. The concentrated media was loaded onto an anion exchange rolumn and the bound proteins were eluted with 1.0 M NaCl. The 1.0 M NaCl elution peak was found to cause mortality against Southern corn rootworm (SCR) larvae Table 30). The 1.0 M NaCl fraction was dialyzed against 10 mM sodium phosphate buffer pH 7.0, concentrated, and subjected to gel filtration on Sepharose CL-4B (Pharmacia, Piscataway, Mew Jersey). The region of the CL- 4B elution profile corresponding to calculated molecular weight (about 900 kDa) as the native W-14 toxin complex was collected, concentrated and bioassayed against larvae. The collected 900 kDa fraction was found to have insecticidal activity (see Table 30 below), with symptomology similar to that caused by native W- 14 toxin complex. This fraction was subjected to Proteinase K and heat treatment, the activity in both cases was either eliminated or reduced, providing evidence that the activity is proteinaceous in nature. In addition, the active fraction tested immunologically positive for the TcbA and TcbA_iii peptides in immunoblot analysis when tested with an anti-TcbA_iii monoclonal antibody (Table 30).

The cell pellet was resuspended in 10 mM sodium phosphate buffer, pH=7.0, and lysed by passage through a Bio-Neb™ cell nebulizer (Glas-Col Inc., Terra Haute, IN). The pellets were treated with DNase to remove DNA and centrifuged at 17,000 x g ro separate the cell pellet from the cell supernatant. The supernatant fraction was decanted and filtered through a 0.2 micron filter to remove large particles and subjected to anion exchange chromatography. Bound proteins were eluted with 1.0 M NaCl, dialyzed and concentrated using Biomax™ (Millipore Corp, Bedford, MA) concentrators with a molecular mass cut-off of 50,000 Daltons. The concentrated fraction was subjected to gel filtration chromatography using Sepharose CL-4B beaded matrix. Bioassay data for material prepared in this way is shown in Table 30 and is denoted as " TcbA Cell Sup".

In yet another method to handle large amounts of material, the cell pellets were re-suspended in 10 mM sodium phosphate buffer, pH = 7.0 and thoroughly homogenized by using a Kontes Glass Company (Vineland, NJ) 40 ml tissue grinder. The cellular debris was pelleted by centrifugation at 25,000 x g and the cell supernatant was decanted, passed through a 0.2 micron filter and subjected to anion exchange chromatography using a Pharmacia 10/10 column packed with Poros HQ 50 beads. The bound proteins were eluted by performing a NaCl gradient of 0.0 to 1.0 M.

Fractions containing the TcbA protein were combined and

concentrated using a 50 kDa concentrator and subjected to gel filtration chromatography using Pharmacia CL-4B beaded matrix. The fractions containing TcbA oligomer, molecular mass of approximately 900 kDa, were collected and subjected to anion exchange chromatography using a Pharmacia Mono Q 10/10 column equilibrated with 20 mM Tris buffer pH = 7.3. A gradient of 0.0 to 1.0 M NaCl was used to elute recombinant TcbA protein.

Recombinant TcbA eluted from the column at a salt concentration of approximately 0.3-0.4 M NaCl, the same molarity at which native TcbA oligomer is eluted from the Mono Q 10/10 column. The recombinant TcbA fraction was found to cause SCR mortality in bioassay experiments similar to those in Table 30.

Claims (1)

1. We claim :

   1. A composition, comprising an effective amount of a Phot orhabdus protein toxin that has functional activity against an insect.

   2. The composition of Claim 1, wherein the Phot ornabdus toxin is produced by a purified culture of Photorhabdus, a transgenic plant, Baculovirus, or heterologous microbial host.

   3. The composition of Claim 2, wherein the Photorhabdus toxin produced by a purified culture of Photorhabdus l uminescens .

   4. The composition of Claim 2, wherein the toxin is produced from a purified culture of Photorhabdus luminescens strain designated ATCC 55397.

   5. The composition of Claim 2, wherein the toxin is produced by a purified culture of Photorhabdus l uminescens strain designated W-14.

   6. The composition of Claim 1, wherein the toxin is produced by a purified culture of Photorhabdus strain designated WX-1, WX-2, WX-3, WX-4, WX-5, WX6 , WX-7, WX-8, WX-9, WX-10, WX-11, WX-12, WX-14, WX-15, H9 , Hb, Hm, HP88, NC-1, W30, WIR, ATCC# 43948, ATCC# 43949, ATCC# 43950, ATCC# 43951, or ATCC# 43952.

   7. The composition of Claim 2, wherein the toxin is produced from a purified culture of Photorhabdus l uminescens strain designated WX-1, WX-2, WX-3, WX-4, WX-5, WX-6, WX-7, WX-8, WX-9, WX-10, WX-11, WX-12, WX-14, WX-15, H9 , Hb, Hm, HP88, NC-1, W30, WIR, ATCC# 43948, ATCC# 43949, ATCC# 43950, ATCC# 43951, or ATCC# 43952. 8. The composition of Claim 1, wherein the toxin is respresented by amino acid sequence is SEQ ID NO: 12.

   9. The composition of Claim 6, wherein the composition is a mixture of one or more toxins produced from purified cultures ot Photorhabdus .

   10. The composition or Claim 1 or 6, therein the insect is ot the order Lepidoptera , Col eoptara , Hymenoptera , Diptera , Di ctyoptera , Acarina or Homoptera .

   11. The composition ot Claim 1 or 6 wherein the insect species is trom order Col eoptera and is Southern Corn Rootworm, Western Corn Rootworm, Colorado Potato Beetle, Mealworm, Boll Weevil or Turf Grub.

   12. The composition of Claim 1 or 6, wherein the insect species is from order Lepidoptera and is Beet Armyworm, Black Cutworm, Cabbage Looper, Codling Moth, Corn Earworm, European Corn Borer, Tobacco Hornworm, or Tobacco Budworm.

   13. The composition of Claim 1 or 6, wherein the toxin is rormulated as a sprayable insecticide.

   14. The composition of Claim 1 or Claim 6, wherein the toxin is formulated as a bait matrix and delivered in an above ground or below ground bait station.

   15. A method of controlling an insect, comprising orally delivering to an insect an effective amount of a protein toxin that has functional activity against an insect, wherein the protein is produced by a purified bacterial culture of the genus Photorhabdus .

   16. The method of Claim 15, wherein the bacterium is a purified culture of Photorhabdus l uminescens .

   17. The method of Claim 15, wherein the toxin is produced trom a purified culture of Photorhabdus l uminescens strain designated ATCC 55397

   18. The method of Claim 16, wherein the toxin is produced rrom a purified culture of Photorhabdus l uminescens strain designated W-14.

   15. The method of Claim 15, wnerein tne toxin is produced from a purified culture of Phot orhabdus strains designated WX-1, WX-2, WX-3, WX-4, WX-5, WX-6, WX-7, WX-3, WX-9, WX-13, WX-11, WX- 12, WX-14, WX-15, H9, Hb, Hm, HP88, NC-1, W30, WIR, ATCC# 43948, ATCC# 43949, ATCC# ATCC# 43950, ATCC# 43951, or ATCC# 43952.

   20. The method of Claim 15, wherein the toxin is produced trom a purified culture of Photorhabdus l uminescens strains des ignated WX- 1 , WX-2 , WX- 3 , WX- 4 , WX-5, WX-6, WX-7, WX-8, WX-9, WX-10, WX-11, WX-12, WX-14, WX-15, H9 , Hb, Hm, HP88, NC-1, W30,

   WIR, ATCC# 43948, ATCC# 43949, ATCC# ATCC# 43950, ATCC# 43951, or ATCC# 43952.

   21. The method of Claim 19, wherein a mixture of one or more toxins is produced from a purified culture of Photorhabdus and said toxins are orally delivered to an insect.

   22. The method of Claim 15, wherein the toxin is produced by a prokaryotic host transformed with a gene encoding the toxin.

   23. The method of Claim 15, wherein the toxin is produced by a eukaryotic host transformed with a gene encoding the toxin.

   24. The method of Claim 23, wherein the eukaryotic host is baculovirus.

   25. The method of Claim 15 or 19, wherein the insect is ot the order Lepidoptera , Coleoptera , Hymenoptera , Diptera ,

   Dictyoptera , Acarina or Homoptera .

   26. The method of Claim 15 or 19, wherein the insect species is from order Coleoptera and is Southern Corn Rootworm, Western Corn Rootworm, Colorado Potato Beetle, Mealworm, Boll Weevil or Turf Grub.

   27. The method of Claim 15 or 19, wherein the insect species is from order Lepidoptera and is Beet Armyworm, Black Cutworm, Cabbage Looper, Codling Moth, Corn Earworm, European Corn Borer, Tobacco Hornworm, or Tobacco Budworm.

   28. The method or Claim 15 or 19, wherein the toxin is formulated as a sprayable insecticide.

   29. The method of Claim 15 or Claim 19, wherein the toxin is formulated as a bait matrix and delivered in an above ground or below ground bait station.

   30. A method of isolating a gene coding for a protein subunit, comprising the steps of: constructing at least one RNA or DNA oligonucleotide molecule that corresponds to at least a part of a DNA coding region of an amino acid sequence selected trom a group consisting of SEQ ID NO:1, SEQ ID NO:2, SEQ ID NO: 3, SEQ ID NO:4, SEQ ID NO:5, SEQ ID NO:6, SEQ ID NO:7, SEQ ID NO: 8, SEQ ID NO: 9, SEQ ID NO: 10, SEQ ID NO: 13, SEQ ID NO: 14, SEQ ID NO:15, SEQ ID NO:16, SEQ ID NO:17, SEQ ID NO:18, SEQ ID NO:19 SEQ ID NO:20, SEQ ID NO:21, SEQ ID NO:22, SEQ ID NO:23, SEQ ID NO: 24, SEQ ID NO: 38, SEQ ID NO: 39, SEQ ID NO: 40, SEQ ID NO: 41, SEQ ID NO: 42, and SEQ ID NO: 43, wherein the nucleotide molecule is used to isolate genetic material from Photorhabdus or

   Photorhabdus l uminescens .

   31. A method for expressing a protein produced by a purified bacterial culture of the genus Photorhabdus in a prokaryotic or eukaryotic host in an effective amount so that the protein has functional activity against an insect, wherein the method comprises: constructing a chimeric DNA construct having 5' to 3' a promoter, a DNA sequence encoding a protein, a transcription terminator, and then transferring the chimeric DNA construct into the host.

   32. The method of Claim 31, wherein the protein has functional activity against insects selected from a group consisting of Coleoptera , Lepidoptera , Diptera , Homoptera ,

   Hymenoptera , Dictyoptera , and Acarina .

   33. The method of Claim 31, wherein the protein encoded by the DNA sequence has an N-terminal amino acid sequence selected rrom the group consisting or SEQ ID NO:1, SEQ ID NO:2, SEQ ID NO: 3, SEQ ID NO: 4, SEQ ID NO: 5, SEQ ID NO : 6 , SEQ ID NO:7, SEQ ID NO:8, SEQ ID NO: 9. SEQ ID NO: 10, SEQ ID NO: 13, SEQ ID NO: 14, SEQ ID 11O:15, SEQ ID 13O: 16 , SEQ ID NO:17, SEQ ID NO : 13 , SEQ ID NO:19, SEQ ID NO:20, SEQ ID NO : 21, SEQ ID NO:22, SEQ ID NO:23, SEQ ID NO:24, SEQ ID NO : 38 , SEQ ID NO : 39, SEQ ID NO:40, SEQ ID NO:41, SEQ ID NO: 42, and SEQ ID NO: 43.

   34. The method of Claim 31, wherein the protein encoded by the DNA sequence includes the amino acid sequence selected from the group consisting of SEQ ID NO: 12, SEQ ID NO:26, SEQ ID NO:23, SEQ ID NO:30, SEQ ID NO:32, SEQ ID NO:34, SEQ ID NO:35, SEQ ID NO:47, SEQ ID NO:49, SEQ ID NO:51, SEQ ID NO:53, SEQ ID NO:55, SEQ ID NO: 57, SEQ ID NO: 59 and SEQ ID NO: 61.

   35. A chimeric DNA construct, adapted for expression in a prokaryotic or eukaryotic host comprising, 5' to 3' a

   transcriptional promoter active in the host; a DNA sequence encoding a Photorhabdus protein that has functional activity against an insect; and a transcriptional terminator.

   36. A chimeric DNA construct of Claim 35, wherein the protein encoded by the DNA sequence has an N-terminal amino acid sequence selected from the group consisting of SEQ ID NO:1, SEQ ID NO: 2, SEQ ID NO: 3, SEQ ID NO: 4, SEQ ID NO: 5, SEQ ID NO: 6, SEQ ID NO: 7, SEQ ID NO: 8, SEQ ID NO: 9, SEQ ID NO: 10, SEQ ID NO: 13, SEQ ID NO: 14, SEQ ID NO: 15, SEQ ID NO: 16, SEQ ID NO: 17, SEQ ID NO:18, SEQ ID NO:19, SEQ ID NO:20, SEQ ID NO:21, SEQ ID NO:22, SEQ ID NO: 23, SEQ ID NO: 24, SEQ ID NO: 38, SEQ ID NO: 39, SEQ ID NO:40, SEQ ID NO:41, SEQ ID NO:42, and SEQ ID NO:43.

   37. The chimeric DNA construct of Claim 35, wherein the protein encoded by the DNA sequence has an amino acid sequence selected from the group consisting of SEQ ID NO: 12, SEQ ID NO: 26, SEQ ID NO:28, SEQ ID NO:30, SEQ ID NO:32, SEQ ID NO:34, SEQ ID NO: 35, SEQ ID NO: 47, SEQ ID NO: 49, SEQ ID NO: 51, SEQ ID NO: 53, SEQ ID NO:55, SEQ ID NO:57, SEQ ID NO:59, and SEQ ID NO:61.

   38. The chimeric DNA construct of Claim 35, wherein the DNA sequence encoding the Photorhabdus l uminescens protein is selected from the group comprising SEQ ID NO: 11, SEQ ID NO: 25, SEQ ID NO:27, SEQ ID NO:29, SEQ ID NO : 31 , SEQ ID NO : 33 , SEQ ID NO: 46, SEQ ID NO: 48, SEQ ID NO: 50, SEQ ID MO: 52, SEQ ID NO: 54, SEQ ID NO:56, SEQ ID MO: 58, and SEQ ID NO:60.

   39. The chimeric DNA construct of Claim 35, wherein the host is baculovirus.

   40. An isolated and substantially purified preparation comprising, a DNA molecule capable of encoding an effective amount of a protein that is produced by a bacterium of tne genus Photorhabdus and that has functional activity against an insect.

   41. The preparation of Claim 40, wherein the bacterium is Photorhabdus luminescens . 42. A purified preparation comprising, a protein produced by Photorhabdus or Photorhabdus l uminescens having an N-terminal amino acid sequence selected from the group consisting of SEQ ID NO:1, SEQ ID NO: 2, SEQ ID NO: 3, SEQ ID NO: 4, SEQ ID NO: 5, SEQ ID NO: 6, SEQ ID NO: 7, SEQ ID NO: 8, SEQ ID NO: 9, SEQ ID NO: 10, SEQ ID NO: 13, SEQ ID NO: 14, SEQ ID NO: 15, SEQ ID NO: 16, SEQ ID NO: 17, SEQ ID NO:18, SEQ ID NO: 19, SEQ ID NO:20, SEQ ID NO:21, SEQ ID NO:22, SEQ ID NO:23, SEQ ID NO:24, SEQ ID NO:38, SEQ ID NO:39, SEQ ID NO:40, SEQ ID NO:41, SEQ ID NO:42, and SEQ ID NO:43. 43. A purified protein preparation comprising, a protein that has an N-terminal amino acid sequence selected from the group consisting of SEQ ID NO:1, SEQ ID NO: 2, SEQ ID NO: 3, SEQ ID NO: 4, SEQ ID NO: 5, SEQ ID NO: 6, SEQ ID NO: 7, SEQ ID NO: 8, SEQ ID NO: 9, and SEQ ID NO: 10, SEQ ID NO: 13, SEQ ID NO: 14, SEQ ID NO: 15, SEQ ID NO: 16, SEQ ID NO: 17, SEQ ID NO: 18, SEQ ID NO: 19, SEQ ID NO:20, SEQ ID NO:21, SEQ ID NO:22, SEQ ID NO:23, SEQ ID NO:24, SEQ ID NO: 38, SEQ ID NO: 39, SEQ ID NO: 40, SEQ ID NO: 41, SEQ ID NO: 42, and SEQ ID NO: 43. 44. A purified protein preparation comprising, a protein selected from the group of SEQ ID NO: 12, SEQ ID NO: 26, SEQ ID NO: 28, SEQ ID NO: 30, SEQ ID NO: 32, SEQ ID NO: 34, SEQ ID NO : 35, SEQ ID NO: 47, SEQ ID NO: 49, SEQ ID NO : 51, SEQ ID NO: 53, SEQ ID NO: 55, SEQ ID NO: 57, SEQ ID NO: 59, and SEQ ID NO: 61.

   45. A purified DNA preparation comprising, a DNA sequence selected from the group consisting of SEQ ID NO:11, SEQ ID NO:15, SEQ ID NO:27, SEQ ID NO : 29 , SEQ ID NO : 31, SEQ ID NO:33, SEQ ID NO:46, SEQ ID NO:48, SEQ ID NO:50, SEQ ID NO: 52, SEQ ID NO:54, SEQ ID NO: 56, SEQ ID NO: 58 and SEQ ID NO: 60, wherein the DNA sequence is isolated from its native host

   46. A purified protein preparation comprising, a

   Photorhabdus luminescens protein with at least one subunit having an approximate molecular weight between 18 kDa to about 230 kDa; between about 160 kDa to about 230 kDa; 100 kDa to 160 kDa; about 80 kDa to about 100 kDa; or about 50 kDa to about 80 kDa.

   47. A purified protein preparation comprising, a

   Photorhabdus luminescens protein with at least one subunit having an approximate molecular weight of about 280 kDa.

   48. A substantially pure microorganism culture comprising, ATCC 55397.

   49. The culture of Claim 48, wherein the culture is a derivative of ATCC 55397 that produces a protein toxin that has functional activity against an insect. 50. A substantially pure microorganism culture comprising, H9.

   51. A substantially pure microorganism culture comprising, Hb.

   52. A substantially pure microorganism culture comprising, Hm.

   53. A substantially pure microorganism culture comprising, HP88.

   54. A substantially pure microorganism culture comprising, NC-1. 55. A substantially pure microorganism culture comprising, W30.

   56. A substantially pure microorganism culture comprising, WIR.

   57. A transgenic plant comprising in its genome, a chimeric artificial gene construction imbuing the plant with an ability to express an effective amount of a Photorhabdus protein that has functional activity against an insect.

   58. The transgenic plant of Claim 57, wherein the plant is transformed using acceleration of genetic material coated onto microparticles directly into cells, Agrobacteria, whiskers, or electroporation techniques

   59. The transgenic plant of Claim 57, wherein the

   selectable marker is selected from the group consisting of kanamycin, neomycin, glyphosate, hygromycin, methotrexate, phosphinothricin (bialophos), chlorosulfuron, bromoxynil, dalapon and the like.

   60. The transgenic plant of Claim 57, wherein the promoter is selected from the group consisting of octopine synthase, nopaline synthase, mannopine synthase, 35S, 19S, ribulose-1,6-bisphosphate (RUBP) carboxylase small subunit (ssu), beta-conglycinin, phaseolin, alcohol dehydrogenase (ADH), heat-shock, ubiquitin, zein, oleosin, napin, or acyl carier protein (ACP).

   61. The transgenic plant of Claim 57, wherein embryogenic tissue, callus tissue type I or II, hypocotyl, meristem, or plant tissue during dedifferentiation is used in preparing the

   transgenic plant.

   62. The transgenic plant of Claim 57, wherein the chimeric gene is a DNA sequence which encodes a Photorhabdus protein that has functional activity against an insect and at least one codon of the gene has been modified so that the codon is a plant preferred codon.

   63. A method of controlling an insect comprising orally delivering to an insect an effective amount of a protein toxin, wherein the protein is produced by a transgenic plant, which said insect feeds.

   64. A composition of matter, comprising a purified DNA sequence from a purified bacterial culture from the genus

   Photorhabdus.