GOVERNMENT RIGHTS

This invention was made with United States Government support under NIH Tropical Diseases Research Unit Program Project grant POl AI28778 and NIH grants ROl AI08614 and AI10984. The United States Government has certain rights in this invention.

BACKGROUND OF THE INVENTION

Field of the Invention

The present invention concerns methods for the facile stable introduction of foreign genes into the symbiotic bacteria of arthropods and helminths using a symbiont transformed with a shuttle plasmid vector.

Background Information

Insects transmit diseases to man and livestock, but the permanent control of harmful insects has not yet been achieved.

In the short run, chemical methods are the most practical and economic methods for the control of harmful insects. Unfortunately long-acting chemicals enter the environmental food chain and may be ecologically unacceptable. However, the major difficulty both with short and long acting insecticides is that insects have numerous mechanisms for developing

insecticide resistance. Among the most powerful resistance mechanisms is the expression of enzymes which degrade insecticides. The genes for such enzymes either reside in the insect or can be spread laterally among the insect population by bacterial plasmid vectors.

To replace chemical control mechanisms, physical methods such as the spreading of oil for mosquito larval control and insect traps continue to be used. Trapping methods have become more sophisticated with the development of attractants and phero ones. These methods are highly effective both in theory and practice, but they are consumptive of effort. In areas where the economy is marginal and all available effort is needed to raise enough food for survival, it is frequently difficult to motivate people to expend effort to mitigate a nuisance which is seen as a natural concomitant of life.

A third important approach has been the development of biological control elements. The release of sterile male insects, the cultivation of plants expressing gene products which deter or kill insects, or insect-borne pathogens, the importation or raising of insect predators and the introduction or manipulation of insect parasites or lethal bacteria have been widely employed. These methods, however, are not uniformly successful.

The introduction of sterile males was originally successful in the control of the cattle screw worm, but was totally ineffective in controlling tsetses carrying human and cattle trypanosomes. These methods are very expensive. Both insect predators and insect parasites depend for their continued existence on a stable balance between uninfected

insects and the parasite/predator. Such stable balances are difficult to achieve.

Both the trapping and the biological control mechanisms are subject to the sink effect. Human habitation over much of the world is scattered and is surrounded by regions such as forests and deserts. Even when a village decreases the number of tsetses by trapping, unless the effort is continuous, the numbers will quickly be replenished by insects from the surrounding environment. Hence effective insect control often means applying the control measures to large regions and this is economically unrealistic.

In summary, control of harmful insects—as represented by mosquito and tsetse control—have made few real practical advances in the last 45 years; in many third world countries there is now less insect control than in the colonial era.

Given the enormous adaptive capacities of insects, methods have been aimed not at the reduction of insect numbers, but at changing specific insect functions. Such functions include the ability of insect vectors to carry disease-causing parasites or to develop resistance to insecticides.

Attempts at the genetic modification of insects have been made. Certain insects, such as Drosophila melanogaster have a large complement of mobile genetic elements in their genome. There have been efforts to introduce foreign genes into insects such as osquitos, using the Drosophilamobile elements as vectors. These efforts have had few practical results. One important difficulty is with the design of the project (Chandler, A. M.

and O'Brochta, D. A., (1991), "Prospects for Gene Transformation in Insects", Ann. Rev. Entomol.. 36. 159-163). Insects are highly evolved systems. Alteration of a single gene and introduction of foreign genes generally affect many other genes in the insect. Thus, for survival, a neutral or advantageous gene has to be introduced. However, most genetic alterations are harmful. Thus the introduction of each new gene into somatic tissues of an insect is an experiment in genetic selection. Each experiment takes a long time and by definition most experiments will introduce harmful mutations.

In recent years increasing stress has been laid on the modification of specific insect functions, which, in the anthropocentric view, are not desirable.

Approaches to expressing foreign genes in insect vectors which are under investigation elsewhere involve creating transgenic insects by injecting foreign DNA directly into embryos or using insect mobile elements for this purpose (Miller, L.H. , Sakai, R.K., Romans, P., Owadz, R. . , Kantoff, P., Coon, H.G., (1987), "Stable Integration and Expression of a Bacterial Gene in the Mosquito Anopheles Gambiae , Science. 237:779-781; McCrane, V., Carlson, J.P., Muller, B.R. , Beaty, B.J., (1988), "Micro-injection of DNA into Aedes triseriatus ova and Detection of Integration", Am. J. Trop. Med. Hγq.. 39:502-510; Morris, A.C., Eggleston, P., Crampton, J.M., (1989), "Genetic Transformation of the Mosquito Aedes aegypti By Micro-Injection of DNA", Med. Vet♦ Entomol.. 3.:1-7). These techniques have been used with mosquitoes and are potentially valuable for achieving the common goal of modifying vector competence; however, they are hindered primarily by the lack of effective transformation vectors.

The use of mobile elements such as the p-element, to carry transgenes is an elegant method which so far has produced only a few results in mosquitoes; the insects in which most research efforts have been concentrated. While rare transformations have been observed in Aedes aegypti, Aedes triseriatus and Anopheles gambiae , their frequency has been much lower than would be expected of a vector as efficient as the P- element and it seems possible that the integrative events were not P-element mediated. With further effort these methods may eventually result in the production of transgenic insects. Yet, however, if these methods are refined, insect mobile elements are essentially promiscuous and insert, if not completely randomly, at least into many different DNA sites in their insect host. In an heterologous insect host, unless the site of DNA insertion can eventually be controlled, each transformation will result in "random" mutations. The vast majority of random mutations are unfavorable. This will bring problems at the next stage, where the inserted transgene needs to be spread through the insect population and an unfavorable mutation may inhibit spread. Thus, only when the procedures of transgene insertion have been completed and the transgene has been shown to be stable can the transgenic insect host be assessed for fitness. The length of these procedures combined with the effect of unfavorable mutations on gene spread, makes it desirable that other methods of transgene insertion, which do not involve insertion into germ-line or somatic eukaryotic insect cells, should be developed. These methods should have strong selection properties in the heterologous host and should be both rapid and simple.

Blood-feeding insects are vectors of protozoan parasites that cause a number of important human and animal diseases. Many hematophagous insects also contain symbiotic bacteria that supply nutrients that are essential for their hosts (Buchner, P., (1965), Endosymbiosis of Animals with Plant Microorganisms, New York: John Wiley & Sons Dasch, G.A. , Weiss, E., Chang, K. (1984), "Endosymbionts of Insects", Krieg, N.R. , Cd., Bergey's Manual of Systematic Bacteriology. Vol. 1, Baltimore: Williams & Wilkins, 811-833; Ishikawa, H. , (1989) , "Biochemical and Molecular Aspects of Endosymbiosis in Insects," Int. Rev. Cvtol.. 116; 1-45).

Over 100 years ago Blochmann described bacteria-like bodies in insects (Blochmann, F., Bacterienahnliche Kόrperchen in den Geweben und Eiern, (1887) , Biologisches Zentralblatt. Z, 606) . It has been estimated that 300 million years ago some insects were invaded by bacteria, which, by managing to evade the host's immune system and by establishing a maternal transmission route (typically transovarial or coprophagous transmission) became obligatory insect symbionts.

Symbionts within single insects may be of multiple bacterial and microorganism species and in the insect kingdom many species of bacteria are endosymbiotic. Symbionts include intra-cellular species permanently associated with an insect cell and of which no free-living bacterial forms have ever been found. It has been held that the mitochondrion is an extreme example of such ancient symbiosis. At the other end of the range there are free-living forms which are readily cultured in bacterial media in vitro . Some of the cultured insect symbionts have both intra-cellular and extracellular forms (Welburn, S.C., Maudlin, I. and Ellis, D.S., (1987), In vitro

Cultivation of a Rickettsia-like Organism From Glossina spp .", Ann. Trop. Med. Parasitol.. 81, 331-335; Beard, C.B., Mason, P.W., Aksoy, S., Tesh, R.B. and Richards, F.F., (1992), "Transformation of an Insect Symbiont and Expression of a Foreign Gene in the Chagas disease vector, Rhodnius prolixus" , J. Trop. Med. Hyg.. 46. 195-200) .

The classification of species of bacteria which have symbiotic habitats in insects, has been studied phylogenetically by analysis of their 16S rRNA genes. These include Wolbachia-like species (O'Neill, S.L., Giordano, R. , Colbert, A.M.E., Karr, T.L. and Robertson, H. (1992)" 16S RNA Phylogenetic Analysis of the Bacterial Endosymbionts Associated with Cytoplasmic Incompatibility in Insects", Proc. Natl. Acad. Sci. (USA). 89_:2699-2702) which are related to the Rickettsia , enterocyte-like and actino ycotic species (Beard, C.B., Mason, P.W. , Aksoy, S., Tesh, R.B. and Richards, F.F., (1992) , "Transformation of an Insect Symbiont and Expression of a Foreign Gene in the Chagas Disease Vector, Rhodnius prolixus" , J. Trop. Med. Hyg. 46. 195-200; O'Neill, S.L., Giordano, R. , Colbert, A.M.E., Karr, T.L. and Robertson, H. , (1992), "16S RNA Phylogenetic Analysis of the Bacterial Endosymbionts Associated with Cytoplasmic Incompatibility In Insects", Proc. Natl. Acad. Sci. (USA), 89:2699-2702) and undoubtedly there are many more.

The functions of symbionts vary both in symbiont species and in different insects. Probably all monophagous insects (which have a single food source) including reduviid bugs, tsetses, aphids, leaf-hoppers and ticks, have symbionts. The symbionts synthesize insect nutrients not found in the single food source; i.e., blood or plant juice. Symbionts synthesize

amino acids, especially the sulfur-containing amino acids. For instance, Myzus persicae , the green peach aphid obtains only three essential amino acids from its plant juice diet; the rest are synthesised by its symbionts (Dadd, R.H. and Krieger, D.L., (1968), "Dietary Amino Acid Requirements of the Aphid Myzus persicae" , J. Insect. Physiol.. 14. 71-74). Sterols, pantothenic acid, biotin and choline are just a few of the substances synthesised by symbionts (Ishikawa, Hajime. In Insect Endosymbiosis. Ed. Schwemmler, W. , CRC Press, (1989), pp. 123-143) . Symbionts may also be active in catabolism, breaking down uric acid produced by insects (Lamb, K.P., (1959) , "Composition of the Honeydew of the Aphid Brevicoryne brassica (L) Feeding on Swedes (Brassicae neobrassicae D.C. ) ", J. Insect. Phvsiol.. 15, 155-162).

Most studies on symbiont function employ antibiotics which remove symbionts and produce aposymbiotic insects which then lack the nutrients described. In aphids, symbionts provide the machinery to synthesise a major 65 KD storage protein, yet the genes directing its synthesis are insect genes (Ishikawa, Hajime, (1984) , "Control of Macromolecule Synthesis in the Aphid Endosymbiont by the Host Insect", Comp. Biochem. and Physiol. , 73, 51-56) . An example of an intricately linked symbiont function is described by Schwemmler (Reviewed: Tiivel, Toomas, "Endocytobiosis of Leafhoppers with Prokaryotic Microorganisms", In Insect Endosymbiosis. Ed. Schwemmler, W. , CRC Press, (1989) , 111-122) . If certain symbionts are removed with lysozyme from the eggs of Euscelidius variegatus , a leafhopper, the resulting imagos (adult insects) are cephalothorax forms, i.e. they lack an abdomen. If other symbionts are removed from eggs with antibiotics, the insect's head anlage is missing.

These phenomena may be due to the presence of insect homeobox or structural genes in the symbionts or it may be that the symbiont produces factors which control insect structural gene expression. Insect symbiont functions are closely integrated with insect host functions and there is a complicated mutual dependence which anchors the symbiont to its insect host. In any genetic manipulation of symbionts, it is therefore necessary to preserve their physiological functions vis-a-vis their insect host.

A highly relevant function of certain other symbionts carried in insect gonads is to confer on the infected insect host, a reproductive advantage. These symbionts mediate a mechanism whereby the only fertile matings occur between insects carrying the same gonadal symbiont. Matings with uninfected insects of the same species or with insects infected with another strain of gonadal symbiont, are infertile. A number of non-gonadal insect symbionts carry very large plasmids over 150 Kb in size. For most of the symbiont functions described above, it is not clear whether the functions are coded by symbiont plasmid or genomic DNA. Arthropod resistance to insecticides is of great commercial importance, and in only a few instances the resistance mechanism has been localized to the symbiont or to the symbiont-plasmid for (Lalithakumari, D. and Annamalai, P., "Edifenphos Resistance in Pyricularia oryzae and Drechslera oryzae' in "Managing Resistance to Agrochemicals" Eds., Green, M.D., LeBaron, Homer, M. and Moberg, W.K., Am. Chem. Soc.. Washington, DC, (1990), pp 249-263).

The known functions of symbiont plasmids include coding for enzymes which detoxicate exobiotic (insecticidal)

substances in the environment. Copper, for instance, is present as copper sulfate in Bordeaux mixture, an anti-aphid spray which was once effective. A plasmid codes for enzymes which produce H ₂ S from sulfate, converting soluble copper salts toxic to the insect into insoluble non-toxic copper sulfide (Bender, CL. and Cooksey, D.A. , (1986), "Indigenous Plasmids in Pseudomonas syringae p.v. Tomato Conjugative Transfer and Role in Copper Resistance", J. Bacteriol.. 165. 534-541; Erardi, F.X., Failla, M.L., Falkinham, J.O. III., (1987), "Plasmid-encoded Copper Resistance and Precipitation by Mycobacterium Scrofulaceum", APPI. Envir. Microbiol.. 53. 1951-1954) . Detoxication of a prevalent exobiont may itself be a gene-driving mechanism.

Other bacterial symbiont functions include effects on host range and infectivity (Takai, Shinji, Sekizaki, T. , Ozawa, T., Sugawara, T., Watanabe, Y. and Tsubaki, S., (1991), "Association Between a Large Plasmid and 15-17 KD Antigens in Virulent Rhodococcus equi. " , Inf. Imm.. 59. 4056-4060; Bowles, P.D., Woolcock, J.B. and Muti er, M.D., (1987), "Experimental Infection of Mice With Rhodococcus equi'. Differences in Virulence Between Strains", Vet. Microbiol. 14. 259-268). The synthesis of antibiotics is often a function of bacterial plasmids and is presumably a device to control the biotic environment of the bacterium (Hopwood, D.A. , (1978) , "Extrachromosomally Determined Antibiotic Production", Ann. Rev. Microbiol.. 32. 373-392) . In general, bacterial plasmids tend to code for "dispensible" functions, which while very important, are not needed at all times by the bacterium.

An example of a symbiont is Rhodococcus rhodnii , which is an actinomycete symbiont that inhabits Rhodnius prolixus and is essential for normal growth and development of the bug (Baines, S., (1956), "The Role of the Symbiotic Bacteria in the Nutrition of Rhodnius prolixus (Hemipteran) , J. Exp. Biol..

3_3:533-541) . The symbiont lives extracellularly in the lumen of the insect gut, and is transmitted from adult to progeny by egg shell contamination or by coprophagy. Bugs can be made aposymbiotic (free of symbionts) by surface sterilizing the eggs and rearing the resulting progeny in a clean environment free from adult bugs, their wastes and remains. Aposymbiotic individuals have higher stage-specific mortality rates and never reach reproductive maturity. Since both Trypanosoma cruzi and R. rhodnii grow in proximity within the insect gut, the symbiont is potentially useful as a vehicle for expressing a gene product within the gut lumen that would affect attachment or development of the parasite and thus render the insect incapable of transmitting the protozoan parasite.

Arthropods are vectors of important diseases of man, animals and plants. In recent years it has become clear that total eradication of disease-carrying arthropods is not feasible. Consequently, increasing attention is being given to the specific molecular and biochemical mechanisms of arthropod-parasite interaction as potential points for disease intervention. Attempts have been made to introduce foreign genes (and eventually anti-parasitic genes) into insects using embryonic icroinjection (Miller, L.H. , Sakai, R.K., Romans, P., Gwadz, R.W. , Kantoff, P.and Coon, H.G., (1987), "Stable Integration and Expression of a Bacterial Gene in the Mosquito Anopheles gambiae" , Science. 237: 779-781; McGrane, V., Carlson, J.O., Miller, B.R. and Beaty, B.J. , (1988), "Microinjection of DNA into Aedes triseriatus ova and detection of integration", Am. J. Trop. Med. Hyg. 3_9:502-510; Morris, A.C., Egglestoη, P. and Cra pton, J.M., (1989), "Genetic Transformation of the Mosquito Aedes aegypti by micro-injection of DNA", Med. Vet. Entomol. , 3 _. :1-7). These methods have been only of limited success in non-drosophilids.

A recent letter in Nature. (June 11, 1992) , 357. 450 discusses the possibility of inserting a gene for inability to transmit malaria into the Wolbachia of an Anopheles mosquito. This is not an easy task since presently Wolbachia cannot be grown free of insect cells. Furthermore, the mosquito probably does not have the usual type of bacterial symbionts.

A shuttle plasmid is an extrachromosomal DNA element that is capable of replicating in two very different microorganisms. It contains the plasmid replication origins of both microbes, allowing it to maintain itself in either host. It can also contain antibiotic resistance genes that are expressed differentially in transformed bacteria and thus allow for selection in both systems.

SUMMARY OF THE INVENTION

An object of the present invention is to alter or introduce specific functions into arthropods or helminths.

Another object of the present invention is to modify the ability of arthropods or helminths to carry disease-producing parasites.

A further object of the present invention is to delay the development of pesticide resistance in arthropods or helminths.

A still further object of the present invention is to retard the ability of arthropods and insects to metabolize non-toxic compounds or toxic catabolites.

Still another object of the present invention is to genetically modify the symbiont bacteria of helminths or arthropods.

Another object of the present invention is to produce helminths or arthropods which cannot carry parasites causing animal and human diseases (vector incompetent insects) .

These objects, as well as other objects, aims and advantages are satisfied by the present invention.

Many blood-feeding insects, because of their restricted diet, maintain endosymbiotic bacteria which provide their host with nutritional supplements.

The method of the present invention uses genetic alteration of the endosymbiotic bacteria (symbionts) of arthropods or helminths to introduce foreign genes into the symbionts.

In the present invention, plasmid vectors serve to carry genes whose products can modify particular insect functions.

The present invention is based on a system for using bacterial symbionts as vehicles for expressing foreign genes in insects, with the ultimate goal of expressing a gene product or products that would reduce the vector competence of the insect for a specific pathogenic agent.

Instead of producing transgenic insects by introducing foreign genes directly into insect somatic tissues via mobile elements, the method of the present invention uses a simpler and much more versatile technique. The method of the present invention involves transformed symbiotic bacteria (symbionts) , stably associated with an insect species. One example of the present invention employs for transfection a shuttle plasmid vector carrying the origin of replication of both, for

example, E. coli, and of the symbiont and multiple selectable marker genes.

Such genetically altered symbionts are stable in their hosts and carry out the physiological functions of the symbiont in the insect.

The methods of the present invention avoid the complications of the random insertion of mobile elements into eukaryotic cells and substitute instead the well developed methods of, for example, bacterial genetics.

These methods can be used to introduce genes corresponding to synthetic peptides related to the pore- forming antibiotics. These can have highly selective anti- parasitic actions.

These methods can be extended to introduce genes into insects whose products activate pro-insecticides in the insect and also genes whose products slow the development of insecticide resistance by acting on insect detoxicating exobiotic substances.

The ultimate aim of the present invention is to drive newly introduced genes throughout an insect population. Certain insects obtain reproduction advantage by being infected with a different class of symbiont which resides in gonadal tissues. When an infected insect mates either with an uninfected insect or an insect carrying another strain of this class of symbionts; the mating is sterile. Matings of two insects carrying the same gonadal symbiont are fertile, causing the spread of the gonadal symbiont through a

population. Other gene-driving mechanisms, based on symbiont plasmids expressing products which detoxify common exobiotic substances such as heavy metals, can also be utilized.

The present invention concerns a method of expressing gene products in host insects, i.e., arthropods or helminths, which products are detriment to disease causing parasites, bacteria or viruses carried by the host, which products interfere with disease transmission in plants, animals or humans. The method comprises:

(a) removing endosymbiotic bacteria (symbionts) from the host by administering an antibiotic,

(b) transforming a symbiont with a shuttle plasmid vector carrying the origin of replication of both (i) a shuttle plasmid DNA amplifier selected from the group consisting of yeasts and bacteria and (ii) the symbiont, the shuttle plasmid vector also carrying multiple, selectable marker genes mediating antibiotic resistance, the shuttle plasmid vector carrying at least one foreign gene whose product has anti- parasitic, anti-bacteria or anti-viral properties or genes whose products retard the development of insect-resistance to insecticides,

(c) infecting the host with the transformed symbiont from step (b) and

(d) introducing to the host a strain of Wolbachia pipientis which gives reproductive advantage to the host.

BRIEF DESCRIPTION OF THE DRAWINGS

Fig. 1 is a schematic diagram showing the protocol used for developing an Escherichia coli-Rhodnius rhodnii shuttle plasmid. Amp ^R = ampicillin resistant gene; Thio ^R =thiostrepton resistant gene; Ori = origin of replication.

Fig. 2 is a schematic diagram showing the physical map of the shuttle plasmid pRrl. The large arrows indicate the direction of transcription of the Escherichia coli pBR 322-derived ampicillin resistance gene (Amp ^R ) and the thiostrepton resistance gene (Thio ^R ) from Streptomyces azureus (hatched region). The small arrow indicates the £ coli pBR322-derived origin of replication (ori) . The stippled region is derived from the endogenous 6.5-kb plasmid from Rhodnius-rhodnii ATCC strain 35071. Restriction enzymes sites are indicated; unique sites are underlined.

Fig. 3 is a light micrograph showing intracellular and extracellular tsetse symbionts.

Fig 4 is electron micrograph of tsetse endosymbionts in C6/36 cells. Arrow: endosymbiont in vacuoles; N, C6/36 nucleus.

Fig. 5 is electron micrograph of tsetse endosymbiont. Pointers, fimbrae.

Fig. 6 is a dendogram showing the phylogenetic relationship of tsetse endosymbionts and other Gram-negative bacteria based on 16S rRNA sequence.

Fig. 7 is a restriction enzyme analysis of plasmid DNA isolated from tsetse endosymbionts. Lanes 1 & 6, Hindlll-digested lambda marker; Lane 2 , GPOl, Sal-1-digested; Lane 3, GM02, Sal-1-digested; Lane 4, GM02 transformed with pSUP204, Sal-1-digested; Lane 5, pSUP204, Sal-1-digested; Lane 7, GPOl, EcoRl-digested; Lane 8, GM02, EcoRl- digested; Lane 9, GM02 transformed with pSUP204, EcoRl-digested; Lane 10, pSUP204, EcoRl-digested. Arrow: linearized pSUP204, approximately 12kb.

DETAILED DESCRIPTION OF THE INVENTION

The present invention provides for the use of genetically altered symbiotic bacteria of insects as carriers for genes, whose expressed products can modify insect function. Attempts to produce insects carrying foreign genes (transgenic insects) have been made in the past. These attempts have introduced foreign genes via mobile genetic elements or introduced DNA directly. The advantages of symbiotic bacteria as carriers of transgenes are as follows:

(a) Mobile genetic elements insert randomly into somatic cell DNA of insects and most random insertions are harmful mutations. Heretofore there has not been any technology which allows the placement of foreign genes into insect cells, and the productive expression of such genes.

(b) The present inventions provides for the placement of foreign genes into symbiotic bacteria such that genetically altered symbionts (GAS) can be introduced stably into insects and that the products of such foreign genes can be expressed.

The present invention serves to introduce foreign genetic material in a stable fashion into bacteria present in the tissues of arthropods or helminths. This is done via plasmid

shuttle vectors. Transformed symbiotic bacteria have been shown to be capable of spreading laterally into insect populations. Heretofore there have been introduced into insects, genes whose products are selectively toxic to human disease or to animal disease-causing parasites, carried by the insect. The potential capabilities of the present method are, however, much wider. By the present invention, genes may be introduced whose products delay the onset of insecticide- resistance or which, inside the insect, produce toxic metabolites from non-toxic insecticides sprayed into the environment. It is also possible to introduce into symbionts of those arthropods or helminths which transmit bacterial, viral or fungal diseases of plants, particular genes whose products decrease vector competence and/or the transmission of disease to plants. Since genetically altered symbionts are generally species specific, it is possible by the present invention to target individual insect species without destroying other insects or allowing toxins to enter the food chain. A goal of the present invention is selective modification of undesirable insect functions, not at the general elimination of insects. To be successful a scheme to introduce foreign genes into arthropods or helminths should have the following characteristics:

(1) Introduction and expression of the gene should be simple and easy to reproduce. Non-transient expression of genes in eukaryotic cells is still difficult to achieve and is not a routine procedure.

(2) Once the foreign gene has been introduced into the insect, the gene has to be able to spread throughout the insect species population, yet not spread beyond the targeted insect species.

The present invention concerns a method which by-passes these aforementioned difficulties. The foreign gene is not introduced into the tissues of insects, but is introduced into the stable symbiotic bacteria present in insects. Such symbiotic bacteria have the function of providing insects with nutrients not present in their natural food sources. Foreign genes are introduced via bacterial plasmid shuttle vectors. These plasmids contain origins of replication both for, for example, E. coli and the symbiotic bacterium. Plasmid constructs are made and grown in, for example, E. coli. Thus, they use the simple molecular biology procedures adapted to £ coli which have been used for decades. A typical construct will contain one or two selection markers—generally a thiostrepton and/or an ampicillin resistance gene, one appropriate for each bacteria, £ coli and symbiotic bacterium. An arthropod or helminth is made aposymbiotic (its native symbionts are removed with antibiotic) . It is then infected with cultured symbionts which have been transfected with the plasmid construct containing selection markers. Such transformed symbionts will stably colonize aposymbiotic insects for the lifetime of the insect.

The plasmid construct contains sites at which further genes can be introduced. These are genes designed to alter specific insect functions. An embodiment of one system according to the present invention can be aimed at reducing or eliminating the capacity of the assassin bug (Rhodnius prolixus) to carry the human trypanosome parasite (T. crυzi) ; the cause of Chagas disease. The gene for the inactive form of a pore- forming protein, Cecropin A together with a promoter are introduced into the bacterial plasmid in addition to the selection markers. Rhodococcus rhodnii , a symbiotic bacterium of

the assassin bug, is transformed with the plasmid and assassin bugs are infected. Cecropin A is a small peptide which consists of two short alpha helices joined by a hinge-like region. One helix is amphipathic, the other is hydrophobic. The peptide inserts into the membrane of certain cells destroying the integrity of the membrane and killing the cell. Cecropins are natural products of insects secreted into the hemolymph, but are not constitutively expressed in the gut of the insect. Both T. cruzi and symbionts are found in the gut of the insect. Cecropin A has a two order of magnitude therapeutic index; it kills T. cruzi at concentrations of 1/1000 of the concentration at which it kills the symbiont. It is considered that most of the inactive preproCecropin A released into the gut is by symbiont turnover. Symbionts may constitute about 10-15% of the biomass of the insect. The gut of the assassin bug probably contains a dipeptidyl peptidase. Such an enzyme is necessary to convert the inactive preproCecropin A to the active Cecropin A. It is also possible to supply the gene for this enzyme to transformed symbionts.

Foreign genes can be spread into the insect population by several different methods. In the Rhodnius prolixus system, symbionts are transferred from mother to offspring by a transovarial route. The egg is coated with symbionts and on emergence the larva is infected. Coprophagy is another common mode of spread. In the Rhodnius prolixus system nothing needs to be done to spread the transformed symbiont locally The transformed symbiont competes with the untransformed symbiont very favorably and the transformed symbiont will spread through a Rhodnius colony.

Symbionts are highly specific for each insect species. In several instances a different gonadal symbiont (Wolbachia) is actually responsible for cytoplasmic incompatibility; the inability of certain insects to interbreed. When the symbionts are removed, breeding occurs. The genes responsible for cytoplasmic incompatibility can be used as a driving mechanism to spread genetically altered symbionts through insect populations. By inserting origins of replication into plasmids relevant to certain other symbionts the host symbiont range of the plasmid can be extended. There are also certain broad bacterial host-range plasmids such as RSF 1010. Thus the spread of the genetically altered symbionts can be tailored exactly to the insect species which are targets.

The present invention can also be employed for introducing additional genes into symbionts, whose expression would favor insect nutrition or other functions and which would act as a strong selection mechanism.

The following Table I represents a non-limiting list of man and animal parasitic vector-borne diseases.

The following Table II is a non-limiting listing of human animal vector-borne viral diseases for which the present invention can be directed to.

The following Table III is a non-limiting list of arthropod-vector borne viral rice diseases.

The following Table IV is a non-limiting list of vector- borne plant diseases.

Table I

c ro

H c

H m cn 3: m

H

Table II HUMAN AND ANIMAL VECTOR-BORNE VIRAL DISEASES

Animal Disease Vectors Virus Anti-virus Target Symbionts Species Gene

Pig African Tick ASF Virus Single-chain Viral Tick Swine Fever antibody surface Symbiont protein (not characterized) | Cattle Blue Tongue Mosquito BT Virus as above Viral Spiroplasma £" Sheep (Culicoides) Orbivirus surface protein

I

H Man and Venezuelan Mosquito VEE Virus as above Viral Spiroplasma M O Horse Equine (many) Alpha Virus surface I Encephalo- protein myelitis

Man and Eastern Mosquito EEE Virus as above Viral Horse Equine (many) Alpha Virus surface Spiroplasma

X encephalo- protein myelitis

-**-_*» Man Dengue Mosquitoes Dengue Virus as above Viral Spiroplasma Fever (Aedes aegypti) Flavivirus surface protein

Table III Arthropod-vector Borne Viral Rice Disease

Plant Disease Vectors Viruses Anti-virus Target Symbionts Gene

Rice Rice dwarf Nephotettix RDV Single-chain (a) helper All leaf-hoppe

(Oryza sativa) virus cincipeps ds RNA antibody factor symbionts incl, disease N. nigropictis 12 segs. (b) viral a, a ^{1 ■} r and +

N. viriscens surface types

Recidia dorsalis proteins

All leafhoppers (7 proteins) w c Rice gall N. nigropictis RGDV Single-chain (a) helper as above m dwarf virus N. virescens ds RNA antibody factor ω disease N. malyanus 10 segs (b) viral

N. cincipeps surface proteins

I c Rice stripe Laodelphax RSV Single-chain (a) helper as above t

*-. virus disease striatellus ss + ds antibody factor I Unkanodes RNA (b) viral sapporonius 4 + 4 surface x U. albifascia proteins proteins Terthron albovittatus (Plant hoppers) as above

There is a total of 13 insect-borne virus diseases of rice of which 4 representatives are named. Similar target mechanisms apply to each of them.

Table IV

Representative Vector-borne Plant Diseases

Plant Disease Vectors Virus or Anti-parasitic Target Symbionts Microorganism Gene

Cucurbits Cucumber Aphids CM virus Single-chain Viral Un-named

Mosaic antibodies to surface mycoplasma-

Virus to viral like surface organisms m Spotted Tomatoes Thrips SW disease Un-named _t «- wilt virus symbionts disease m Peach Peach Plum leaf- PY virus yellows hoppers Macropsus I c trimaculata Ui I

Grapes Pierce ' s Leafhoppers Bacteria Single-chain Bacterial Disease e.g. antibodies surface Carneocephala to bacterial

IT" draeculacephala surface

The tsetse vectors of African trypanosomiasis contain a rich flora of bacterial endosymbionts, reported from various tissues including midgut epithelium, fatbody, milk glands, and ovaries (Dasch, G.A. , Weiss, E. and Chang, K. , (1984), "Endosymbionts of Insects," p.811-833 In N.R. Krieg, ed. Bergey's Manual of Systematic Bacteriology. Vol. 1. Williams and Wilkins, Baltimore) . Ultrastructural analysis indicates that at least two different symbiont populations may infect the tsetse. A large bacterial symbiont, 5-9 μm in length, has been described from midgut mycetomes (Southwood, T.R.E., Khalaf, S., and Sinden, R.E. (1975), "The Microorganisms of Tsetse Flies", Acta. Trop. 32: 259-266) and a smaller, <2-3 μm symbiont has been described from a number of tissues including midgut epithelium and ovaries (Reinhardt, C. , Steiger, R. and Hecker, H. , (1972), "Ultrastructural Study of the Midgut Mycetome-Bacteroids of the Tsetse Flies Glossina morsitans , G. fuscipes and G. brevipalpis (Diptera, Brachycera)", Acta. Trop. , 29: 280-288) . Both light and electron microscope studies of these latter symbionts in the ovaries, eggs and in nurse and follicle cells reveal that they are pleomorphic, occurring mostly as gram-negative rods that resemble rickettsia. Welburn, S.C., Maudlin, I. and Ellis, D.S.,. (1987), "In vitro Cultivation of Rickettsia-like Organisms from Glossina spp.", Ann. Trop Med Parasitol, 81: 331-335 have cultured a bacterial symbiont from hemolymph of Glossina spp . in insect cell lines. Morphologically, this isolate appears to be similar to earlier described "rickettsia-like organisms" or RLOs (Pell, P.I. and Southern, D.I. (1976), "Effect of the Coccidiostat, Sulphaquinoxaline, on Symbiosis in The tsetse Fly, Glossina species, Microbios Letters. 2 _. : 203-211) . Although the endosymbionts of tsetse have been termed "rickettsia-like" because of their small size, intracellular association with

insect cells and pleomorphic appearance, their taxonomic position is uncertain. For example, it is not clear to what extent these agents are related to the rickettsial pathogens of humans and animals or to the nonpathogenic rickettsia-like Wolbachia symbionts from other arthropods. In fact, their exact position within the eubacteria as a whole is unknown.

The shuttle plasmid vector DNA amplifier is a bacteria or yeasts into which shuttle plasmid vector can introduce transgenes. The shuttle plasmid vector DNA amplifier comprise yeasts and bacteria such as Saccharomyces cerevisiae and genetically defined gram positive and gram negative bacteria such as B. subtilis and E. coli .

The method according to the present invention of genetically altering arthropod or helminth symbionts to express foreign proteins in arthropods and helminths has a wide spectrum of use in agriculture as well as in medicine.

The products of genes expressed in insects can be used as follows:

(a) To produce incompetent parasite vector insects. Gene products such as the pore-forming antibiotics such as Cecropin A 1→13, Mellitin 1→13 hybrids (CMHO or Cecropin A, B or their analogs) , as well as Attacins, Defendins and Magainins which have highly specific anti-parasitic activity can be expressed. Specific single-chain antibodies directed against parasites can be expressed, as well as enzymes such as he e methylase which drive necessary nutrients into non- utilizable pathways. Direct anti-parasite toxins from bacterial and other sources can be expressed.

Resistance to certain insecticides can be mediated by inducible enzymes in the insect which detoxicate the insecticide. Recent advances in immunoglobulin genetics and expression make it relatively simple to introduce and express anti-enzyme or anti-protein antibody genes. Single chain antibody genes which have the light and heavy chain variable regions linked by a spacer on one DNA chain. The antibody product could inhibit enzymic activity and delay the development of insecticide resistance.

(b) Genes can be inserted whose enzyme products could convert non-toxic pro-insecticides into toxic insecticides within the arthropod or helminth. Other products could be expressed in insects which would inactivate enzymes used to detoxify exobiotic substances (insecticides) and thus slow the development of insecticide resistance.

Symbionts

Non-limiting examples of symbionts include Rhodococcus rhodnii, tsetse symbionts , tick symbionts, plant-hopper symbionts, aphid symbionts and spiroplasma (especially spiroplasma of mosquitos) .

Foreign Genes

The foreign gene is foreign to the symbiont or to the strain of Wolbachia pipientis .

The foreign gene for utilization in the present invention can be a protein or peptide which inhibits the survival of a parasite, virus, bacteria or fungus and which decreases the transmission of disease by such microorganism.

Foreign genes for use in the present invention include antibodies to detoxication enzymes. The foreign genes thus can be antibodies to multifunctional esterases or oxidases.

The foreign gene can be an anti-parasitic gene including ATF-2, Cecropin A or Cecropin-mellitin hybrid.

The foreign gene can also be an immunoglobin.

The foreign gene can furthermore code for a polypeptide or protein which neutralizes a viral helper factor. The viral helper factor can be associated with a poty virus, e.g., tobacco vein mottling virus; a caulimovirus; a closterovirus; or a potex virus; or a specific virus such as cauliflower mosaic virus, pea enation virus, anthracis yellows virus or parsnip yellow fleck virus.

The foreign gene can also code for a polypeptide protein which interacts with a viral surface protein, wherein the polypeptide or protein is an antibody or lectin. The viral surface protein is of a virus selected from the group consisting of caulimovirus, commelina yellow mottle virus, geminivirus, cryptovirus, reovirus, rhabdovirus, bunyavirus, carmovirus, luteovirus, maravirus, M.C.D.V. group virus, necrovirus, potexvirus, PYFV group virus (parsnip yellow fleck virus) , sobemovirus, tobamovirus, tombus virus, tymovirus, capillovirus, closterovirus, carlaviros, poty virus, cosmovirus, dianthovirus, fabavirus, nepovirus, P.E.M. virus (pea enation mosaic virus) , furovirus, tobravirus, bromovirus, cucumovirus, harvirus, hordeivirus and tenvivirus.

Antiparasitic Molecules And Their Genes

There is a wide variety of genes whose macromolecular products are potentially inhibitory or toxic to insect-borne parasites. Those include enzymes which deviate essential substances into non-utilizable intermediates. One example is heme methylase which deviates protoporphyrin down the chlorophyll pathway (Blanche, F. , Debussche, L. , Thibaut, D. , Crouzet, J. and Cameron, B. , (1989) , J. Bacteriol. , 171. 4222- 4231; Warren, M.J. , Stolowich, N.J. , Santander, P.J., Roessner, CA. , Sowa, B.A. and Scott, A.I., (1990), "Enzymatic Synthesis of Dihydrosirohydrochlorin (precorrin-2) and of a Novel Pyrrocorphin by Uroporphyrinogen III Methylase", FEBS. Lett. , 261. 76-80) . Others are genes which produce T. cruzi toxins (Mercado, T.I., Butany, J.W. and Ferrans, V.J. , (1986), "Trypanosoma cruzi: Ultra-structural Changes Produced by an Anti- trypanosomal Factor From Pseudomonas fluorescens" , Experimental Parasitol. , j51, 65-75) , or control the production of lectins necessary for parasite adherence (Maudlin, I. (1991), "Transmission of African trypanosomiasis: Interactions among Tsetse Immune Systems, Symbionts and Parasites", p 117-140. In Vol. 7, Advances in Disease Vector Research. Ed. Harris, K.F., Springer-Verlag, NY) .

Pore-forming polypeptides have direct toxicity to T., cruzi, while they have less toxicity to insect symbionts and additionally we know that they can be non-toxic to many insects in usable concentrations. They can be expressed in the insect gut (where they are not normally found) in close proximity to the T. cruzi parasites and there is evidence that they may be processed from the inactive prepro-form to the active peptide in the insect gut. While Cecropin A can be moderately toxic to mosquitoes (Gwadz, R.W. , Kaslow,

D. , Lee, J.Y. , Maloy, W.L., Zasloff, M. and Miller, L.H. , (1989) , "Effects of Magainins and Cecropins on the Sporogenic Development of Malarial Parasites in Mosquitoes", Inf. Immun.. 57. 2628-2633), the bug Rhodnius prolixus appears to be resistant to Cecropin A. Although Cecropin A is a natural insect product, it is usually found in the hemolymph.

The Nature of Pore-forming Polypeptides

The Cecropins are a family of insect-derived, inducible, antibiotic pore-forming basic peptides with a common structural organization. They consist of two small alpha- helical segments joined by a flexible non-helical, non-beta sheet region with beta-turn potential, which may serve as a hinge. The N-terminal alpha helix is _} amphipathic and the C- terminal helix is hydrophobic.

Cecropins contain of the order of 35-40 amino acid residues and are extremely stable, resisting hydrolytic attack from most proteases (Boman, H.G., (1991), "Anti-bacterial Peptides: Key Components Needed in Immunity", Cell. 65. 205- 207) . They are made as inactive prepropeptides which are split to the active mature peptides by a dipeptidyl peptidase enzyme. Originally isolated by Boman from the Hyalophia cecropa moth, related peptides are now also known to--exist in mammals (Hultmark, D. , Steiner, H. , Rasmuson, T. and Boman, H.G., (1980) , "Insect Immunity. Purification and Properties of Three Inducible Bactericidal Proteins from Hemolymph of Immunized Pupae of Hyalophora cecropia" , Europ. J. Biochem.. 106. 7- 16; Lee, J-Y. , Boman, A., Chuanxin, S., Andersson, M. , Jόrnvall, H. , Mutt, V. and Boman, H.G., (1989), "Antibacterial Peptides from Pig Intestine: Isolation of a Mammalian Cecropin", Proc. Nat. Acad. Sci. (USA) . 86, 9159-9162).

There are other classes of pore-forming proteins with antibiotic activity such as the Magainins, Attacins and Defensins, as well as numerous synthetic peptides based on the overall structural plans of Cecropins and related compounds (Kanost, M.R. , Kawooya, J.K., Law, J.H. , Ryan, R.O., Van Heusden, M.C and Ziegler, R. , (1990), "Insect Haemolymph Proteins", Adv. in. Insect Physiol. 22. 299-396). Some of these peptides induce in lipid planar membranes, time-variant and voltage-dependent ion channels. The single-channel conductances of up to 2.5 nS observed, suggest a channel diameter of approximately 4 μm (Christensen, B. , Fink, J., Merrifield, R.B. and Mauzerall, D. (1988), "Channel-forming Properties of Cecropins and Related Model Compounds Incorporated into Planar Lipid Membranes", Proc. Natl. Acad. Sci. (USA) . 85. 5072-5076) . Cecropins release carboxyfluorescein from liposomes containing this dye (Steiner, H. , Andreu, D. and Merrifield, R.B., (1988), "Binding and Action of Cecropin and Cecropin Analogues: Anti¬ bacterial Peptides from Insects", Biochem. Biophvs. Acta. 939. 260-266) . The insertion of the N-terminal helix into different cell membranes is highly dependent on the exact amino acid sequence of the peptide.

Strategies For Spreading Genes Through Insect Populations

Once anti-parasitic genes have been expressed in insects it becomes absolutely necessary to spread these genes throughout the insect population. There exist natural mechanisms in insects for gene-spreading and these mechanisms can be harnessed for the spreading of anti-parasitic genes.

Cytoplasmic Incompatibility-Mediated Reproductive Advantage

Cytoplasmic incompatibility (Cl) is a process in insects in which during fertilization, the sperm does not contribute

its genetic material effectively to the ovum and the zygote does not develop. This occurs when two different, cytoplasmiσally incompatible strains of the same insect species are mated. Cl is associated with the presence of symbionts in the gonads of the insect strain and disappears when the symbiont is removed with antibiotics (Hoffmann, A. , Turelli, A.M. and Simmons, G.M. , (1986), "Unidirectional Incompatibility between populations of Drosophila simulans" , Evolution. 40. 692-701) .

The symbiont responsible for Cl is the non-pathogenic Wolbachia pipientis which is located in gonadal tissues and is inherited via a transovarial route.

Wolbachia mediated C.I. is known to occur in the mosquitoes of the Culex and Aedes spp. (Yen, J.H. and Barr, A.R. , (1973), "The Etiological Agent of Cytoplasmic Incompatibility in Culex pipiens" , J. Invertebr. Pathol.. 22; 242-250; Subbarao, S.K., "Cytoplasmic incompatibility in mosquitoes", In Recent Developments in the Genetics of Insect Disease Vectors. Ed. Steiner, W.W.M. , Tabachnick, K.R. and Narang, S., (1982), pp 313-342, Stipe Publishing Co., Champaign, Illinois), the parasitic wasps Nasonia (Breeuwer, J.A.J. and Werren, J.H. , (1990) , "Microorganisms Associated with Chromosome Destruction and Reproductive Isolation between Two Insect Species", Nature, 346, 558-564); Drosophila simulans (Hoffmann, A.A. and Turelli, M. , (1988), "Unidirectional Incompatibility in Drosophila simulans : Inheritance, Geographic Variation and Fitness Effects", Genetics. 119. 435-444); the beetle Triboiium confusum (Wade, M.J., Stevens, L. , (1989), "Microorganism-mediated Reproductive Isolation in Flour Beetles (Genus Triboiium)", Science. 227, 527-528; O'Neill, S.L. (1989), "Cytoplasmic Symbionts in Triboiium confusum"; J. Invertebr. Pathol. 53. 132-134); the moths Ephestia cautella , E. figulilella (Kellen, W.R. , Hoffman, D.F.

and Kwock, R.A. , (1981), J. Invertebr. Pathol. 37. 273-283) and the beetle Hypera postica (Hsiao, C and Hsiao, T.H. , (1985), J. Invertebr. Pathol. , 45, 244-246) .

It seems likely that C.I. occurs in many arthropods carrying the Wolbachia symbiont (O'Neill, S.L., Giordano, R. , Colbert, A.M.E., Karr, T.L. and Robertson, H.M. , (1992), "16S rRNA Phylogenetic Analysis of the Bacterial Endosymbionts Associated with Cytoplasmic Incompatibility In Insects", Proc. Natl. Acad. Sci. USA. (1992), 89, 2699-2702. An analysis of 16S rRNA sequences of Wolbachia from several arthropod species has shown that all Wolbachia from insects should be classified as a single species; Wolbachia pipientis (O'Neill, S.L., Giordano, R. , Colbert, A.M.E., Karr, T.L. and Robertson, H.M. , (1992), "16S rRNA Phylogenetic Analysis of the Bacterial Endosymbionts Associated with Cytoplasmic Incompatibility In Insects", Proc. Natl. Acad. Sci. USA. (1992), 89, 2699-2702). Further work has shown that W. pipientis from an infected arthropod species, can be transferred by micro-injection into uninfected insect species. The Wolbachia infection once established in the previously uninfected species, is stable and confers onto the new insect host the properties of C.I. Transfer of Wolbachia is accomplished by collecting pre-syncitial blastoderm stage embryos and by removing from the posterior region a small amount of cytoplasm with a fine glass needle on a micro- injector and injecting it into host arthropod embryos at a similar developmental stage.

The C.I. Gene

It has been observed that the presence of C.I. correlates consistently with the presence of a single low molecular weight protein in the sperm from the Wolbachia infected male

Drosophila simulans (Chang, L. , O'Neill, S. and Karr, T., (1991), "Cytoplasmic Incompatibility is Correlated with the Expression of a Single Protein in the Sperm of Incompatible Males", No. 23.178, p 58 in 32nd Annual Drosophila Research Conference Program and Abstract, Volume 1991, Chicago, IL.), in the wasps Nasonia vitripennis and N. giraulti , paternal chromosomes present in the "fertilized" egg form tangled masses and are eventually eliminated Reference (Breewer et al, (1990), Nature. 558-564).

Undirectional C.I. Is a Gene-Spreading Mechanism

Since crosses of uninfected females and infected males are "sterile", uninfected females are at a reproductive disadvantage when compared to infected females, who can successfully mate with either infected or uninfected males. This reproductive disadvantage of uninfected females ensures the rapid spread of the maternally inherited symbiont into the insect population. Thus a symbiont Cl. infection is expected to spread from the point of infection and will tend to eliminate polymorphism in the insect population. Turelli, M. and Hoffmann, A.A. , (1991), "Rapid Spread of an Inherited Incompatibility Factor in California Drosophila , Nature, 353. 440-442 have monitored Californian populations of Drosophila simulans . The R type of D. simulans is a Wolbachia infected strain, believed to have entered the Central Valley of California by invasion from the west through the coast range. Repeated monitoring of D. simulans populations shows that the R strain has spread at a rate of more than 100 km per year; in three years the R strain has become common in locations where it was once absent.

C.I. Can Be Harnessed to Spread Foreign Genes

When an insect strain becomes infected with a Wolbachia symbiont which mediates C.I., mitochondrial DNA markers already present in the newly infected insect and detected by restriction fragment length polymorphism, are maternally inherited in a constant manner, together with the Wolbachia symbiont. The constant association of a mtDNA marker with a particular symbiont can be adequately explained by the C.I. effect alone without invoking mtDNA marker selection (Nigro, L. and Prout, T., (1990), "Is there Selection on RFLP (Restriction Fragment Length Polymorphism) Differences in Mitochondrial DNA?", Genetics, 125_, 551-553).

The constant association of a mitochondrial DNA marker with transmission of a particular symbiont is seen even in insects which have two separate Cl systems (Baba-Aϊssa, F., Solignac, M. , Dennebouy, N. and David, J.R., (1988), "Mitochondrial DNA Variability in Drosophila simulans : Quasi Absence of Polymorphism Within Each of the Three Cytoplasmic Races", Heredity 61. 419-426; Montchamp-Moreau, C, Ferveur, J-F. and Jacques, M. , (1991), "Geographic Distribution and Inheritance of Three Cytoplasmic Incompatibility Types in Drosophila simulans" , Genetics, 129. 399-407) . These observations strongly suggest that if an anti-parasitic gene is introduced in a stable fashion into an insect via one type of maternally inherited symbiont, the addition of C.I. via a second maternally inherited Wolbachia symbiont not already present in the insect population, will drive the anti-parasitic gene throughout the insect population (Turelli, M. and Hoffmann, A.A. , (1991), "Rapid Spread of an Inherited Incompatibility factor in California Drosophila" , Nature 353, 440-442).

Rhodnius prolixus

Rhodococcus rhodnii - Trypanosoma cruzi system R. rhodnii is an actinomycete which lives extracellularly within the lumen of the bug gut in close apposition to T. cruzi. The mass of this symbiont is large; after a blood meal approximately 10-15% of the body weight may be due to the symbiont. R. rhodnii is known to turn over rapidly after the blood meal, presumably liberating our recombinant gene products. Both Cecropin A and CMH peptides are highly resistant to proteolytic digestion (Boman, H.G., Boman, A., Andreo, D. , Li, Z-Q, Merrifield, R.B., Schlenstedt, G. and Zimmermann, R. , (1989), "Chemical Synthesis and Enzy ic Processing of Precursor Forms of Cecropin A and B", J. Biol. Chem.. 264. 5582-5586; Kreil, G. , Maiml, L. and Suchanek, G. , (1980), "Stepwise Cleavage of the Pro Part of Promelittin by Dipeptidylpeptidase IV", Eur. J. Bioch .. Ill, 49-58.), yet active peptides are derived from their inactive precursors by dipeptidyl peptidase activity (Boman, H.G., Boman, A., Andreo, D. , Li, Z-Q, Merrifield, R.B., Schlenstedt, G. and Zimmermann, R. , (1989), "Chemical Synthesis and Enzymic Processing of Precursor Forms of cecropin A and B", J. Biol. Chem.. 264. 5582-5586; Kreil, G. , Maiml, L. and Suchanek, G. , (1980), "Stepwise Cleavage of the Pro Part of Promelittin by Dipeptidylpeptidase IV, Eur. J. Biochem. , 111, 49-58) and this enzyme activity is present in the gut of the bug.

Chagas' Disease

Chagas' disease is widespread in South and Central America. It has been estimated that 70 million people have had a self-limiting primary infection with T. cruzi, of which 10% develop the secondary disease after an interval of many years. The secondary disease presents as cardiac arhythmias,

cardiomegaly and may have the "mega disease" symptoms consisting of enlargement and inco-ordination of hollow viscuses. The underlying lesion is the degeneration of autonomic nervous tissue in Auerbach's and Meissner's plexi and of the cardiac conducting tissues. There is no known cure for Chagas disease.

Non-Toxic Precursors of Insecticides

Some insecticides have non-toxic precursors which can be made into toxic metabolites by the action of an enzyme. Such an enzyme can be introduced into the symbionts of a harmful arthropod or helminth species. Since symbionts and vectors can be made highly specific for a single species of an arthropod or a helminth, there would be no danger of transmission of the enzyme to non-harmful insects. A possible non-toxic precursor could be a derivative of a commonly used organophosphorus insecticide. This derivative is non-toxic since a protecting group prevents interaction with the esteratic site of choline esterase. When an enzyme is expressed in the insect which cleaves off the protecting group, the compound becomes an irreversible choline esterase inhibitor inside the insect. Thus generally useful insecticides can be developed which will affect only a single or several harmful species which carry the enzyme. Since enzymes are destroyed when ingested, there would be no toxic residues in the food chain from ingestion of live insects.

The invention will now be described with reference to the following non-limiting examples.

EXAMPLES Example 1: Construction of the Shuttle Plasmid

A shuttle-plasmid was constructed which replicates stably in R. rhodnii as well as in £ coli.

Plasmid DNA from R. rhodnii (strain 35071; American Type Culture Collection [ATCC] , Rockville, MD) was isolated by the method of Vogt Singer and Finnerty (Vogt Singer, M.E., Finnerty, W.R. , (1988), "Construction of an Escherchia coli- Rhodococcus Shuttle Vector and Plasmid Transformation in Rhodococcus spp.", J. Bacteriol; 170:638-645) with the following modifications. Following lysozyme incubation, the bacteria were centrifuged gently, resuspended in GTE solution (50 mM glucose, 25 mM Tris and 10 mM EDTA pH 8.0) containing 50 μg/ml of proteinase K, and incubated at 55°C for 1.5 hours. The bacteria then were lysed with NaOH and SDS for 30 minutes at 55"C Genomic DNA and cellular debris were precipitated with a solution that was 3 M with respect to potassium and 5 M with respect to acetate (pH 4.8), and plasmid DNA was recovered following phenol-chloroform extraction, and precipitation with isopropanol.

The resulting plasmid DNA was digested individually with the restriction enzymes Sma I, Pvu II, Hind II to give approximately 12 fragments collectively, ranging in size from 400 to 3,500 basepairs. Phosphorylated Eco RI linkers (New England Biolabs, Beverly, MA) were added to the blunt-end fragments, which were then ligated into the Eco RI site of pIJ30 (Thompson, C.I., Kieser, T. , Ward, J.M. , Hopwood, D.A. , (1982) , "Physical Analysis of Antibiotic-Resistance Genes from Streptomyces and their use in vector construction", Gene 30:61-62) transfected into £ coli strain DH5αmcr (Bethesda Research

Laboratories Life Technologies, Gaithersburg, MD) using the standard product protocol and selected on LB plates containing 100 μg/ml of ampicillin. Plasmid DNA was prepared from the resulting library of ampicillin-resistant transformants (Mierendorf, R.C, Pfeiffer, D. (1987) "Direct Sequencing of Denatured Plasmid DNA," Methods Enzvmol. 152:556-562) and used to transform protoplasts of R. rhodnii (Okanishi, M. , Suzuki, K. , Umezawa, H. , (1974), "Formation and Reversion of Streptomyces Protoplasts: Cultural Condition and Morphological Study," J. Gen Microbiol. 8^:389-400) with selection of R2YE plates (Thompson C.I., Ward, J. M. , Hopwood, D.A. , (1980)" DNA Cloning in Streptomyces ; Resistance Genes from Antibiotic- Producing Species, "Nature. 286:525-527) containing 50 μg/ml of thiostrepton. The shuttle plasmid was then placed back into £ coli , re-isolated, and mapped by restriction enzyme digestion.

Analysis of plasmid DNA from R. rhodnii strain ATCC 35071 revealed an endogenous plasmid of approximately 6.5 kb. Construction of the shuttle plasmid was achieved by means of the strategy outlined in Fig. 1. A shuttle plasmid was isolated, and its ability to replicate in both bacterial systems was confirmed. The plasmid, pRrl.l (Fig. 2), contains replication origins from both £ coliand R. rhodnii and the ampicillin and thiostrepton resistance genes from pIJ30. In £ coli, the recombinant plasmid uses the pUC19 origin for replication and the bla gene product for selection, amp ^R . The plasmid confers ampicillin resistance to £ coli and thiostrepton resistance to R. rhodnii. It also contains seven unique restriction enzyme sites that are potentially valuable for inserting additional genes.

In order to isolate an origin that would allow replication in R. rhodnii, random restriction enzyme fragments of a purified endogenous R. rhodnii plasmid were subcloned into the pUC-Thio construct pRrl.l and one of these was found to replicate efficiently in R. rhodnii. The 3 kb insert was further subcloned and the R. rhodnii origin was localized to a 1 kb DNA fragment.

Example 2: Transfection of the Symbiont

In order to generate transformation competent protoplasts, mid-exponential R. rhodnii cells were grown for 2 hours in the presence of ampicillin; cells were harvested, washed, and incubated for 2 hours with hypertonic protoplast buffer containing lysozyme (10 mg/ml) . Cells were then spun down, resuspended in the rotoplast buffer, transfected with plasmid DNAs and transformants were selected on R2YE plates containing 50 μg/ml thiostrepton. Cells were propagated for 40 generations in the absence and presence of the antibiotic and were plated onto thiostrepton containing plates. When the number of thiostrepton selected and non-selected colonies formed were compared, no significant difference was observed indicating that the construct is stable in R. rhodnii even in the absence of the selectable antibiotic marker.

Example 3: Development. Infection, and Selection of Aposymbiotic Bugs

Aposymbiotic Rhodnius prolixus bug colonies were established by surface sterilizing freshly laid eggs in a povidone-iodine solution (Betadine: Purdue Fredrick Co., Norwalk, CT) , and transferring the eggs into sterile glass medium bottles for rearing at 38"C The emerging aposymbiotic insects were infected with the R. rhodnii strain RP003, which had been

transformed with the functional thiostrepton resistance gene- carrying shuttle plasmid (pUCRrl.l) described above in Example 1 by brief immersion of the bugs in broth cultures of modified symbionts, or by feeding them through "PAR _A FILM" ⁽ American Can Co., Greenwich, CT) membranes on suspensions of symbionts in defibrinated rabbit blood. Selection of transformed ft rhodnii was achieved by feeding the bugs monthly on defibrinated rabbit blood that contained 50 μg/ml of thiostrepton.

The results of the experimental infection assays are shown in Table V.

Table V

Summary of Experimental Infection Studies with Transformed Rhodococcus rhodnii in aposymbiotic Rhodnius prolixus

Fifty micrograms per milliliter delivered in a bloodmeal , once a month t Negative bugs had sterile guts t Not tested

SUBSTITUTE SHEET

Thirty days following infection, during which the infected bugs were given one bloodmeal containing thiostrepton and underwent 1 molt, transformed symbionts were cultured, and the shuttle plasmid was reisolated from two of five bugs that had been infected by immersion. After 2.5 months, following two bloodmeals and two successive molts, transformed symbionts were recovered from the guts of nine of 15 bugs in this group. The nine bugs that did not harbor transformed symbionts (three at 30 days and six at 2.5 months) had sterile guts, indicating that they had probably never been infected. After 6.5 months and four additional bloodmeals, eight of the remaining nine bugs harbored transformed symbionts. In the bugs that had been orally infected, after 6.5 months and six bloodmeals, 13 of 13 bugs were positive for transformed symbionts.

Example 4: Assessment of Infected Bugs

At various times following infection, individual bugs were removed and surface-sterilized in 70% ethanol and in 1.3% sodium hypochlorite (Clorox bleach, 1:4 dilution; The Clorox Co., Oakland, CA) (10 min/solution) . The bugs were then pierced using 26 gauge needles, and the gut contents were aspirated and added to 100 μl of brain heart infusion (BHI) broth. This material was then spread on the surface of BHI plates with and without thiostrepton, incubated at 28°C and observed for growth.

Example 5: Assessment of Plasmid Stability

The in vitro stability of the shuttle plasmid was determined for three different strains of ft rhodnii ATCC 35071, RP002B (isolated from a ft prolixus colony obtained from Dr. Alberto Morales of the National Institute of Health, Bogota,

Colombia) , and RP003 (isolated from a ft prolixus colony obtained from Dr. Antonio D'Allesandro, Tulane University, New Orleans, LA) . Single, transformed colonies of each bacterial strain were picked from BHI agar plates that contained 50 μg/ml of thiostrepton. The colonies were inoculated into BHI broth without antibiotics and maintained in exponential growth by passage at regular intervals. After the culture had been maintained for 40 generations (generation time being determined as approximately 12 hours on the basis of optical densities at 600 nm) , bacterial suspensions were spread on BHI agar plates and incubated at 28°C Replicas of the resulting colonies were made on BHI plates with and without 50 μg/ml of thiostrepton using accutran replica platers (Schleicher & Schuell, Keone, NH) , and the stability was calculated for each of the transformed strains using the mean of four replicates.

In vivo stability was examined by infecting aposymbiotic bugs with transformed symbionts (strain RP003) as previously described, and feeding them defibrinated rabbit blood that contained no antibiotics. After 6.5 months and six bloodmeals, the bugs were surface-sterilized and 50 μl of the gut contents were streaked on BHI plates with and without 50 μg/ml of thiostrepton. The plates were incubated at 28°C and observed for growth, which was graded from + to ++++.

Sixteen of 16 bugs were positive for transformed symbionts. In 10 of these 16, there was no apparent loss of the plasmid, as indicated by the ratio of thiostrepton- sensitive to thiostrepton-resistant colonies (++++ growth on plates with and without thiostrepton) . Two of 16 showed moderate loss (++ or +++ on thiostrepton plates compared with

++++ on BHI control plates) , and the remaining four showed substantial loss (+ compared with ++++ on control plates) .

The m vjvQ stability of the shuttle plasmid in three strains of ft Rhodnii was determined by maintaining the cultures for 40 generations (20 days) in log phase growth in the absence of antibiotic selection (Table VI) .

Table VI

In vitro stability of pRrl.1 in three strains of Rhodo¬ coccus rhodnii over 40 generations in the absence of thiostrepton selection*

No.(%) of colonies growing Mean % plasmid Strain on Thio/BHI plates loss/generation

RP003 1 42/59 (71.2) 2 30/48 (62.5) 0.82 3 52/68 (76.5)

208/268 (77.6)

ATCC35071 1 80/106(75.5) 2 76/109 (69.7) 0.92 3 51/90 (56.7) 4 58/80 (72.5)

RP002B 1 13/44 (29.5) 2 48/220(21.8) 3.6 3 7/35 (20.0) 4 12/55 (21.8)

*Thio denotes the number of colonies growing on brain heart infusion (BHI) plates with 50 μg/ml of thiostrepton. BHI denotes the number of colonies growing on plates without thiostrepton. This mean number of colonies (4 replicates each) was 72 for strain RP003, 69 for strain ATCC 35071, and 23 for strain RP002B.

SUBSTITUTE SHEET

The highest stability, 0.82% loss per generation, was observed in strain RP003. Stabilities of 0.92% and 3.6% loss per generation were seen in strains ATCC 35071 and RP002B, respectively.

Discussion of the Results for Examples 1-5

Examples 1-5 show that a genetically altered symbiont can be introduced into its aposymbiotic host and can persist over successive molts to adulthood. The genetic alterations introduced consist of antibiotic resistance genes; however, the shuttle plasmid retains the potential for incorporating additional genes.

The advantages of using bacterial symbionts as vehicles for expressing foreign genes in medically important arthropods are the effectiveness, ease and reproducibility of the system. The important specific parameters are the transformation efficiency, infection rates in bugs and plasmid stability. The infection rate in bugs infected by immersion was 40% (two of five) at 30 days following infection. 60% (nine of 15) at 2.5 months, and 89% (eight of nine) at 6.5 months. The infection rate in orally infected bugs assayed at 6.3 months was 100% (13 of 13) . The apparent increase in infection rate over time in the group of bugs that was infected by immersion suggests the occurrence of horizontal transmission of transformed symbionts between bugs. This phenomenon is to be expected on the basis of the biology of the symbiont-host relationship, and could be of significant usefulness in establishing modified symbionts within a population of bugs. The stability of the shuttle plasmid in at least one strain (RP003) is high both in vitro and in vivo , suggesting that maintenance of transformed

symbionts in insects may be possible for extended periods of time, with little or no selection.

Bacterial symbionts have been isolated from several species of medically important arthropods and observed in many others. (Welburn, S.C, Maudlin, I., Ellis, D.S., (1987), 'In vitro Cultivation of rickettsia-like Organisms from Glossina spp.", Ann. Trop, Med. Parasitrol., .81:331-335; Beard, C.B., Butler, J.F., Hall, D.W. , (1990), "Prevalence and Biology of Endosymbionts of Fleas (Siphonaptera; Pulicidae) from dogs and cats in Atachua County, Florida. J. Med. Entomol, 27: 1050-1061) .

The Rhodococcus symbionts of ft prolixus (and perhaps of other triatomines) are particularly useful as potential vehicles for expressing foreign genes in these insects, since the symbionts are required by the host for development and occur in large numbers within the gut of the bug in close proximity to T. cruzi. These symbionts can easily be maintained in culture, transformed, and established in aposymbiotic bugs. Furthermore, it is simple to generate large numbers of transformants and test a variety of different genes and constructs, thus providing a powerful potential tool for modifying vector competence.

Example 6: The Genetically Altered Symbiont (GAS) Is Stable In Its Host And Fulfills Normal Symbiont Functions

Initial selection of GAS was by feeding the infected bugs twice at monthly intervals, on defibrinated rabbit blood containing 50 μg/ml thiostrepton. Symbionts from the gut of ft prolixus were grown on BHI and BHI plus thiostrepton plates. At 2 1/2 months, 9 of the 15 bugs contained the genetically altered symbiont (GAS) ; the remaining six bugs showed no

growth of either transformed or untransformed symbionts; suggesting that in these six bugs no infection had occurred. At six months, the nine remaining imagos (adult insects) , still carried the GAS. This preliminary study showed that the GAS were able to establish themselves in the host and that the GAS were able to fulfill symbiont physiological functions required for repetitive moults and for formation of the imago. This was a small-scale study and will have to be repeated with symbionts infected with the shuttle vector carrying the anti- parasitic gene. The initial results are very encouraging.

Example 7: Competition Between Transformed And Untransformed Symbionts When Both Are Introduced Into The Gut Of Rhodnius Prolixus

When the bug emerges from its egg, it is transiently aposymbiotic. The bug picks up its initial symbiont infection by coprophagy. It was investigated whether a genetically altered symbiont would compete efficiently with a native symbiont. Newly emerged symbiont bugs were exposed to the GAS by coprophagous infection and subsequently fed native symbionts in the blood meal. We also did the reverse procedure. The results are given in Table VIII.

TABLE VIII- Competition Between Growth Of Genetically Altered Symbionts And Native Symbionts In Rhodnius Prolixus Gut

COLUMN 1

% of R prolixus guts harboring symbionts at 6 months

Primary secondary fecal infection Genetically Genetically infection (symbionts altered Native altered and after fed in symbiont symbiont native symbionts emergence blood mean rmixed infectionsϊ

GROUP A genetically native 100 19 altered symbionts 19 symbionts

B genetically none 100 altered symbionts native genetically 44 100 symbionts 44 altered native none 100

The following conclusions were drawn: a) In short-term laboratory conditions the GAS competes on approximately equal terms with the native symbiont. b) The initial coprophagous infection seems to predominate both for GAS and for the native symbiont infections. The secondary (blood meal) infection, however, will cause mixed infections. If anything, the GAS seems to be a little more competitive, since in 44% it replaces the primary infection with native symbionts. Replacement in the reverse experiment is 19%. These initial results suggest that

SUBSTITUTE SHEET

the GAS does not seem to carry features which cause short-term negative genetic selection.

Example 8: Cultures of Endosymbionts were established from both Tsetse Species

Non-motile rod-shaped bacteria were observed in C6/36 cells on day 3 following inoculation with hemolymph extracted from tsetse. Giemsa-stained preparations of cultures from both Glossina pallidipes (designated GPOl) and from G. morsitans (designated GM02) revealed intra- and extracellular bacteria (Fig. 3) which subsequently stained Gram-negative. GPOl grew more slowly, causing cytopathic effects (swelling, vacuolation and cellular detachment) in C6/36 cells in 4 to 5 days, while GM02 produced the same effect in 3 to 4 days. These microorganisms also grew in cell-free medium. Inoculation of noninfected C6/36 cell cultures with symbionts derived from cell-free cultures also gave rise to intracellular infections, indicating that GPOl and GM02 were capable of both intra- and extracellular growth and were not mixed cultures of intra- and extracellular organisms. Repeated attempts to grow these microbes on various types of agar plates under aerobic, anaerobic, or microaerophilic conditions failed.

Electron microscopy of the cultures revealed the presence of rod-shaped bacteria in cytoplasmic vacuoles (Fig. 4) . Infected

C6/36 cells were swollen and heavily vacuolated, and contained numerous bacteria. Different forms of the bacteria were present with regards to size and electron density, reminiscent of the pleomorphism seen in tsetse symbionts in vivo (Pell, P.I. and Southern, D.I., (1976), "Effect of the Coccidiostat, Sulphaquinoxaline, on Symbiosis in the Tsetse Fly, Glossina species", Microbios Letters. 2 : 203-211) and in vitro (Welburn, S.C, Maudlin, I. and Ellis, D.S., (1987), "In vitro Cultivation of rickettsia-like Organisms from Glossina spp.", Ann. Trop. Med. Parasitol. 81: 331-335) . High magnification of bacterial cell walls revealed what appear to be fimbriae (Fig. 5) . Similar structures have been reported in tsetse symbionts colonizing the milk glands and are thought to facilitate attachment and penetration of new cells (Ma, W.C and Denlinger, D.L., (1974), "Secretory Discharge and Microflora of Milk Gland in Tsetse Flies," Nature, 247: 301-303).

In order to confirm the association of GPOl and GM02 with tsetse tissue in vivo , bacterial DNA fragments were cloned into the E. coli plasmid vector pUClδ and these constructs were used as hybridization probes against restricted tsetse DNA. The probes recognized DNA fragments of the same size in isolated bacterial strains and in tsetse DNA, indicating that the GPOl and GM02 are present in the fly and are not culture contaminants.

Example 9: GPOl and GMQ2 are Members of the Gamma Subdivision of the Proteobacteria

The sequence of the 16S rRNA gene is useful for establishing phylogenetic relationships among microorganisms Woese, C R. , (1987), "Bacterial Evolution", Microbiol. Rev. 51: 221-271. Comparison of the 16S rRNA sequence from GPOl and GM02 shows that the two microorganisms are very closely related, displaying only one base substitution over the 1117 bp sequenced from GM02. Such a low level of divergence suggests that these isolates almost certainly belong to the same genus. Whether or not they should be considered conspecific will require additional data. From Table IX and Fig. 6, it can be seen that the tsetse endosymbionts are members of the phylogenetic grouping commonly known as the enteric bacteria, a subgroup within the gamma subdivision of the Proteobacteria (Stackebrandt, E., Murray, R.G.E. and Truper, H.G., (1988), "The Proteobacteria Class is Now a Home for the Phylogenetic Taxon that includes the Purple Bacteria and Their

<*.

Relatives", Int. J. Syst. Bacteriol.. 38: 321-325; Fox, G.E. , Stackebrandt, E., Hespell, R.B., Gibson, J., Maniloff, J. , Dyer, T.A., Wolfe, R.S. Balch, W.E., Tanner, R. , Magrum, L. , Zablen, L.B., Blakemore, R. Gupta, R. , Bonen, L. , Lewis, B.J. , Stahl, D.A. , Luehrsen, K.R., Chen, K.N. and Woese, C.R. (1980), "The Phylogeny of Prokaryotes", Science. 209: 457-463; Woese, C.R. , (1987,) "Bacterial Evolution", Microbiol. Rev. 51: 221- 271).

Table IX

Within that subgroup, however, GPOl and GM02 do not appear to _b e specifically related to any organism for which 16S rRNA sequences are known, including in addition to those organisms shown in Table VII, representatives of the genera Salmonella, Kiebsieiia, Yersinia, Hafnia , as well as Arsenophonus nansoniae the organism responsible for the "son killer" trait (Werren, J.H. , Skinner, S.W. and Huger, A.M. (1986), "Male-killing Bacteria in a Parasitic Wasp." _S cience. ₂₃ 1: 990-992; Gherna, R.L., Werren, J.H., Weisburh, W. , _C olte, R. , _W oese, C.R. , Mandelco, L. , and Brenner, D.J., (1991), •Arsenophonus nansoniae , New genus New Species, The Causative Agent of the Son-killer Trait in the parasitic Wasp Nansonia vitripennis', Int. J. _S vs. _B acteriol, 4_1: 563-565) in wasps and the endosymbionts of aphids ₍ Munson, M.A. , Bau ann, P., Clark, M.A. , Baumann, L. ,

SUBSTITUTE SHEl ^~~ γ

Moran, N.A. , Voegtlin, D.J. and Campbell, B.C., (1991), "Evidence for the Establishment of Aphid-eubacterium Endosymbiosis in an Ancestor of Four Aphid Families", J. Bacteriol. 173: 6321-6324). At present, free-living specific relatives of GPOl and GM02, if they exist, are not known.

It is likely that GPOl and GM02 are similar to the organisms that were cultured by Welburn, S-C, Maudlin, I. and Ellis, D.S. (1987), "In vitro Cultivation of rickettsia-like organisms from Glossina spp.", Ann. Trop. Med. Parasitol. 81: 331-335 from G. pallidipes and G. morsitans . Presumably the symbionts they isolated from other tsetse species are also members of this same group of eubacteria; however, this awaits confirmation. It should also be noted that GPOl and GM02, which qualify, morphologically, as "rickettsia-like organisms", bear no relationship to the rickettsias and various other bacteria capable of invading eukaryotic cells (Weisurg, W.G., Dobson, M.E., Samuel, J.E., Dasch, G.A. , Mallavia, L.P. , Baca, O. , Mandelco, L. , Sechrest, J.E., Weiss, E. and Woese, C.R. (1989) "Phylogenetic diversity of the Rickettsiae", J. Bacteriol.. 171: 4202-4206).

Members of the genus Rickettsia belong to the alpha subdivision of the Proteobacteria, while those members of the family Rickettsiaceae that do belong to the gamma subdivision

branch more deeply than the outgroups (Aeromonas hydrophila and Oceanospirillum linum) shown in Fig. 6 (Weisurg, W.G., Dobson, M.E., Samuel, J.E., Dasch, G.A. , Mallavia, L.P. , Baca, O. , Mandelco, L. , Sechrest, J.E., Weiss, E. and Woese, C.R. , (1989), "Phylogenetic Diversity of the Rickettsiae", J. Bacteriol.. 171: 4202-4206) .

Both GPOl and GM02 harbor large endogenous plasmids. Both endosymbionts have extrachromosomally replicating supercoiled plasmid sequences (Fig. 6) . Based on hybridization intensity of the restricted total plasmid DNAs, there appear to be at least two different plasmids maintained in different copy number.

It can be seen from Fig. 7 that the restriction enzyme patterns of plasmid DNAs from GPOl and GM02 while very similar, have distinct differences. Based on pulsed field gel electrophoresis analysis of EcoRl generated fragments, the large plasmid contains at least one 80 kb EcoRl fragment. If all of the observed DNA fragments of the same intensity are generated from one plasmid, it could potentially be as large as 135 kb.

In Rhodococcus equi strains containing such large fragments, there is evidence suggesting that the plasmids are essential for encoding virulence factors (Takai, S., Sekizake, T. , Ozawa, T. ,

Sugawara, T. , Watanabe, Y. and Tsubaki, S., (1991), "Association Between a Large Plasmid and 15- to 17-kilodalton Antigens in Virulent Rhodococcus equi", Infect. Immun.. 59: 4056-4060).

In Mycobacteriυm scrofulaceum , large plasmids encode for heavy metal resistance, mercuric reductase (Meissner, P.S. and Falkinhan III, J.O., (1984), "Plasmid-encoded Mercuric Reductase in Mycobacterium scrofulaceum", J.Bacteriol. 157: 669-672) and copper sulfide production (Erardi, F.X., Failla, M.L. and Falkimham, J.O., III, (1987), "Plasmid encoded copper resistance and precipitation by Mycobacterium scrofulaceum" , Appl. Environ. Microbiol. 53.8: 1951-1954). The 140 kb plasmid associated with epithelial cell invasion of Shigella has been shown to code for proteins involved in the invasion determinants (Maurelli, A.T. and Sansonette, P.J. , (1988) , "Genetic Determinants of Shigella Pathogenicity", Annu. Rev. Microbiol.. 42 _. :127-150) . The function of these large plasmids in tsetse symbionts remains to be seen.

GM02 cultures were transformed with a heterologous plasmid. The shuttle plasmid, pSUP204, was introduced successfully into GM02 following calcium chloride treatment. The plasmid pSUP204 is a derivative of the IncQ plasmid RSFIOIO, isolated originally from Pseυdomonas aeroginosa (Priefer, U.B., Reinhard, S. and Phler, A. (1985), "Extension of the Host Range of Escherichia coli Vectors by

Incorporation of RSF 1010 Replication and Mobilization Functions", J. Bacteriol. 163: 324-330), and the E. coli cloning vector, pBR325. It contains the broad host range replication origin, orrV, the E. coli pMBl origin, genes that code for resistance to the antibiotics ampicillin, tetracycline and chloramphenicol, and unique restriction sites that can be used potentially for inserting additional DNA sequences.

Transformed cultures expressed resistance to ampicillin, tetracycline HC£ and chloramphenicol up to 200 μg/ml, 5 μg/ml and 10 μg/ml, respectively while the growth of non-transformed cells was inhibited completely at these antibiotic concentrations. The plasmid replicated extrachromosomally in high copy number, as shown in Fig. 7. In addition to the endogenous plasmid fragments, a 12 kb band that represents linearized pSUP204 can be seen in transformed cells (Lanes 4 and 9) that is not present in non-transformed cultures (Lane 3 and 8) .

It is interesting to note that repeated attempts to transform GPOl using variations of the general protocol were unsuccessful. This could be due to incompatibility between the exogenous plasmid pSUP204 and the naturally occurring plasmid(s) of GPOl which differ by restriction analysis to those in GM02 (Fig. 7).

In order to test the stability of the exogenous plasmid, pSUP204, transformed GM02 cultures were propagated in the absence of ampicillin selection. After 4 weeks, the cultures remained transformed as shown by direct plasmid DNA analysis and antibiotic resistance phenotype. These results suggest that pSUP204 is stable in GM02.

Tsetse symbionts and the susceptibility of flies to trypanosome infection. Maudlin, I. and Ellis, D.S., (1985), "Association Between Intracellular Rickettsial-like Infections of Midgut Cells and Susceptibility to Trypanosome Infection in Glossina spp." , Z. Parasitenkd. 71: 683-687 and Maudlin, I., Welburn, S.C. and Mehlitz, D. , (1990), "The relationship Between Rickettsia-1ike-organisms and Trypanosome Infections in Natural Populations of Tsetse in Liberia", Trop. Med. Parasitol. 41: 265-267 have reported an association between bacterial endosymbionts and susceptibility of trypanosome infection in teneral flies.

The mechanism proposed involves the production by symbionts of chitinase which degrades chitin in the larval foregut to glucosamine. These amino sugar residues subsequently bind lectins that would otherwise agglutinate trypanosomes present in the tsetse's first bloodmeal (Maudlin, I. and Welburn, S.C, (1988),

"Tsetse Immunity and the Transmission of Trypanosomiasis", Parasitol. Today. 4 _. : 109-111) . The importance of this mechanism has been questioned by Moloo, S.K. and Shaw, M.K. (1989) , "Rickettsial Infections of Midgut Cells are Not Associated with Susceptibility of Glossina morsitans centraiis to Trypanosoma congolense Infection", Acta. Trop.. 46: 223-227; they reported that they could find no association of bacterial symbionts with the susceptibility of G. m. centraiis to Trypanosoma congolense infection on the basis of electron microscopic analysis. It is likely, however, that vector competence may be an interplay of a combination of factors contributed by all three organisms: fly, parasite and symbiont.

Bacterial symbionts are potential tools for altering an insect's capacity to transmit disease. Naturally-occurring symbiotic bacteria can be used for expressing new genes in disease-carrying arthropods, potentially for studying insect- pathogen interaction and for controlling disease transmission.

The tsetse can be cured of its symbionts through administration of a diet that includes oxytetracycline (Nooge, G. (1978), "Aposymbiotic Tsetse Flies, Glossina morsitans Obtained by Feeding on Rabbits Immunized Specifically with Symbionts", J. Insect, Physiol, 24: 299-304) . Since GM02 transformants

express resistance to the antibiotics ampicillin, tetracycline, and chloramphenicol, antibiotic selection can be utilized to replace natural symbiont populations of flies with genetically modified symbionts.

Example 10: Symbiont Isolation and Maintenance

The tsetse symbionts were isolated from laboratory-reared Glossina pallidipes and G. morsitans, obtained from the Tsetse Research Laboratory, Bristol, U.K., by a modification of the technique used by Welburn, S.C, Maudlin, I. and Ellis, D.S. (1987), "In vitro Cultivation of Rickettsia-like Organisms from Glossina spp.", Ann. Trop. Med. Parasitol. 81: 331-335. Twenty adult flies were surface sterilized for 10 minutes respectively in 70% ethanol and in a solution of 1% sodium hypochlorite. A sterile Pasteur pipette, drawn out to a fine point, was used to puncture the thorax and aspirate hemolymph from the fly. The hemolymph samples were inoculated into a single 0.5 ml aliquot of Mitsuhashi-Maramorosch medium supplemented with 20% heat-inactivated fetal bovine serum (M-M medium) (Mitsuhashi, J. and Maramorosch, K. , (1964), "Leafhopper Tissue Culture: Embryonic, Nymphal, and Imaginal Tissues from Aseptic Insects", Contrib. Bovce Thompson Inst. 22: 435-460) . The pooled hemolymph was added to a 5 ml suspension of Aedes albopictus cells, clone C6/36 (Igarashi, A., (1978), "Isolation of a Singh's Aedes albopictus

Cell Clone Sensitive to Dengue and Chikungunya viruses", J. Gen. Virol. 40: 531-544) and centrifuged at 500 x G for 10 minutes. The pellet was incubated at 28°C for 1.5 hours, resuspended, transferred to a 25 cm ² tissue culture flask, and maintained at 28°C Bacteria from the supernatants of symbiont-positive cultures were subsequently maintained in cell-free M-M medium at the same temperature and passaged weekly.

Example 11: Morphological studies

For light microscopy preparations, C6/36 cells were grown on 4-chamber Lab-Tek tissue culture slides (Nunc, Inc.). When the cells had grown to confluence, the chambers were inoculated with 100 μl of symbiont culture and incubated at 28°C for 2 days. The slides were then washed with phosphate buffered saline (PBS) , pH 7.2, dried, fixed with methanol, and stained with Giemsa stain. Gram stains were made of symbiont cultures which were spotted onto glass slides and heat fixed. For electron microscopy, cultures of C6/36 cells infected with symbionts were shaken to detach the cells, decanted, and centrifuged at 500 x G for 10 minutes. The pellets were fixed in a solution of 2.5% glutaraldehyde and 2.0% paraformaldehyde in 0.1 M sodium cacodylate buffer (pH 7.2) and postfixed in 1.0% osmium tetroxide. The cells were then pelleted in low melting point agar and cut into small blocks which were dehydrated in graded ethanol solutions, and embedded in Luffs resin (Luft, J.H.

(1961), "Improvements in Epoxy Resin Embedding Methods", J. Bio. Phγs. Biochem. Cvtol.. 9: 409-414). Thin sections were cut on a Reichert OM V3 microtome and viewed on a Phillips 201 electron microscope.

Example 12: 16s rRNA Sequence Analysis

The 16S rRNA of GPOl was sequenced directly by the dideoxynucleotide chain termination method, as described previously (Sanger, F. , Nicklen, S. and Coulson, A.R., (1977) "DNA Sequencing with Chain-terminating Inhibitors", Proc. Natl. Acad. Sci. USA. 74.:5463-5467; Lane, D.J. , Pace, B. , Olsen, G.J., Stahl, D. , Sogin, M.L. and Pace, N.R. , (1985), "Rapid Determination of 16S Ribosomal RNA Sequences for Phylogenetic Analysis, Proc. Natl. Acad. Sci. USA. 82:6955-6959). The sequences were aligned against a representative collection of enteric eubacteria and several other related gamma purple bacteria (Woese, C.R. , Gutell, R. , Gupta, R. and Noller, H.F., (1983) , "Detailed Analysis Higher-order Structure of l6S-like Ribosomal Ribonucleic Acids", Microbiol. Rev. 47: 621-669; Weisburg, W.G. , Tully J.G., Rose, D.L. , Petzel, J.P., Oyaizu, H. , Yang , D. , Mandelco, L. , Sechrest, J., Lawrence, T.G., Van Etten, J. , Namiloff, J. and Woese, C.R. (1989), "A Phylogenetic Analysis of the Mycoplasmas: Basis for their Classification", J. Bacteriol. 171: 6455-6467.

Pairwise evolutionary distances (expressed as estimated changes per 100 nucleotides) were computed from percent similarities using the correction of Jukes, T.H. and Cantor, C.R., (1969), "Evolution of Protein Molecules, p.2 1-132 In N. Munro, ed. , Mammalian Protein Metabolism. Academic Press, New York, as modified to accommodate the actual base ratios (Weisburg, W.G., Tully J.G., Rose, D.L., Petzel, J.P., Oyaizu, H. , Yang , D. , Mandelco, L. , Sechrest, J., Lawrence, T.G., Van Etten, J., Namiloff, J. and Woese, C.R., (1989)," A Phylogenetic Analysis of the Mycoplasmas: Basis for their Classification", J. Bacteriol _f 171: 6455-6467) . Dendrograms were constructed from evolutionary distance matrices by the method of De Soete, G. (1983) , "A Least Squares Algorithm for Fitting Additive Trees to Proximity Data", Psychometrika. 48: 621-626.

To determine the relatedness of GM02 to GPOl, an approxi¬ mately llδObp region of the 16S rRNA gene of GM02 was amplified by PCR from DNA extracted from cultured GM02 cells using conserved eubacterial primers (5'-GCTTAACACATGCAAG corresponding to E. coli positions 45-61 forward and 5•ACGGGCAGTGTGTACAAGACC corresponding to E. coli positions 1242-1227 reverse) (O'Neill, S.L. , Giordano, R. , Colbert, A.M.E., Karr, T.L. and Robertson, H.M. , (1992), "16S rRNA Phylogenetic Analysis of the Bacterial Endosymbionts Associated with Cytoplasmic

Incompatibility in Insects", Proc. Natl. Acad. Sci. USA, (1992) 89, 2699-2702) and cloned into t-tailed Bluescript Marchuk, D. , Drumm, M. , Saulino, A. and Collins, F.S. (1990), "Construction of T-vectors, a Rapid and General System for Direct Cloning of Unmodified PCR Products", Nuc. Acids. Res.. 19: 1154. Multiple clones were sequenced to identify Taq polymerase errors and/or cistron variability and a consensus sequence was derived from these clones which was then compared to the sequence of GPOl.

Example 13: Genetic Transformation

The plasmid pSUP204 was obtained from Dr. Louis Mallavia, Washington State University, and was used in transformation experiments which were carried out by a variation of the procedure of Sambrook, J., Fritsch, E.F. and Maniatis, T., (1989) , Molecular Cloning: A Laboratory Manual. Cold Spring Harbor Laboratory. Briefly, 15 ml cultures of the symbionts were grown to an 0D600 of approximately 0.3, pelletted and resuspended in 3 ml of cold F buffer [50 mM calcium chloride, 100 mM potassium chloride, 10 mM potassium acetate and 10% (w/v) glycerol (pH 6.2)]. The cells chilled on ice for 45 minutes, pelletted and resuspended in 200μl of cold F buffer. 100 μl of competent cells were transformed with 100 ng of pSUP204 DNA. The bacteria were incubated 30 minutes on ice followed by a heat shock at 42"C for 45 seconds. The tubes were returned to ice, and 0.9 ml of M-M medium was added. Following a 1 hour incubation at

28°C, 0.5 ml of the preparations were added to 25 cm ² tissue culture flasks containing 4.5 ml of M-M medium and 20 μg/ml ampicillin.

Example 14: Plasmid DNA Isolation and Analysis

Small-scale preparations of plasmid DNA from E. coli and from tsetse endosymbionts were performed by the boiling miniprep method (Sambrook, J., Fritsch, E.F. and Maniatis, T. , (1989), Molecular Cloning: A laboratory Manual. Cold Spring Harbor Laboratory) . Larger preparations were isolated by a modification of the alkaline extraction procedure (Sambrook, J., Fritsch, E.F. and Maniatis, T., (1989), Molecular Cloning: A Laboratory Manual. Cold Spring Harbor Laboratory) . Plasmid DNA was examined by restriction enzyme digestion and gel electrophoresis.

Example 15: Export. Turnover and Processing

Normally, expression of a gene within a cell has to be coupled to an export mechanism before the gene produced can be effective in the extracellular milieu. It is known that in prepromellitin (a related peptide) , the export mechanism requires ATP but does not require either a signal recognition particle or a docking protein (Mύller, G. and Zimmermann, R. , (1978), EMBO. J. , 6_, 2099-2107) . In many instances, activation of the pre-peptide to the active form has to occur. Symbionts in the

gut of Rhodnius prolixus represent approximately 10-15% of the total insect weight.

The mass of symbionts depends on the bug's state of nutrition. Five to seven days after a blood meal there may be an increase of six orders of magnitude in the number of symbionts in the gut and this number declines over the next ten days to resting levels. This cyclical return to normal levels is probably due to the lysis and turnover of symbionts in the insect gut. In aphids, lysozyme has been involved in the turnover process. It is considered that turnover will export the recombinant peptide into the gut. The processed Cecropin peptide is quite stable in the presence of proteases (Boman et al, (1989), J. Biol. Chem.. 264, 5582-5586), and symbiont turnover should not degrade Cecropins or their analogues.

The Cecropin A gene incorporated into the symbiont codes for all forms of the inactive and active Cecropin A peptide. In insects, step-wise proteolysis by a dipeptidyl peptidase converts inactive preprocecropin A into the active mature cecropin A peptide. Gut extracts from R. prolixus were tested for their ability to act on a specific chromogenic dipeptidyl peptidase substrate ALA-PRO-para-phenyl anilide. Dipeptidyl peptidase activity was present in 50 mM pH 5.0 acetate buffer in which washed R. prolixus

gut had been suspended for 20 minutes. Ninety microliters of the suspensate released 0.00002 moles chromogenic product anion at 120 minutes above control levels, giving an approximate proteolytic rate constant of 21 ^' Vi&in ^"1 . If the hydrolysis rate of preprocecropin A were similar to that of the chromogenic substrate, and if substrate was not rate-limiting, approximately 2.9 μg cecropin A could be released in one bug gut per minute. There are of course many caveats to such calculations. Almost certainly, suspending the gut in 5.0 ml buffer for 20 mins does not extract all the enzyme. The hydrolytic rates for the chromogenic substrate may be higher or lower than for preprocecropin and the gut may contain substances which do not diffuse easily and may affect substrate hydrolysis in vivo. Nevertheless, substantial dipeptidyl peptidase activity is associated with R. prolixus gut, and it seems likely that the product of the preprocecropin A gene will be processed to active cecropin in the R. prolixus gut.

Example 16: Cecropin A and Cecropin/Mellitin hybrid molecules will kill T. cruzi in vitro under Conditions in which the Rhodococcus rhodnii coccal Forms Survive

In vitro experiments tested the ability of L-Cecropin A and a L- Cecropin A (1-13)/Mellitin (1-13) hybrid peptide to kill both 7. cruzi and the Rhodococcus rhodnii symbiont. T. cruzi cultures were exposed

to the peptides for 24 hours and intact T. cruzi were counted on a hemocytometer. This method, which takes total disintegration as the end-point, probably over-estimates the peptide concentration needed for T. cruzi killing. Thus the actual MIC concentrations are probably lower than the ones needed for cell disintegration. Symbiont bacterial cultures were exposed for 24 hours to the peptides and then spread on BHI plates. Colony counts were performed at 72 hours. A summary of results is given in Table X.

Table X. Toxicity of L-Cecropin a and Cecropin (1-13)-Mellitin (1-13) Hybrid (CMH) Peptides on Micro Organisms - Summary of Data.

ORGANISM: MIC IN MICROMOLAR CONCENTRATIONS L-CECROPIN A CM HYBRID

T. cruzi 200 10

R. rhodnii ( 3-day culture) >500 47**

R. rhodnii (28 day culture) I 127**

£ coli (D21) * 0. 2 0. 5

* Reported by Wade et al. PNAS. 87:4761-4765 (1990) **MIC (99%) - survivors withstand >400 μM I=Insufficient peptide available to reach MIC

The conclusions drawn from Table X were as follows: (a) Toxic concentrations of L-Cecropin A for 7. cruzi epimastigotes were from 50-100 μM with complete lysis at 200 μM. The L-Cecropin/Mellitin hybrid (CMH) showed toxic effects on 7. cruziat 2-5 μM with complete lysis at 10 μM. CMH was

approximately two orders of magnitude more effective than L- Cecropin A against 7. cruzi.

(b) The gut of R. prolixus is normally inhabited by the coccal form of Rhodococcus rhodnii . Toxic concentration of both L Cecropin A and the CMH for the symbiont R. rhodnii depended on the age of the culture. Older (4 week) cultures were more resistant to the peptides and were enriched in coccal forms of the symbiont, as compared to young (3 day) cultures. In 3-day colonies toxic effects on the symbiont were observed at 200 μ Molar L-Cecropin A, but colonies in the coccal form continued to survive at 400 μ Molar Cecropin A concentration. The CMH showed toxicity to the symbiont in concentrations between 1-10 μ Molar, but even at a 100 μM some coccal survivor colonies remained. CMH gave approximately the same dosage effects on later 4 week R. rhodnii cultures except that the number of colonies at 100-400 μ Molar concentrations were considerably increased. A preliminary conclusion is that the coccal form of the R. rhodnii symbiont which is found in the bug gut, is resistant to the pore-forming peptides at a concentration of at least 400 μ Molar. Thus, it seems very likely that concentrations of the peptides which can be obtained in gut, will kill or inhibit T. cruzi without removal of all symbiotic bacteria.

Example 17: Toxicity of Cecropin A for Rhodnius prolixus.

Ten R. prolixus bugs were each fed with approximately 1.0 ml blood containing 3 x 10 ^"5 moles of Cecropin A per ml. There was no acute toxicity and compared with a control group fed blood not containing Cecropin A. There was no excess mortality in the experimental group over a period of 21 days.

It will be appreciated that the instant specification and drawings are set forth by way of illustration and not limitation, and that various modifications and changes may be made without departing from the spirit and scope of the present invention.