Release of Recombinant Microorganisms

0 downloads 0 Views 2MB Size Report
Page 1 ... This review addresses current environmental applications of naturally occur- ... recombinant microorganisms may be released in the future, while an exami- ... 27 distinct releases of recombinant microorganisms have occurred to date. ..... ment of the organism, as well as the resolution of problems resulting from.
Annual Reviews www.annualreviews.org/aronline Annu.Rev. Microbiol.1993.47.’913~14 Copyright©1993by AnnualReviewsInc. All rights reserved

Annu. Rev. Microbiol. 1993.47:913-944. Downloaded from arjournals.annualreviews.org by Dr Mark Wilson on 01/12/06. For personal use only.

RELEASE OF RECOMBINANT MICROORGANISMS M. Wilson and S. E. Lindow Department of Plant Pathology, University of California, 94720

Berkeley, California

KEYWORDS: field testing, fungi, bacteria, bioremediation, biological control

CONTENTS INTRODUCT]ION .......................................... CURRENT ENVIRONMENTALAPPLICATIONS OF NATURALLYOCCURRING MICROORGANISMSAND ANTICIPATED USES OF RECOMBINANT STP, AINS ........................................ Biological Control of Plant Diseases ............................ Biological Control of lnsect Pests and Vectors ...................... Biological Control of Weeds ................................. Symbiotic Nitrogen Fixation ................................. Bioremediation .......................................... HISTORY AND ANALYSIS OF THE RELEASE OF NONENGINEERED MICROORGANISMS ................................. Frequency of Release of Nonengineered Microorganisms ................ Assessment of the Fate of Released Nonengineered Microorganisms ......... RELEASES OE RECOMBINANT MICROORGANISMS ................. Frequency of Release of Recombinant Microorganisms ................. Assessment of the Fate of Released Recombinant Microorganisms .......... CONCLUDING REMARKS ............. ’ ......................

914

914 914 919 921 922 924 925 925 929 930 930 932 934

ABSTRACT This review addresses

current

environmental

applications

of naturally

occur-

ring, nonrec,ambinant microorganisms and potential future genetic modifications of such organisms, as well as releases of recombinant microorganisms that haveoccurredto date. Awareness of the current uses of nonrecombinant microorganisms provides insight into the diversity of habitats in which recombinant microorganisms may be released in the future, while an examination of potential and realized genetic modifications provides insight into the variety of applications for which recombinant microorganisms may be used. 913

0066-4227/93/1001-0913502.00

Annual Reviews www.annualreviews.org/aronline 914

WILSON & LINDOW

Annu. Rev. Microbiol. 1993.47:913-944. Downloaded from arjournals.annualreviews.org by Dr Mark Wilson on 01/12/06. For personal use only.

Analysis of the behavior, persistence, and dispersal of nonrecombinantstrains further provides valuable information required for the assessmentof the risk involved in release of recombinantderivatives of those strains. Approximately 27 distinct releases of recombinant microorganismshave occurred to date. This review assesses what has been learned from such releases regarding persistence, dispersal, and potential deleterious environmentaleffects. INTRODUCTION Microorganismshave evolved adaptations permitting them to colonize most aquatic and terrestrial habitats. Thus, the adaptations of naturally occurring microorganisms are highly diverse and represent an invaluable genetic resource. Whilemanynaturally occurring microorganismswith desirable traits are already used for various purposes, molecular techniques allow the manipulation of strains for more effective or varied applications. Genetic techniques are most often used to make subtle changes in microorganisms that already have manyof the attributes required for a particular use. Current molecular techniques allow directed modification of microorganisms by removal or addition of one or a few genes (109). Advancesare being made in the manipulation of fungi and yeasts, but genetic manipulation of these organisms lags behind that of bacteria. This review emphasizes current environmental applications of nonrecombinant microorganisms to try to anticipate likely future applications of recombinant organisms, primarily bacteria, that will require their release into the open environment.Because very few recombinant microbes have yet been introduced into the open environment, this activity remains primarily a speculative venture. We therefore analyze the releases of both unaltered and recombinantmicroorganisms to better address the risks associated with the numerousintroductions of recombinantmicrobes to natural habitats that are anticipated to occur in the future. Weattempt to place the release of recombinant microbes in perspective with the manyunaltered microbesthat have been and will be field tested. CURRENT ENVIRONMENTAL APPLICATIONS OF NATURALLY OCCURRING MICROORGANISMS AND ANTICIPATED USES OF RECOMBINANT STRAINS Biological

Control of Plant Diseases

CONTROL OF DISEASESONTHEAERIALSURFACES OF PLANTS The control of diseases on the aerial surfaces of plants is limited by the availability of effective chemicals, particularly in the case of bacterial diseases. Few

Annual Reviews www.annualreviews.org/aronline

Annu. Rev. Microbiol. 1993.47:913-944. Downloaded from arjournals.annualreviews.org by Dr Mark Wilson on 01/12/06. For personal use only.

RELEASE OF RECOMBINANTMICROORGANISMS 915 bactericides are available; growers rely on copper compounds and antibiotics such as st:reptomycin and oxytetracycline. Resistance to these bactericides is becomingmore widespread and alternatives are nowbeing sought. Biological control of aerial diseases, particularly those caused by phytopathogenic bacteria, should be an important future application of recombinantmicroorganisms. Manynaturally occurring microorganisms have been used to control diseases on the aerial surfaces of plants (7, 17, 179). The more common bacterial species that have been used for the control of diseases in the phyllosphere include Pseudomonassyringae, Pseudomonas fluorescens, Pseudomonascepacia, Erwinia herbicola, and Bacillus subtilis. One strain of P. fluorescens has recently been registered for the commercialcontrol of fire blight on pear. Fungal genera that have been used for the control of diseases in the phyllosphere include Trichoderma, Ampelomyces,and the yeasts Tilletiopsis and Sporobolomyces.The mechanismsof action proposed for these bliological-control agents, whichinclude competitionfor sites and/or nutrients, ~ntibiosis, and hyperparasitism,represent targets for strain improvement through genetic manipulation. Several phytopathogenic bacteria exhibit an epiphytic phase prior to invasion, during which time they are susceptible to competition from other microorga~isms. While preemptive competitive exclusion of phytopathogenic bacteria in the phyllospherecan be achievedusing naturally occurring strains, avirulent mutants of the pathogen, in whichdeleterious phenotypictraits have been remoxred, may be more effective because they occupy the same niche as the parental strain. Phytopathogenicbacteria possess several genes that encode phenotypesthat allow themto parasitize plants and overcomedefense responses elicited by the plant (132). In addition, phytopathogenicbacteria possess pathogenicity genes such as hrp (191). Isogenic, avirulent mutants can be prod~ucedby insertional inactivation of genesinvolvedin pathogenicity. A nonpathogenic strain of Pseudomonassyringae pv. tomato, produced by Tn5 insertional mutagenesis, prevented growth of pathogenic strains in the tomato phy]ilosphere, presumablyby preemptive competitive exclusion (35). Nonpathogenic mutants of Erwinia amylovora, produced by transposon mutagenesis,havealso been used in the biological control of fire blight (125). Antibiosis has been proposed as the mechanismof control of several bacterial (183) and fungal (103) diseases in the phyllosphere. Molecular techniquescould possibly be used to enhancethe efficacy of biological-control agents whose primary mode of action is antibiosis. For example, the transcriptional regulation of genes conferring antibiotic production could be altered by substitution of a constitutive regulatory region or by replacing the regulatory region with promotersknownto direct high levels of transcription. It mayalso be possible to transfer the genesrequired for antibiotic production

Annual Reviews www.annualreviews.org/aronline

Annu. Rev. Microbiol. 1993.47:913-944. Downloaded from arjournals.annualreviews.org by Dr Mark Wilson on 01/12/06. For personal use only.

916

WILSON & LINDOW

from an organism that colonizes the phyllosphere only poorly to one that colonizes more aggressively. Although genes responsible for antibiotic production in E. herbicola have been cloned (90), they have not yet been expressed in other epiphytic species. Biological-control agents must normally achieve a high population in the phyllosphere in order to control other strains, but colonization by the agent may be reduced by competition with the indigenous microflora. Application of a bactericide to which most membersof the indigenous microflora are sensitive, but to whichthe biological-control agent is resistant, can maximize colonization by the biological-control agent. Integration of chemicalpesticides and biological-control agents has been reported with Trichodermaspp. (71); and P. syringae pv. tomato(35). Biological-control agents tolerant to specific bactericides could be constructed using molecular techniques. Copper-tolerance determinants occur as a single operon in P. syringae (15), which could be transferred to a copper-sensitive organism (164), thereby producing biological-control agent that can be applied simultaneously with a copper bactericide. Moleculartechniques mayeventually be used to transfer several beneficial traits, such as the productionof one or moreantibiotics and pesticide resistance, to an aggressive phyllosphere colonist, which also posses~e~ a conditional suicide systemrestricting it to one specific host plant. BIOLOGICAL CONTROL OFFROST INJURY Frost injury of sensitive crop plants is incited by ice nucleation-active (Ice +) bacterial species, particularly P. syringae, which inhabit the aerial surfaces of plants (107). Such plants can avoid damageby supercooling if Ice+ bacterial populations are low or absent. Biological control of frost injury can be achieved by the prophylactic application of naturally occurring Ice- strains to the uncolonizedblossomor leaf tissue, in a mannersimilar to that employedin the biological control of phytopathogenic bacteria. Preemptive competitive exclusion of Ice+ strains fromthe aerial surfaces by the Ice- strains reduces the probability of freezing injury. P. fluorescens strain A506has recently been registered commercially for the control of frost injury of pear and is available as FrostbanB). While biological control of frost injury can be achieved using naturally occurring Ice- strains, isogenic Ice- P. syringae mutants were hypothesized to be most effective in the exclusion of a parental Ice+ P. syringae strain. The gene conferring ice nucleation activity in P. syringae was cloned and deletions internal to the structural gene were produced in vitro (130). Reciprocal exchange of the deletion-containing ice gene for the native chromosomalgene was accomplished by a process of homologous recombination (108). The resulting Ice- P. syringae mutants were phenotypically and behaviorally identical to the Ice÷ parental strains, except for the inability to nucleate ice (108). These Ice- P. syringae mutants effectively reduced the

Annual Reviews www.annualreviews.org/aronline RELEASE OF RECOMBINANT MICROORGANISMS 917

Annu. Rev. Microbiol. 1993.47:913-944. Downloaded from arjournals.annualreviews.org by Dr Mark Wilson on 01/12/06. For personal use only.

populations of isogenic Ice + P. syringae strains both in the laboratory and the field (112-114, 117). Althoughmolecular techniques have been successful in constructing a biological-control agent of frost injury, biological control of a heterogeneousmixture of indigenous Ice+ strains can be accomplishedwith a naturally occurring strain such as P. fluorescens A506.The use of mixtures of complementaryrecombinantIce- P. syringae strains for biological frost control mayimprove control. BIOLOGICAL CONTROL OF SOIL-BORNE DISEASES Chemical control of soilborne plant diseases is frequently ineffective because of the physical and chemicalheterogeneity of the soil, whichmayprevent effective concentrations of the che~nical from reaching the pathogen. Chemicalsused in the soil are also generally resistant to degradation and therefore tend to persist as environmentalpollutants. Biological-control agents in contrast specifically colonize tl~te rhizosphere, the site requiring protection, and leave no toxic residues. The localization of the biological-control agent in the rhizosphere allows the application of a smaller quantity of the biological agent than the more broadly spread chemicals. Microorganismshave been used extensively for the biological control of soil-borne plant diseases and also for plant-growth promotion(12, 169, 187). Fluorescent pseudomonads are the most frequently used bacteria for biological control and plant-growth promotion, but Bacillus and Streptomyces species have also been commonlyused. Trichoderma, Gliocladium, and Coniothyrium species are the most frequently used fungal biological-control agents. Perhaps the most successful biological-control agent of a soil-borne pathogen is Agrobac~teriumradiobacter strain K84, used against crowngall disease caused by Agrobacterium tumefaciens (53). A great deal of interest has focused on the mechanismsof action of these biological-control agents, particularly the role played by antibiotics. Consequently, manyworkers in this area have reported modifications aimedat efficacy enhancement.Biological control with A. radiobacter strain K84is mediated primarily by the bacteriocin agrocin 84 synthesis, which is directed by genes on a plasmid, pAgK84.This plasmid also bears genes for resistance to agrocin 84 and conjugal transfer capacity. Consequently, pAgK84 maybe transferred to A. tumefaciens, which then becomesresistant to agrocin 84. To prevent this resistance, a transfer-deficient mutant of Strain 1(84 was constructed. A. radiobacter strain K1026is identical to the parental strain, except that the agrocin-producing plasmid, pAgK1026,has had the conjugal-transfer region deleted (88). This strain is nowused commerciallyworldwidefor the control of crowngall disease (152, 184). Competitionas a mechanismof biological control has been exploited with soil-borne pliant pathogensas with pathogenson the aerial surfaces of plants.

Annual Reviews www.annualreviews.org/aronline

Annu. Rev. Microbiol. 1993.47:913-944. Downloaded from arjournals.annualreviews.org by Dr Mark Wilson on 01/12/06. For personal use only.

918

WILSON & LINDOW

Naturally occurring, nonpathogenicstrains of Fusariumoxysporumhave been used to control wilt diseases caused by pathogenic Fusarium spp. (4). Moleculartechniques have been used to removevarious deleterious traits of soil-borne phytopathogenicbacteria to construct a competitive antagonist of the pathogen. RandomTn5 insertions into the genome of Pseudomonas solanacearum(174, 175) or insertion of an interposon (~-km) into hrp cluster (60) produced avirulent mutants. The avirulent mutants exhibited various levels of invasiveness of tomatoplants and providedprotection against bacterial wilt disease caused by the virulent pathogen (60). The phytopathogenic bacterium Erwinia carotovora subsp, carotovora secretes various extracellular enzymes, including pectinases, cellulases, and proteases. The pectinases are knownto be a major pathogenicity determinant in soft rot disease of potato. E. carotovora subsp, carotovora mutants defective in the production of pectate lyase (140) have been used in the biological control this disease (161). Out- mutants of E. carotovora subsp, betavasculorum, which produce but do not secrete pectinases, have been constructed (36) and mayalso have potential for the biological control of potato soft rot. Moleculartechniques have also facilitated the introduction of beneficial traits into rhizosphere-competentorganismsto producepotential biologicalcontrol agents. Chitin is a major structural component of many plant pathogenic fungi. Biological control of somesoil-borne fungal diseases has beencorrelated with chitinase production(23) and chitinolytic bacteria exhibit antagonismin vitro against fungi (66). The importanceof chitinase activity was further demonstratedby the loss of biological-control efficacy in Serratia marcescens mutants in which the chiA gene had been inactivated (89). Chitinase genes have been cloned from several genera including Streptornyces (139), Vibrio (193), and Cellvibrio (194). A recombinant Escherichia coli containing the chiA gene from S. marcescenswas effective in reducing disease incidence caused by Sclerotium rolfsii and Rhizoctonia solani (129, 148). another study, chitinase genes from S. marcescens were expressed in Pseudomonas sp. conferring the ability to control the pathogens F. oxysporum f.sp. redolens and Gauemannomyces graminis var. tritici (162, 163). Various extracellular antibiotics producedby Pseudomonas spp. have been shownto be involved in the biological-control ability of soil-borne plant pathogens (58), including phenazine-l-carboxylic acid (PCA)(168, oomycin A (74, 87), pyoluteorin (PLT) (75, 91), and 2,4-diacetylphloroglucinol (PHL)(75, 91). In systems where antibiosis has been shown to play a primary role, molecular techniques can be used to enhance biological-control efficacy by increasing levels of antibiotic synthesis, either by increasing the copy numberof the biosynthetic genes or by achieving constitutive synthesis. For example, increased production of PLTand PHL and superior control of Pythium ultirnurn damping-off of cucumber was

Annual Reviews www.annualreviews.org/aronline

Annu. Rev. Microbiol. 1993.47:913-944. Downloaded from arjournals.annualreviews.org by Dr Mark Wilson on 01/12/06. For personal use only.

RELEASE OF RECOMBINANT MICROORGANISMS 919 achieved by increasing the numberof antibiotic biosynthesis genes in P. fluorescens strain CHA0 (75, 119). Constitutive synthesis of oomycinA P. fluorescens strain HV37awas achieved by insertion of a strong promoter from E. coli upstream of the afuE locus. The recombinant P. fluorescens strain containing the tac-afuE construct producedsignificantly higher levels of oomycinthan the parental strain and provided greater control of P. ultimurn infection (7’2, 73). Alternatively, biosynthetic genescan be introducedinto strain deficient in antibiotic production,or into one that producesa different antibiotic in order to increase the spectrum of activity. ClonedPCAbiosynthetic genes were transferred from P. fluorescens 2-79, which exhibits poor rhizosphere competence,into P. putida and P. fluorescens strains that exhibit superior rhizosphere competence.The recombinantstrains, which synthesized PCAin vitro, are potentially superior biological-control agents because of their ability to colonize the rhizosphere (22). In a similar study, a cloned genomic fragment from Pseudomonas Fl13 was transferred into various Pseudomonas strains, one of which was subsequently able to synthesize PHL and inhibit P. ultimum damping-off of sugar beet (56). The spectrum activity through expression of an additional antibiotic increased whengenes conferring PHLsynthesis were mobilized from P. aureofaciens into P. fluorescens 2-79, which normally only produces PCA. This procedure increased activity against G. graminis var. tritici, P. ultimum, and R. solani (185). Genetic modifications of biological-control fungi so far have been limited by the availability of appropriate methodologies. While transformation systems are being developed(153), genetic manipulation of Trichodermaand Gliocladiumspecies is currently restricted to physical and chemicalmutation and protoplast fusion (77, 79, 159). Trichodermaharzianumisolates tolerant to fungicides have been generated by spontaneous mutation (1) and exposure (134). In another study, improved root colonizing ability and biological-control efficacy was achieved by protoplast fusion with two T. harzianumstrains (78). Moleculartechniques permit the construction of a superior biological-control agent by such approachesas the introduction of one or morebeneficial traits into an. organismexhibiting high levels of rhizosphere competenceor inactivation of the pathogenicity genes in a pathogen. A biological-control agent could potentially be tailored to each host/pathogen system. However, the superior ability of such strains ur~der field conditions remains to be demonstrated. Biological Control of Insect Pests and Vectors While many insect pests can be controlled through the use of chemical pesticides, bioaccumulation of pesticide residues and the contamination of

Annual Reviews www.annualreviews.org/aronline

Annu. Rev. Microbiol. 1993.47:913-944. Downloaded from arjournals.annualreviews.org by Dr Mark Wilson on 01/12/06. For personal use only.

920

WILSON & LINDOW

soils and groundwater is a major environmental concern. In a rapidly developing field, investigators have focused on biological alternatives to pesticides, particularly the use of Bacillus toxins and of entomopathogenic fungi. Several genera of entomopathogenic fungi have shownpotential for the control of insect pests of agricultural crops, including Aschersonia,Beauveria, Hirsutella, Metarhizium,and Verticillium (26, 67, 120). Beauveriabassiana, one of the most extensively studied mycoinsecticides, has been used to control Europeancorn borer (Ostrinia nubilalis), Coloradopotato beetle (Leptinotarsa decemlineata), and codling moth (Cydia pomonella) (67). Although entomopathogenic fungi have been subjected to physical and chemical mutagenesis, the literature reports few examplesof transformation (80). The first step toward enhancing the efficacy of entomopathogenicfungi lies in the identification of traits determiningpathogenicity such as spore adhesion and germination, penetration and production of cuticle-degrading enzymes, production of toxins and enzymes, and avoidance of host recognition (80). Biological-control activity could possibly be enhancedthrough genetic manipulations affecting such features as sporulation, spore dispersal, and stress tolerance of spores (80). The developmentof fungal transformation vectors (153,196) mayallow the transfer of genes conferring, for example, lipase protease activity to a strain that exhibits near optimal pathogenicity but is deficient in a single trait (80). Fungicideresistance will be an importanttrait for any mycoinsecticide planned for an agricultural application. The entomopathogen Metarhiziurn anisopliae has been transformed to benomyl resistance using the benA3gene from Aspergillus nidulans (68). To date, recombinantentomopathogenicfungi have been released in the field. Biological pest control agents based on Bacillus thuringiensis are produced commerciallyand used worldwide(40, 54). B. thuringiensb strains produce insecticidal crystal proteins (ICPs) that exhibit specific activity against different insect orders, includingDiptera (B. thuringiensis subsp, israelensis), Lepidoptera(B. thuringiensis subsp, kurstaki), and Coleoptera(B. thuringiensis subsp, tenebrionis) (40). Anotherspecies, Bacillus sphaericus, exhibits toxic activity against mosquitoes of the genera Aedes, Anopheles, and Culex (14, 133). The crystal protein (cry) genes have been studied extensively their potential for genetic manipulationreviewed(64, 85, 195). B. thuringiensis strains exhibiting novel ICP gene combinationshave been producedby plasmid curing and conjugal transfer (25, 64). This methodology has been used to produce the Condor® and Foil® bioinsecticides. In the case of the Foil ® bioinsecticide, conjugal transfer was used to develop a B. thuringiensis strain with a broader spectrum of activity. Plasmids encoding activity against the Europeancorn borer (O. nubilalis) were combinedwith a plasmid-encoding activity against the Colorado potato beetle (L.

Annual Reviews www.annualreviews.org/aronline

Annu. Rev. Microbiol. 1993.47:913-944. Downloaded from arjournals.annualreviews.org by Dr Mark Wilson on 01/12/06. For personal use only.

RELEASE OF RECOMBINANT MICROORGANISMS 921 decemlineata)(25, 64). Field trials with the Foil® bioinsecticide demonstrated the superior performanceof this product comparedwith other B. thuringiensis-based ]products (64, 65). Recently, the developmentof B. thuringiensis cloning vectors (63, 64, 195) and an efficient transformation system for thuringien’.ris based on electroporation (19) has permitted the construction recombin~uatB. thuringiensis strains. A broader spectrum of activity can be achieved in recombinantB. thuringiensis strains expressing both the native and cloned ICPs (37, 102). The poor persistence of B. thuringiensis products in the environmenthas promptedthe developmentof various carder or cellular delivery systems, in which the ICP genes are expressed in a novel host chosen to enhance persistence of the toxin in the environmentof the target pest (62). Effectiveness of mosquito control using B. sphaericus and B. thuringiensis subsp. israelensis is limited by sedimentationof the spores out of the larval feeding zone (14, 133). Expression of the ICPs in various cyanobacterial species may improvepersistence of the bioinsecticide in the larval feeding zone (8, 30, 41a, 167). B. thuringiensis cry genes have been expressed in several hosts for use in agricultural ecosystems(55, 62). Rhizobacterial species transformed to express B. thuringiensis cry genes include P. fluorescens (69, 127, 186), P. cepacia (160), and Rhizobiumspp. (154). A novel methodof delivery of ICPs using endophytic bacteria was recently developed. The endophytic bacterium Clavibacter xyli subsp, cynodontis (Cxc), originally isolated from Bermudagrass, serves as a systemic endophyte in corn (136). Different ICP gene constructions from B. thuringiensis subsp. kurstaki HD73were introduced into the chromosomeof Cxc (42, 52, 181). The recombinantCxc/Btexpresses the ICP ofB. thuringiensis subsp, kurstaki, which is toxic to the Europeancorn borer. Systemic colonization of corn plants by the recombinantCxc/Bt (98, 137) prevented or reduced corn borer damageto corn in the field (52, 99, 173). Current efforts are directed increasing the production of the ICP in recombinantCxc/Btin order to enhance potency of ingested bacterial cells (52). Recombinant B. thuringiensis strains and other epiphytic and rhizobacterial species expressing ICPs will undoubtedly be used for pest control in the future. The use of strains expressing multiple or hybrid ICPs mayhelp to delay the developmentof pest resistance. Biological’

Control of Weeds

While weed control with chemical herbicides has been largely successful, herbicides suffer from a lack of selectivity, and their persistent nature often makesthem serious environmentalpollutants. Biological weed-control agents can potential.ly be highly specific and generally leave no toxic residues. Many species of plant pathogenicfungi havebeen tested for their ability to control a variety of weeds, and several of these are nowcommerciallyavailable (29,

Annual Reviews www.annualreviews.org/aronline

Annu. Rev. Microbiol. 1993.47:913-944. Downloaded from arjournals.annualreviews.org by Dr Mark Wilson on 01/12/06. For personal use only.

922

WILSON & LINDOW

165, 166). Phytophthora palmivora (DeVine*) has been used commercially since 1981to control strangler vine (Morreniaodorata) in citrus groves (165). Colletotrichum gloeosporoides f. sp. aeschynomene(Collego®) has been used commerciallysince 1982 for control of northern joint vetch (Aeschynomene virginica) (165). Other mycoherbicidesthat are either commerciallyavailable or under testing include C. gloeosporoidesf.sp. cuscutae (Luboa2) for control of dodder, Alternaria cassiae (CASST)for control of sickelpod, and Colletotrichurn coccodes (VELGO) for control of velvet leaf (165). The optimal pathogen for a mycoherbicideapplication should exhibit a high level of virulence and a narrow host range; however,few naturally occurring pathogens exhibit both attributes (142). Genetic manipulationmaybe used enhancethe virulence of a narrow-host-rangepathogen, or alternatively reduce the host range of a highly virulent pathogen. Chemical and physical mutagenesisof Sclerotinia sclerotiorum, a highly virulent pathogenof Canada thistle, produced mutants with reduced host ranges that have potential for control of this weed(142). Several authors have suggested possible approaches to enhance the virulence of fungal pathogens (28, 70, 92, 165), including introduction of genes for the production of specific phytotoxins, or enzymes, such as cutinases, pectinases, and cellulases. Such modifications, however, will not be possible until appropriate fungal transformation systems have been developed(196) and genes for the biosynthesis of these virulence factors have been cloned. Symbiotic

Nitrogen

Fixation

Leguminous crops form symbiotic associations with Rhizobiurn and Bradyrhizobiutn that fix atmosphericnitrogen in a form that can be used by the plant. Rhizobial inoculants have been used for several years in attempts to increase legumeproductivity. Genetic modificationof rhizobial strains has focused on three areas: host range modification, enhancednitrogen fixation, and enhancedcompetition with indigenous strains for nodulation (84, 110, 131,158). Several types of genes are involved in symbiotic nitrogen fixation, includingnodulation(nod), nitrogen fixation (fix), and nitrogenase(n/f) genes. In Rhizobium species these genes reside on sym plasmids, while in Bradyrhizobiumspecies functional analogues of these genes reside on the chromosome(43). Host-range modifications have been achieved by mutation of the common nodulation genes (nodDABC).The operon nodABCis activated by the gene product of nodDin combination with host-secreted phenolics (43). Chemical mutation of nodDin Rhizobiumleguminosarumby. trifolii produced mutants with inducer-independent ability to activate nod gene expression, thereby extendingthe host range comparedwith the parental strain (121). In a similar study, a flavonoid-independent hybrid NodDprotein constitutively activated

Annual Reviews www.annualreviews.org/aronline

Annu. Rev. Microbiol. 1993.47:913-944. Downloaded from arjournals.annualreviews.org by Dr Mark Wilson on 01/12/06. For personal use only.

RELEASE~OF RECOMBINANT MICROORGANISMS923 nod genes in R. legurninosarurnand R. meliloti, increasing the host range of these strains (156). Host-range modifications have also been achieved manipulatie,n of the host-specific nodulation genes (nodEFGH) (41, 44). The host range ,af R. leguminosarum bv. trifolii was extended to include alfalfa plants following the integration of nodEFG and nodHgenes from R. meliloti into its sym plasmid (57). Transfer of pSymgenes into alternate hosts, such as A. tumefaciens and E. coli, mayeventually permit nitrogen fixation in nonleguminoushosts (83, 180). Twodifferent approaches have been used to enhance nitrogen fixation-modified e~pression of the regulatory gene nifA and improved substrate transport through modifiedexpression of the C4-dicarboxylatetransport (dct) genes (24, 131, 141). In Bradyrhizobiumjaponicum and R. meliloti, the nifA product activates transcription of several genesinvolvedin nitrogen fixation. Increased e~pression of nifA was achieved by chromosomalinsertion of an enhancementcassette containing an additional copy of nifA. RecombinantB. japonicum ~mdR. meliloti strains with additional nifA sequences improved legumeyields in greenhouseand field trials (21,24, 141). Therate of nitrogen fixation maybe additionally limited by the rate of uptake of dicarboxylic acids, the p~hmarysource of energy. Transfer of dct genes from R. meliloti to B. japonicurn resulted in higher nitrogenase activity (18). A recombinant R. meliloti with a chromosomallyinserted enhancementcassette with both dctABDand nifA gave the largest yield increases in recent field trials (T. Wacek, personal communication). Certain strains of rhizobia contain an uptake hydrogenasethat can use H2 to produce ATP. Hydrogengas is a by-product of nitrogenase activity and strains that possess uptake hydrogenage (Hup+) can recycle this hydrogen with the formation of ATP(158). The hup gene, encoding biosynthesis of uptake hydrogenase, has been cloned and used to transform naturally Hupstrains. These Hup+ recombinants have been shown to exhibit increased nitrogen fixation (131). However,most of the strains used commerciallyare Hup+; hence such modifications may not be of value unless gene dosage effects are demonstrated(131). Improvementsin legume yield achieved with rhizobial inoculants are limited by competitionbetweenthe inoculant and indigenousrhizobial strains that are frequently poor at fixing nitrogen (157, 178). Considerableeffort has been directed to the determinationof phenotypesand genotypesthat contribute to superior nodulation competitiveness (178). These phenotypes include antibiosis, cell surface characteristics, and motility (178). R. leguminosarum bv. trifolii strain T24producesa potent antibiotic, trifolitoxin, active against other rhizobia (177), but this strain producesnoduleson clover that fix little nitrogen (176). Insertion of trifolitoxin biosynthesis ((fx) genes into genomeof a symbiotically effective strain of R. leguminosarumbv. trifolii

Annual Reviews www.annualreviews.org/aronline 924

WILSON & LINDOW

produceda recombinantstrain that was highly competitive with respect to a trifolitoxin-sensitive strain in greenhousetests (176, 178). Modifications symbiotically effective strains by addition of traits beneficial in nodulation competitiveness mayultimately produceinoculant strains that achieve acceptable frequencies of nodule occupancy,even whenindigenous population levels are high.

Annu. Rev. Microbiol. 1993.47:913-944. Downloaded from arjournals.annualreviews.org by Dr Mark Wilson on 01/12/06. For personal use only.

Bioremediation The bioremediation of environmental pollutants such as petroleum hydrocarbons or recalcitrant synthetics may in the future become an important application of recombinant microorganisms. Althoughsome toxic wastes can be treated in contained bioreactors (49) or activated sludge systems, in situ bioremediation of petroleumspills or contaminatedsoils or aquifers necessitates the release of organisms into the environment. Nonrecombinantshave been used in the bioremediation of petroleum hydrocarbons (9, 100) and xenobiotics (16). Although microcosmtests of recombinants have occurred (51, 126), and the EPArecently granted permission for a contained test of recombinantE. coli for trichloroethylene (TCE)degradation (192; B. Ensley, personal communication),no field releases of recombinantshave taken place. Few naturally occurring organisms possess the necessary degradative pathwaysfor complete mineralization of the more recalcitrant xenobiotics, such as pentachlorophenol(PCP), 2,4,5-trichlorophenoxyacetic acid (2,4,5-T) and polychlorinated biphenyls (PCB). Operonsencoding steps in the mineralization of xenobiotics are frequently plasmid borne (20, 143). Metabolic pathways can be constructed by in vivo genetic manipulation through conjugational transfer in mixtures of strains, each possessing componentsof the required degradative pathway(20, 27). The degradative strains produced by these techniques have been extensively reviewed (76, 124, 138). In vitro construction of degradative pathwaysthrough genetic engineering has great potential in systems for which sufficient genetic and biochemical informationis available (2, 50, 61,124). Moleculartechniquesare particularly useful for enzymerecruitment. Genes encoding enzymeswith broad substrate specificity were addedto extend the usefulness of the chlorocatechol pathway in several bacteria. Pseudomonas B13 has a pathway for the complete degradation of chlorocatechols, but the first enzymein the chlorobenzoate pathwayhas narrowsubstrate specificity. Recruitmentof a broad-specificity dioxygenase from a TOLplasmid permitted the degradation of a wider range of chlorobenzoates and chloroaromatics (101). In situ bioremediation may unsuccessful if an inducer of a key degradative enzymeis absent. Molecular techniques have been used to alter the regulation of degradative enzymes requiring induction. Insertional mutagenesisin P. cepacia strain G4 produced mutants that constitutively metabolizedTCEwithout aromatic induction (151).

Annual Reviews www.annualreviews.org/aronline

Annu. Rev. Microbiol. 1993.47:913-944. Downloaded from arjournals.annualreviews.org by Dr Mark Wilson on 01/12/06. For personal use only.

RELEASE OF RECOMBINANTMICROORGANISMS925 Successful application of recombinantorganismsfor in situ bioremediation will depend not only on the construction and optimization of degradative pathwaysbutt also on the successful introduction, establishment, and containmentof the organism, as well as the resolution of problemsresulting from the heterogeneous distribution of the target compound,nutrients, and the degrading organism (16, 50, 115). Biological containment systems based lethal genes, possibly coupled to the regulatory systemof the substrate to be degraded, have been proposed (33, 50) and maybe a regulatory requirement of field releases. HISTORY AND ANALYSIS OF THE RELEASE NONRECOMBINANT MICROORGANISMS

OF

Althoughmuchattention has focused recently on the release of recombinant microorganisms, the use of nonrecombinant microorganisms in the open environmenthas a long history and is pertinent to this issue. Wehave much to learn from studying the nature of previous uncontained releases of nonrecombinantmicroorganisms. A commonassertion is that nonrecombinant strains havebeen released frequently without harm. Wetherefore try to assess not only the .frequency with whichnonrecombinantstrains have been released but also address assessmentsof their fate and effects in an effort to apply this knowledgeto the release of recombinantstrains. Frequency

of Release

of Nonrecombinant

Microorganisms

The release of microorganismsinto nature has long been an integral part of the research activities of several biological disciplines includingplant pathology and entomology. For the purpose of this review, we have attempted to identify all releases of microorganisms within the past three years. This review of the literature was restricted to the introduction of bacteria and fungi into the open environment, specifically excluding viruses. Figure 1 summarizes research reports describing the introduction of microorganismsinto the open environment. Plant pathogens are the most commonlyreleased microorganisms, followed closely by nonpathogenic microorganisms used in the biological control of plant diseases. Mycorrhizal fungal species and many strains of Rhizobiumhave also been commonlyreleased as part of research activities. Interestingly, very few bacteria have yet been released for the bioremediation of toxic compoundsor agricultural pesticides. The actual number of yearly releases of nonrecombinant microorganisms may differ substantially from that reflected in Figure 1. Weexpect that the actual number of releases rnay in somecases be higher by a factor of 2 or more because many microorganisms may have been released incidentally to the main purposeof a :study. Plant pathogensare usually released into field sites either

Annual Reviews www.annualreviews.org/aronline

Annu. Rev. Microbiol. 1993.47:913-944. Downloaded from arjournals.annualreviews.org by Dr Mark Wilson on 01/12/06. For personal use only.

926

WILSON & LINDOW

to insure the presence of a pathogen so disease-control strategies can be evaluated or to measure the impact of a disease on crop productivity. Such studies are usually conducted where the previous occurrence of the disease insures the presence of sufficient pathogen inoculum, but such sites do not always exist. The uniform presence of a pathogen can be insured by its inoculation into a field site. Hundreds of releases of plant pathogens occur yearly. In 1992 alone, we identified at least 138 reported studies in which plant pathogens were released into field sites (Figure 1). Because only experiments that are successful are generally reported in the literature, substantially more than the reported number of releases probably occurred. Similarly, most reports of field studies include data from at least two years of field results to allow evaluation of the reproducibility of a result. Therefore, the number of actual

03 150 "--’ 120 "~: 90 ~-

¯ [] [] ~] [] []

gems 13iorem rhizobium mycor biocontrol pathogens

6O

O _~ 30

1987 1988198919901991

1992

Year of Appearance Figure1 Publishedreports on the use of different microorganisms in field studies. Thenumber of publishedstudies involving(left to right) genetically engineeredmicroorganisms (gems), microorganisms for bioremediation,Rhizobium or Bradyrhizobiurn species, mycorrhizalspecies, microorganisms used in biological control of plant pests, andplant pathogensis givenfor the first year of the experimentalstudyreported. Whileall uses of gemsare tabulated, releases of plant pathogensis reportedonlyfor 1992andthe use of the other organisms is reportedonlyfor the past three yearsfor comparative purposes.

Annual Reviews www.annualreviews.org/aronline

Annu. Rev. Microbiol. 1993.47:913-944. Downloaded from arjournals.annualreviews.org by Dr Mark Wilson on 01/12/06. For personal use only.

RELEASE OF RECOMBINANTMICROORGANISMS 927 releases maybe greater than the numberof reports of a pathogen’sintroduction by a factor of two or more. Because field inoculations of plant pathogens have occurred since the turn of the century, thousands of releases of plant pathogens have been made. Theincreased interest in biological control of plant diseases and insects as an alternative to chemicalpesticides in the past 15 years (7, 17, 40, 67, 169, 179, 187) has led to the testing of manynaturally occurring microorganisms. Anaverage of about 55 studies have been reported annually from field tests of fungi and’. bacteria for the control of plant diseases; the numbersare fewer for the control of insects and weedswith pathogenic strains. The targets of biological-control efforts include both foliar and soil-borne plant pathogens, foliage- and root-feeding insects, and terrestrial and aquatic weedspecies. Therefore, indigenous microorganismshave been released into manydifferent habitats. Thelong history of biological-control studies suggest that hundreds of field rele~lses have occurred. For example,as of 1989, 109 pathogensthat could infect a total of 69 different weedplant species were underconsideration for biologice~l weedcontrol (29). Manyof these pathogens had already been investigated under field settings to test their efficacy; manymore probably will be tested for biological-control potential in the field in the future. Mycorrhizal associations of roots continue to be intensively studied to determinetheir effects on plant nutrition and productivity as well as disease avoidance (189). Each of the past three years has seen an average of about 40 research reports on the inoculation of plants under field conditions with mycorrhizal fungi. A wide range of terrestrial plant species have been inoculated with one of several different mycorrhizalfungal species. The infection of leguminousplants with Rhizobiumspecies is an important feature of the nitrogen metabolismof such plants. Therefore, attention has focused on this interaction for over a century (157, 158). Earlier studies Rhizobiumspecies tended to focus on practical aspects of inoculation of plants with this species under field conditions, survival in soil, and productivity of inoculated plants. More recent studies of Rhizobium and Bradyrhizobium species have primarily addressed genetic and biochemical determinants of plant-microbe interactions. Nonetheless, an average of about 40 research reports involving field releases of Rhizobiumspecies have been published in each of the past three years (Figure 1). Becauseinoculation of leguminous plants with ,~hizobium and Bradyrhizobium species is now a common agricultural practice, manythousandsof inoculations occur yearly aroundthe world (157, 158). Not only h~tve more plant pathogens and microorganismswith the potential for biological control been released than other types of agents, the variety of these organisms and of the types of habitats into which they have been introduced is muchwider than for other microorganisms.There are hundreds

Annual Reviews www.annualreviews.org/aronline

Annu. Rev. Microbiol. 1993.47:913-944. Downloaded from arjournals.annualreviews.org by Dr Mark Wilson on 01/12/06. For personal use only.

928

WILSON & LINDOW

of different plant pathogenicfungal and bacterial species, and most have been tested under field situations---clearly considerable numbersof diverse microorganisms have been introduced to the open environment. Nearly every report of a released pathogen involves a genotypically unique strain. In contrast, manystudies of mycorrhizae and Rhizobiumspecies involve repeated inoculations of a few different strains into different habitats. Certainly, however, thousands of habitat-microbe genotype combinationshave been tested as part of field testing of plant pathogensand biological-control agents in nature. In addition to releases of microorganismsduring research studies, several microorganisms are registered for commercial use as pesticides. The US EnvironmentalProtection Agency(EPA) has registered 195 different strains of 11 different bacterial species for control of fungal and bacterial diseases of roots, control of bacterial-incited plant frost damage,and control of insect pests (6). The vast majority of these taxa are B. thuringiensis strains with different spectra of insecticidal activity. Elevendifferent strains of six fungal species are also registered by the EPAfor biological control of weedplants, plant diseases, and insect larvae (6). Severalof these registered pesticides are produced and used commercially on a substantial scale, thus resulting in innumerable applications to field settings. For example, the fungus C. gloeosporoides f.sp. aeschynomene is sold as the product Collego ® and applied widely in the southern USto kill Northernjoint vetch (A. virginica) in rice fields (29, 165, 166). Several commercialformulations of different thuringiensis subspecies are also commonlyused worldwideon agricultural and forest crops to manageinsect pests (29, 40, 54). While most pathogens, biological-control agents, and mycorrhizal or Rhizobiumstrains that have been field tested represent a reintroduction into a habitat that is the same or similar to the one from which the organismwas isolated, a few microorganisms have been introduced from nonindigenous sources. This distinction is clearly illustrated by the use of fungi as biological-control agents of weeds. Most fungal species inoculated onto weeds, such as A. cassiae and C. gloeo6poroides, were isolated in North Americaand then reintroduced onto plants in large numbersto cause severe disease as a "mycoherbicide" (29, 165, 166). In contrast, the rust fungus, Puccinia chondrillina, was isolated in the Mediterranean region of southern Europe, the site of origin of skeleton weed(Chrondrillajuncea), and released in Australia and the westernUSto attack this weedbecause these areas lacked any of the plant’s predators or parasites (3, 38, 39). Rust fungi are easily dispersed by wind, so these pathogenshave spread far from the initial-introduction spot and have successfully reduced the aggressiveness of the weed plants that they attack; no unforeseenimpact on the terrestrial habitats into which they were introduced have been observed (39). Three other fungi have been similarly introduced to various parts of the world in classical biological

Annual Reviews www.annualreviews.org/aronline RELEASE OF RECOMBINANTMICROORGANISMS 929

Annu. Rev. Microbiol. 1993.47:913-944. Downloaded from arjournals.annualreviews.org by Dr Mark Wilson on 01/12/06. For personal use only.

weed control programs. Nonindigenousbacteria have also been introduced for evaluation of the biological control of plant diseases (EPA,unpublished data). The EPAhas authorized the field testing of 14 nonindigenous B. thuringiensis strains in the US. The introduction of nonindigenousmicroorganismsinto the environmentcertainly must pose at least the samerisk as the introduction of genetically engineered microorganisms because the nonindigenous strains have manynovel traits not present in the indigenous microflora. Assessment of the Fate of Released Nonrecombinant Microorganisms Releases of nonrecombinantmicroorganismshave apparently not resulted in significant perturbations of the habitats into whichthey havebeen introduced, but the larg,e majority of release studies have not sought to address this possibility. Mostintroductions of pathogens,for example,have beento limited areas, and no subsequentreports of disease of plants outside the area initially inoculated or of unexpected host plants have appeared. In most studies specifically reisolating the inoculated pathogen wouldbe difficult because genetic markers such as antibiotic resistance traits are not present on the strains. Therefore, moststudies apparently have not attempted to describe the fate of the introduced microorganisms. Several studies, however, have involved the release of pathogens for the purpose of determining their persistence or spread as part of other epidemiological studies (188). Most these reveal that after inoculation in high numbers,the population size of bacterial plant pathogens drops to low, and frequently undetectable, levels within a few days to a few years. In addition, the populationdynamicsof biological-control agents have often been measured. Because a biological-control agent of plant disease must colonize plants to achieve control, manystudies have quantified antagonist populations both spatially and temporally (81, 116, 187). The dispersal organisms, however, has not often been the focus of such studies. Obvious deleterious effects either on plant growth, such as stunting or pathogenicity, or on other components of a field site, such as arthropodmortality attributable to a biological-control agent, have not been reported in the literature. The widespreadcommercialuse of plant pathogensas biological-control agents of weedshas not resulted in unexpectedeffects on nonsusceptible plant species (165). The difficulty in describing changesin the microbial communitiesinto which microorganisms have been introduced explains why more detailed studies of perturbations resulting from released microbes have not been commonlymade. The diversity of microorganismspresent in most terrestrial soil or plant habitats, the difficulty in isolating all microbial components of the habitat, and their uncertain taxonomic position, makea quantitative

Annual Reviews www.annualreviews.org/aronline 930

WILSON & LINDOW

measureof perturbation associated with released microorganismsdifficult to realize (13, 59). Manystudies addressing risks of recombinant microorganisms now focus on improvementof methods for such assessments (13, 45, 59, 144, 147, 182).

Annu. Rev. Microbiol. 1993.47:913-944. Downloaded from arjournals.annualreviews.org by Dr Mark Wilson on 01/12/06. For personal use only.

RELEASES Frequency

OF

RECOMBINANT

of Release

MICROORGANISMS

of Recombinant

Microorganisms

Comparedwith the numerousreleases of nonrecombinantmicroorganismsthat occur yearly, the number of recombinant microorganisms that have been released remains small. For the purposes of this review, microorganisms altered by acquisition of a plasmid, by protoplast fusion, or by chemical or ultraviolet mutagenesis are not considered recombinant. Several microorganisms altered in such ways have been introduced into the open environment. For example, the EPAhas authorized the field testing of 10 transconjugant B. thuringiensis strains containing a different complementof plasmids harboring ICP genes. The EPAhas also authorized the release of two Trichodermastrains resulting from protoplast fusion and 10 strains of either Verticillium lecanii, S. sclerotiorum, or T. harzianum with chemical or ultraviolet-induced mutations. In contrast, since the first field release of a Tn5-containing strain of P. fluorescens in the Netherlands in 1986, in only 27 studies have recombinantmicroorganismsbeen tested in the open environment (Table 1). All field tests have involved recombinantbacteria. A large percentage of the tests involve simple inactivation of genes by transposon insertion (82); genetic tagging of bacteria with selectable or identifiable markers such as lacZY (47, 48, 135), the lux operon(150), or antibiotic-resistance markers (5, 10, 11); or deletion of genes by molecular genetic procedures in vitro (88, 105, 111,117, 184). Moreover,manyof the releases represent the introduction of the samealtered strain to different sites. Initial experiments with recombinantbacteria in the US, especially those of microorganismsmeant to be tested as biological pesticides, were burdened with the necessity of obtaining an Experimental Use Permit (EUP)from the EPA(93, 95, 111). Acceptable documentation for an EUP, normally necessary only for large field tests of chemicalpesticides, required a great deal of information and greatly dela,,ed initial experiments (111). Also required for these experiments was the collection of detailed data on the dispersal and survival potential of the altered bacteria (93, 95, 111). The EPA has subsequently considerably relaxed such requirements, and though the procedures to obtain permits for the testing of recombinant microorganisms from the USDepartmentof Agriculture, Animal, and Plant Health Inspection

Annu. Rev. Microbiol. 1993.47:913-944. Downloaded from arjournals.annualreviews.org by Dr Mark Wilson on 01/12/06. For personal use only.

Annual Reviews www.annualreviews.org/aronline RELEASEOF RECOMBINANT MICROORGANISMS 931

Annual Reviews www.annualreviews.org/aronline

Annu. Rev. Microbiol. 1993.47:913-944. Downloaded from arjournals.annualreviews.org by Dr Mark Wilson on 01/12/06. For personal use only.

932

WILSON & LINDOW

Service (USDA-APHIS) still necessitate considerable effort, they have overall been greatly simplified (149). Despite this apparent relaxation of requirements, the numberof studies of genetically engineered microorganisms(GEMS)that have been initiated has not increased as rapidly as manyprojected (Figure 1). Althoughfour field studies releasing genetically engineeredbacteria beganin 1987, still only four were initiated in 1991. The numberof initiated studies peaked at eight in 1990, and we could find no evidence of new studies initiated in 1992 (Figure 1). The number of releases of recombinant microbes continues to be fewer than the numberof conferences and review articles addressing the safety of such releases. Weagree with the assessment of Shawet al (149, p. 51) that: ¯ . . restrictions on the introductionof GEMS in field experimentsposeda majorproblemto the pursuitof suchstudies. Federalregulationsand,to a lesser extent, state regulationsoften prohibit suchreleases unlessextensivegreenhouse and microcosm studies havebeenpreviously conducted.The burdento develop these data typicallyrests uponthe researcherandis onerousto prohibitive.Even after generatingthe requireddata, there is no guaranteethat a permitwill be issued. For these reasons, academicscientists with limited resources have generally beendiscouragedfromconductingsuch experimentation. Assessment of the Fate of Released Microorganisms

Recombinant

In contrast to most introductirns of nonrecombinantbacteria, the fate of GEMS has been carefully monitored. A necessary and prominent feature of permits obtained from both the EPAand USDA-APHIS is an experimental design and procedures that enable careful description of the survival and dispersal of the introduced microorganism. For example, permits from EPA to conduct releases of ice nucleation-deficient (Ice-) mutants of P. syringae and P. fluorescens in 1987required extensive studies of the aerosol dispersal of spray-applied inoculum and frequent surveys for the presence of these antibiotic resistant-marked strains on plants, soil, insects, and water in and aroundthe plot areas (111). Permits also required that all treated plant material be destroyedat the end of the experimentto test the feasibility of eradication of the introduced bacteria (111). The EPAalso conducted extensive studies of the dispersal of bacteria during and after spray inoculation at field sites to documentthe dispersal potential of the Ice- bacteria applied in such a manner and to test different detection methods (146). The results of monitoring released Ice- P. syringae strains reveal that the rate of dispersal was closely predicted by studies of these strains conducted in the laboratory and greenhouse and by prior field experiments using nonrecombinantbacterial strains (106, 108, 111, 117).

Annual Reviews www.annualreviews.org/aronline

Annu. Rev. Microbiol. 1993.47:913-944. Downloaded from arjournals.annualreviews.org by Dr Mark Wilson on 01/12/06. For personal use only.

RELEASE OF RECOMBINANTMICROORGANISMS933 Release~; of several recombinantPseudomonasstrains have been designed specifically to evaluate their fate and dispersal in field sites. The use of lacZY-encodedI~-galactosidase activity in combinationwith rifampicin resistance mar~:ers allowed rapid and unambiguousselection and identification of introducedstrains (48, 94). Initial trials of Pseudomonas aureofaciensstrains containing lacZY (which was introduced with a disarmed variant of Tn7), involved extensive studies of its survival and movement in the plot area (94, 95). Greenhouseand microcosmstudies predicted that the P. aureofaciens strain would initially colonize inoculated wheat seeds and roots in large numbersbut would slowly decrease in population size with time (93-96). Thesestrains indeed decreased to undetectable levels within 31 weeksafter inoculation~into field sites in SouthCarolina(94). It wasalso anticipated that movement through soil by such strains or along the roots of wheat would be restricted, with the largest populations occurring near the inoculated seeds. The recombinantstrains of P. aureofaciens behaved in a mannersimilar to that projected in each of several field studies (34). Notransfer of Tn7::lacZY from the c]hromosomalinsertion in P. aureofa¢iens into other soil microorganismswas detected (94). The behavior of recombinantC. xyli subsp, cynodontis strains containing genes for an ICP from B. thuringiensis was also predicted from preliminary studies using nonrecombinantstrains or recombinantstrains studied in the laboratory. Greenhousestudies of C. xyli indicated that it could be moved from plant to plant by mechanicaltransmission but not by insects (97). Such findings indicated that the recombinantstrain should be contained within the plants into which it was inoculated or in plants developing from infected seeds. To date, this has been the case in field studies of recombinantstrains of this species (97). Whilelarge populations of recombinantC. xyli developed in inoculated corn plants, it was not detected on trap plants nearby(97). Althoughthe numberof recombinant microorganismsthat have been so far released into the open environment remains small, we agree with the statements by Kluepfel et al (94) regarding conclusions that can be drawn fromthe re:suits of initial experiments(p. 351): Theinitial results of this plannedreleaseof a geneticallyengineered, soil-borne root-colonizingbacteriumhaveshownthat sucha release can be conductedin a safe andre, sponsiblemanner.Though weshouldguardagainst generalizationwe and other~ have not observedany adherent danger in the use of genetically engineeredbacteria in the environmentsimplybecausethe organismhas been modifiedgenetically. This, however,doesnot excludethe necessityof a case by case evaluationof the microorganism and its newgenetic construct before its releaseinto the environment. Bycontinuingto build our databaseonthe microbial ecologyon both native and genetically engineeredmicroorganisms we will move

Annual Reviews www.annualreviews.org/aronline 934

WILSON & LINDOW

towardthe establishmentof scientifically-basedrisk assessmentproceduresthat will guidetheir futurerelease.

Annu. Rev. Microbiol. 1993.47:913-944. Downloaded from arjournals.annualreviews.org by Dr Mark Wilson on 01/12/06. For personal use only.

CONCLUDING REMARKS This review was not intended to address the voluminousliterature on the risk assessment of releases of recombinantbacteria. Manyexcellent treatments of different aspects of this subject have ~ppeared (31, 32, 59, 86, 155, 171, 172, 190). A deterrent to generalized methodsof assessing risk has been the diversity of the microbesthemselves,the large numberof traits that potentially might be modified, and the diverse habitats into whichthey will be introduced. For this reason, different componentsof risk take on paramountimportance for various recombinant microbes, while some risk componentsare common to nearly all GEMS. For example, the risk of horizontal gene transfer should be consideredin all cases. Manystudies haveaddressedthe potential of genetic transfer by conjugation, transformation, transduction, and cell fusion (32, 122, 128, 145), Unfortunately, most studies have focused on horizontal transfer using a plasmid model, often extrapolating from studies of antibiotic resistance plasmids that can undergostrong selection in an antibiotic-containing environment. This model has been strongly criticized (32). Comeaux al (32) provideda very lucid description of genetic transfer in relation to risk assessment of recombinantbacteria. Weagree strongly with their conclusions that (p. 132): ¯ . . for... GEMS, systems maybe used that minimizegenetic exchange. However,total elimination of genetic transfer from or to GEMS released into agriculturehabitats is probablyan unrealistic goal. Therefore,considerationsfor plannedreleases should documentand evaluate gene movement from GEMS into indigenousbacterial populationsand evaluatethe ecologicalimpactof that gene movement on community structure and function. Assessmentof risks should be basedon reasonableinterpretationsof the results of logical tests for potential hazardsand perforceavoidtests for moreesoteric potential hazards. The potential effects of GEMSon a habitat into which they have been introduced are hard to assess. As discussed above, even describing a microbial community,except at a generalized functional level, is largely beyond our current capabilities. Weencourage the developmentof new tools to enable sensitive assays of changes in communityfunction as well as in describing communitycomposition. Such tools will not only prove extremely useful in assessing potential risks associated with the release of genetically engineered microbes,but will be a great aid in all future microbial ecological studies. A commonconcern in releases of GEMSis that they will out-compete

Annual Reviews www.annualreviews.org/aronline

Annu. Rev. Microbiol. 1993.47:913-944. Downloaded from arjournals.annualreviews.org by Dr Mark Wilson on 01/12/06. For personal use only.

RELEASE OF RECOMBINANT MICROORGANISMS 935 indigenou:~ microbes because of the novel trait that they carry. Because nonindige~aousbacteria frequently do not becomeestablished whenintroduced into a newhabitat, indigenousbacteria are typically genetically manipulated for environmental releases; hence, GEMS usually will be introduced into habitats in which genotypically and phenotypically similar organismsalready exist. In rnanycases, a study of the behaviorof the parental strain in a given habitat should be sufficient to evaluate the prospects of the GEM.Successful colonization of the parental strain in a field setting should indicate the basic capability of the genetically altered strain in a given habitat. Comparative microcosrntstudies can then be done to determinethe contribution of the new trait(s) introducedinto the GEM to predict any differences in testable behavior of the GEM in the field. Therefore, one will not have to predict the field behavior of a genetically engineered microbefrom laboratory features of the GEM alone but, instead, can rely on extensive knowledgeof the parental strain or knowledgethat can be gained from the release of the parental strain into a given habitat to assist in prediction of the propensity of a GEM to occupya giiven site and to have effects on indigenouspopulations because of such occupation. Weintended in this review to provide insight into the opportunities and needs for the release of GEMS to advanceour knowledgeof microbial ecology and to benefit society. Ananalysis of previous work demonstrates that GEMS will likely be used for the biological control of plant diseases and pests as well as nitrogen fixation in the future. Weanticipate that the future will also see a great demandfor releases of GEMS for bioremediation. The absence of careful quantitative examinationof the fate of manypreviously introduced nonenginee~edorganisms weakensthe foundation for understanding the fate of released GEMS. However,carefully conducted studies with released GEMS suggests th~.t their release will have little deleterious effect. By summarizing the history .of releases of nonengineeredmicrobes and the current status of the releases of GEMS,we hope to have shown that GEMS can be released without dek;terious perturbation of a given habitat, and that their release, in combination with continued cautious and thorough study, can contribute to superior strategies for their effective application to solve field-based problems. ACKNOWLEDGMENTS

Wethank B. Ensley, T. Wacek,C. Gawron-Burke,S. Kostka, T. Yamamoto, and D. Dra~tos for providing useful information, including unpublisheddata for this review, as well as J. Payne from USDA-APHIS and S. Matten from the EPAfor providingcompilationsof regulatory actions relating to the release of GEMS, which were helpful in assembling the list provided here. Wealso thank G. Beattie for helpful manuscriptsuggestions.

Annual Reviews www.annualreviews.org/aronline

936

WILSON

Literature 1.

Annu. Rev. Microbiol. 1993.47:913-944. Downloaded from arjournals.annualreviews.org by Dr Mark Wilson on 01/12/06. For personal use only.

2.

3.

4.

5.

6.

7. 8.

9.

9a.

10.

11.

& L1NDOW

Cited

Abd EI-Moity, T. H., Papavizas, G. C., Shatla, M. N. 1982. Induction of newisolates of Trichodermaharzianum tolerant to fungicides and their experimental use for control of white rot of onion. Phytopathology 72:396-400 Abramowicz, D. A. 1990. Aerobic and anaerobic biodegradation of PCBs: a review. Crit. Rev. Biotechnol. 10:24151 Adams, E. B., Line, R. F. 1984. Epidemiology and host morphology in the parasitism of rush skeleton weed by Puccinia chondrillina. Phytopathology 74:745-48 Alabouvette, C., Couteaudier, Y. 1992. Biological control of Fusariumwilts with non-pathogenic fusaria. See Ref. 171a, pp. 415-26 Amarger, N., Delgutte, D. 1990. Monitoring genetically manipulated Rhizobiurn leguminosarum biovar viciae released in the field. See Ref_ l17a, pp. 221-28 Anderson, E. L., Betz, F. S. 1990. EPAperspective on risk assessment for environmental introductions: progression from small-scale testing to commercial use. See Ref. l17a, pp. 119-26 Andrews,J. H. 1992. Biologicalcontrol in the phyllosphere. Annu. Rev. Phytopathol. 30:603-35 Angsuthanasombat, C., Panyim, S. 1989. Biosynthesis of 130-kilodalton mosquitolarvicide in the cyanobacterium Agmonellum quadriplicatum PR6. Appl. Environ. Microbiol. 55: 2428-30 Atlas, R. M., Atlas, M. C. 1991. Biodegradation of oil and bioremedlation of oil spills. Curt. Opin. Biotechnol. 2:440-43 Baker, R., Dunn, P. E,, eds. 1990. NewDirections in Biological Control: Alternatives for Suppressing Agricultural Pests and Diseases. NewYork: Liss Bakker, P. A. H. M., Salentijn, E., Hoekstra, W. P. M., Schippers, B. 1990. Fate of transposon Tn5 labelled Pseudomonasfluorescens in the field. In Proc. Int. Workahopon Plant Growth Promoting Rhizobacteria, 2nd, pp. 412-13. Interlaken, Switzerland: Int. Union of Biological Sciences Bakker, P. A, H. M., Schippers, B., Hoekstra, W. P. M., Salentijn, E. 1990. Survival and stability of a Trd tran~poson derivative of Pseudomonas

12.

13.

14.

15.

16. 17.

18.

19.

20.

21.

22.

22a. 23.

fluorescens WCS374 in the field. See Ref. l17a, pp. 201-4 Bakker, P. A. H. M., van Peer, R., Schippers, B. 1991. Suppression of soil-borne plant pathogens by fluorescent pseudomonads: mechanisms and prospects. In Biotic Interactions and Soil-Borne Diseases, ed. A. B. R. Beemster, G. J. Bollen, M. Gedagh, M. A. Ruissen, B. Schippers, A. Tempel, pp. 217-30. Amsterdam:Elsevier Bej, A. K., Perlin, M., Atlas, R. M. 1992. Impact of introducing genetically modified microorganisms on soil microbial communitydiversity. See Ref. 159a, pp. 137-39 Berry, C., Hindley, J., Oei, C. 1991. The Bacillus sphaericus toxins and their potential for biotechnological development. See Ref. l17b, pp. 35-51 Bender, C. L., Cooksey, D. A. 1987. Molecular cloning of copper resistance genes from Pseudomonassyringae pv. tomato. J. Bacteriol. 169:470-74 Bewley, R. J. F. 1992. Bioremediation and waste management.See Ref. 159a, pp. 33--45 Blakeman, J. P., Fokkema,N. J. 1982. Potential for biological control of plant diseases on the phylloplane. Annu. Rev. Phytopathol. 20; 167-92 Birkenhead, K., Manian, S. S., O’Gara, F. 1988. Dicarboxylic acid transport in Bradyrhizobiumjaponicum: use of Rhizobium meliloti dct gene(s) to enhance nitrogen fixation. J. Bacteriol. 170:184-89 Bone, E. J., Ellar, D. J. 1989. Transformation of Bacillus thuringiensis by electroporation. FEMSMicrobiol. Left. 58:171-78 Boyle, M. 1992. The importance of genetic exchange in degradation of xenobiotic chemicals. In Environmental Microbiology, ed. R. Mitchell, pp. 319-33. NewYork: Wiley-Liss Breitenbeck, G. A., Hankinson, T. 1992. Dispersal and persistence of recombinant Bradyrhizobium japonicum. Agron. J. 84:259 Bull, C. T., Weller, D. M., Tomasbow, L. S. 1991. Relation between root colonization and suppression of Gauemannomyces graminis var. tritici by Pseudomonasfluorescens strain 279. Phytopathology 81:954-59 Burge, M. N., ed. 1988. Fungi in Biological Control Systems. Manchester: ManchesterUniversity Press Buxton, E. W., Khalifa, O., Ward,

Annual Reviews www.annualreviews.org/aronline RELEASE

Annu. Rev. Microbiol. 1993.47:913-944. Downloaded from arjournals.annualreviews.org by Dr Mark Wilson on 01/12/06. For personal use only.

24.

25.

26.

27.

28.

29. 30.

31. 32.

33.

34.

OF RECOMBINANT

V. 1’965. Effects of soil amendment with chitin on pea wilt caused by Fusarium oxysporum f.sp. pisi. Ann. appl. Biol. 55:83-88 Cannon,F. C., Beynon,J., Hankinson, T., Kwiatlowski, R., Legocki, R. P., et al. 1988. Increasing biological nitrogen fixation by genetic manipulation. In Nitrogen Fixation: Hundred Years After, ed. I4. Bothe, F. J. de Bruijn, W. E. Newton, pp. 735-40. New York: Gustav Fischer Carlton, B. C. 1988. Developmentof genetically improvedstrains of Bacillus thuringiensis: a biological insecticide. In Biotechnology for Crop Protection. Symposium Series 379, ed. P. A. Hedin, J. J. Menn, R. M. Hollingworth, pp. 260--79. Washington, DC: Am. Chem. Soc. Carruthers, R. I., Hural, K. 1990. Fungi as naturally occurring entomopathogens. See Ref. 9a, pp. 11538 Chal~abarty, A. M., Friello, D. A., Bopp, L. H. 1978. Transposition of plasmid DNAsegments specifying hydrocarbon degradation and their expression in various microorganisms. Proc. Natl. Acad. Sci. USA75:3109-12 Cbarudattan, R. 1985. The use of natural and genetically altered strains of pathogens for weedcontrol. In Biological Control In Agricultural IPM Systems, ed. M. A. Hoy, D. C. Herzog, pp. 347-72. Orlando: Academic Charudattan, R. 1991. The mycoherbicide approach with plant pathogens. See Ref. 164.b, pp. 24--57 Chungjatupomchai, W. 1990. Expression of the mosquitocidal-protein genes of Bacillus thuringiensis subsp. israelensis and the herbicide-resistance gene bar in Synechocystis PCC6803. Curr. Microbiol. 21:283-88 Colwell, R. R. 1991. Risk assessment in envi:ronmental biotechnology, Curr. Opin. Biotechnol. 2:470-75 Comeaux, J. L., Pooranampillai, C. D., Lacy, G. H., Stromberg, V. K. 1990. Transfer of genes to other populations;, and analysis of associated potenti~d risks. See Ref. 118, pp. 13245 Contreras, A., Molin, S., Ramos, J. L. 199l. Conditional-suicide containment system for bacteria which mineralize aromatics. Appl. Environ. Microbiol. 57:1504--8 Cook, R. J., Weller, D. M., Kovace,Ach, P., Drahos, D., Hemming, l:l., Barnes, G., Pierson, E. L. 1990. Establishment, monitoring, and

35.

36.

37.

38.

39.

40.

41.

41a.

42.

43.

MICROORGANISMS

937

termination of field tests with genetically altered bacteria applied to wheat for biological control of take-all. See Ref. l17a, pp. 177-87 Cooksey, D. A. 1988. Reduction of infection by Pseudomonassyringae pv. tomato using a non-pathogenic, copper-resistant strain combined with a copper bactericide. Phytopathology78: 601-3 Costa, J. M., Loper, J. E. 1992. Out-minus mutants of Erwinia carotovora subsp, betavasculorum with potential for biological control of potato soft rot. Phytopathology 82:1121 Crickmore, N., Nicholls, C., Earp, D. J., Hodgman,T. C., Ellar, D. J. 1990. The construction of Bacillus thuringiensis strains expressing novel entomocidal g-endotoxin combinations. Biochem. J. 270:133-36 Cullen, J. M. 1976. Evaluating the success of the programmefor the biological control of Chondrilla juncea L. In Proc. Int. Syrup. on Biological Control of Weeds, 4th, ed. T. E. Freeman, pp. 117-21. Gainesville: Univ. Fla. Cullen, J. M., Kable, P. F., Katt, M. 1973. Epidemic spread of a rust imported for biological control. Nature 244:462-64 Currier, T. C., Gawron-Burke, C. 1990. Commercialdevelopment of Bacillus thuringiensis bioinsecticide products. See Ref. 124a, pp. 111-43 Debelle, F., Maillet, F., Vasse, J., Rosenberg, C., de Billy, F., et al. 1988. Interference between Rhizobium meliloti and Rhizobiumtrifolii nodulation genes: genetic basis of Rhizobium meliloti dominance.J. Bacteriol. 170: 5718-27 de Marsac, N. T., de la Torre, F., Szulmajster, J. 1987. Expression of the larvicidal gene of Bacillus sphaericus 1593Min the cyanobacterium Anacystis nidulans R2. MoLGen. Genet. 209:396-98 Dimock, M. B., Beach, R. M., Carlson, P. S. 1989. Endophytic bacteria for the delivery of crop protection agents. In Proceedings of Biotechnology, Biological Pesticides and Novel Plant-Pest Resistance for Insect Pest Management, ed. D. W. Roberts, R. R. Granados, pp. 88-92. Ithaca, NY: Boyce ThompsonInst. for Plant Research Djordevic, M. A., Gabriel, D. W., Rolfe, B. G. 1987. Rhizobium--the refined parasite of legumes. Annu.Rev. Phytopathol. 25:145-68

Annual Reviews www.annualreviews.org/aronline

Annu. Rev. Microbiol. 1993.47:913-944. Downloaded from arjournals.annualreviews.org by Dr Mark Wilson on 01/12/06. For personal use only.

938

WILSON

& LINDOW

44. Djordevic, M. A., Schofield, P. R., Rolfe, B. G. 1985. Tn5 mutagenesis of Rhizobium trifolii host-specific nodulation genes result in mutants with altered host-range ability. Mol. Gen. Genet. 200:463-71 45. Drahos, D. J. 1991. Current practices for monitoring genetically engineered microbes in the environment. AgBiotech News Inf. 3:39-48 46. Drahos, D. J. 1991. Field testing of genetically engineered microorganisms. Biotechnol. Adv. 9:157-71 47. Drahos, D., Bah’y, G., Hemming,B., Brandt, E. 1988. Field-testing of lacZYmarkedrecombinantsoil bacteria. Phytopathology 78:1544 48. Drahos, D. J., Hemming,B. C., McPherson, S, 1986. Tracking recombinant organisms in the environment: 13-galactosidase as a selectable non-antibiotic marker for fluorescent pseudomonads. Bid~Technology 4:439-44 49. Drahos, D. J., Mueller, J. G., Lantz, S. E., Head, C. S., Middaugh, D. P., Pritchard, P. H. 1992. Microbial degradation of high molecular weight polyaromatic hydrocarbons at the American Creosote Works superfund site, Pensacola, Florida. In Proc. Int. Symp. on the Biosafety Results of Field Tests of Genetically Modified Plants and Microorganisms, 2rid, ed. R. Casper, J. Landsmann,pp. 153-62. (Abstr.) 50. Duque, E., Ramos-Gonzalez, M., Delgado, A., Contreras, A., Molin, S., Ramos,J. L. 1992. Genetically engineered Pseudomonasstrains for mineralization of aromatics: survival, performance, gene transfer, and biological containment. See Ref. 62a, pp. 429-37 51. Dwyer, D. F., Rojo, F., Timmis, K. N. 1988. Fate and behavior in an activated sludge microcosm of a genetically-engineered micro-organism designed to degrade substituted aromatic compounds. See Ref. 164a, pp. 77-88 52. Fahey, J. W., Dimock, M. B., Tomasino,S. F., Taylor, J. M., Carlson, P. 1992. Genetically engineered endophytesas biocontrol agents: a case study from industry. In MicrobialEcology of Leaves, ed. S. S. Hirano, J. H. Andrews, pp. 401-11. NewYork: Springer Verlag 53. Farrand, S. K. 1990. Agrobacterium radiobacter strain K84: a model biocontrol system. See Ref. 9a, pp. 679-91 54. Feitelson, J. S., Payne, J., Kim, L.

55.

56.

57.

58. 59.

60.

61,

62.

62a.

63.

64.

1992. Bacillus thuringiensis: insects and beyond. Bid~Technology10:271-75 Feitelson, J. S., Quick, T. C., Gaertner, F. 1990. Alternative hosts for Bacillus thuringiensis delta-endotoxin genes. See Ref. 9a, pp. 561-71 Fenton, A. M., Stephens, P. M., Crowley, J., O’Callaghan, M., O’Gara, F. 1992. Exploitation of gene(s) involved in 2,4-diacetylphloroglucinol biosynthesis to confer a newbiocontrol capability to a Pseudomonasstrain. Appl. Environ. Microbiol. 58:3873-78 Folch, L L. 1992. Extension of Rhizobium leguminosarumbiovar trifolii RS1051host range to alfalfa plants by integration into its symbiotic plasmid of nodEFG and nodH genes of Rhizobium meliloti. Presented at Int. Symp. Mol. Plant-Microbe Interact., Seattle, Wash. Fravel, D. R. 1988. Role of antibiosis in the biocontrol of plant diseases. Annu. Rev. Phytopathol. 26:75-91 Fredrickson, J. K., Hagedorn,C. 1992. Overview: identifying the ecological effects from the release of genetically engineered microorganisms and microbial pest control agents. In Microbial Ecology: Principles, Methods, and Applications, ed. M. A. Levin, R. J. Seidler, M. Rogul, pp. 559-78. New York: McGraw-Hill Frey, P., Prior, P., Trigalet-Demery, D., Marie, C., Kotoujansky, A., Trigalet, A. 1993. Advances in biological control of tomato bacterial wilt using genetically engineeredavirulent mutants of Pseudomonas solanacearum. In Proc. Int. Conf. Plant Pathogenic Bacteria, 8th, Versailles, FranceIn press Furukawa, K., Suzuki, H. 1988. Gene manipulationof catabolic activities for production of intermediates of various biphenyl compounds.Appl. Microbiol. Biotechnol. 29:363 69 Gaertner, F. 1990. Cellular delivery systemsfor insecticidal proteins: living and non-living microorganisms. In Controlled Delivery of Crop-Protection Agents, ed. R. M. Wilkins, pp. 245-57. London: Taylor and Francis Galli, E., Silver, S., Witholt, B., eds. 1992. Pseudomonas:Molecular Biology and Technology. FEMS Symp. No. 60. Washington: Am.Soc. Microbiol. Gamela,P. H., Plot, J,-C. 1992. Characterization and properties of a novel plasmid vector for Bacillus thuringiensis displaying compatibility with host plasmids. Gene 120:17-26 Gawron-Burke, C., Banm, J. A. 1991. Genetic manipulation of Bacillus

Annual Reviews www.annualreviews.org/aronline RELEASE

65.

Annu. Rev. Microbiol. 1993.47:913-944. Downloaded from arjournals.annualreviews.org by Dr Mark Wilson on 01/12/06. For personal use only.

66.

67. 68.

69.

70.

70a.

71.

72.

73.

74.

75.

OF RECOMBINANT

MICROORGANISMS

939

thuringiensisinsecticidal crystal protein 1991. Secondary metabolites of Pseugenes in bacteria. In Genetic Engineerdomonasfluorescens strain CHAO ining, ed. J. K. Setlow, 13:237-63. New volvedin suppression of root diseases. York: Plenum In Advances in Molecular Genetics of Gawron-Burke, C., Johnson, T. B. Plant-Microbe Interactions, ed. H. 1993. Development of Bacillus Hennecke, D. P. S. Verma, 1:450-56. thuringiensis based pesticides for the Amsterdam: Kluwer Academic control of potato insect pests. Advances 76. Haggblom, M. 1990. Mechanisms of in Potato Pest Biology and Managebacterial degradation and transformament. fin press tion of chlorinated monoaromaticcomGay, P. A., Saikumar, K. V., Clevepounds. J. Basic Microbiol. 30:115land, T. E., Tuzun, S. 1992. Antag41 onistic effect of chitinolytic bacteria 76a. Halvorson, H. O., Pramer, D., Rogul, against toxin-producing fungi. PhytoM., eds. 1985. Engineered Organisms pathology 82:1074 in the Environment:Scientific Issues. Gillespie, A. T. 1988. Use of fungi Washington, DC: Am. Soc. Microbiol. to conl~rol pests of agricultural import77. Harman, G. E., Stasz, T. E. 1991. ance. See Ref. 22a, pp. 37-60 Protoplast fusion for the production of Goettel, M. S., St. Leger, R. J., superior biocontrol fungi. See Ref. Bhairi, S., Roberts, D. W., Staples, 164b, pp. 171-86 R. C. 1989. Transformation of the 78. Harman,G. E., Taylor, A. G., Stasz, entomopathogenic fungus Metarhizium T. E. 1989. Combining effective strains anisopl~:ae using the benJt3 gene from of Trichoderma harzianum and solid Aspergillus nidulans. J. Cellular matrix priming to improve biological Biochern. Suppl. 13A:170 seed treatments. Plant Dis. 73:631-37 Graham,T. L., Watrud, L. S., Perlak, 79. Hayes, C. K. 1992. Improvement of F. J., Tran, M. T., Lavrick, P. B., Trichoderma and Gliocladium by geet al. 1986. A model genetically ennetic manipulation. See Ref. 171a, pp. gineered pesticide: cloning and expres277-86 sion of lhe Bacillus thuringiensis subsp. 80. Heale, J. B. 1988. The potential imkurstaki ~-endotoxin into Pseudomonas pacts of fungal genetics and molecular fluorescens. In Recognition in Microbebiology on biological control with parPlant Symbiotic and Pathogenic Interticular reference to entomopathogens. actions, ed. B. Lugtenberg, pp. See Ref. 22a, pp. 211-34 385-93. Berlin: Springer-Verlag 81. Hirano, S. S., Demars, S. J., Morris, Greaves, M. P., Bailey, J. A., HarC. E. 1981. Survival, establishment greaves, J. A. 1989. Mycoherbicides: and dispersal of Pseudomonassyringae opportunities for genetic manipulation. on snap beans (Phaseolus vulgaris). Pestic. Sci. 26:93-101 Phytopathology 71:881 Gresshoff, P. M., Roth, L. E., Stacey, 82. Hirano, S. S., Willis, D. K., Upper, G., Newton, W. E., eds. 1990. NiC. D. 1992. Population dynamics of trogen f’ixation: Achievementsand Oba Tn5-induced non-lesion forming mujectives. NewYork: Chapmanand Hall tant of Pseudomonas syringae pv. Gullino, M. L., Garibaldi, A. 1988. syringae on bean plants in the field. Biological and integrated control of Phytopathology 82:1067 grey mould of grapevine: results in 83. Hirsch, A. M., Wilson, K. J., Jones, Italy. EI~POBull. 18:9-12 J. D. G,, Bang, M., Walker, V. V., Gutterson, N. 1990. Microbial fungiAusubel, F. M. 1984. Rhizobium cides: recent approachesto elucidating meliloti nodulation genes allow mechani:~ms.CRCRev. Biotechnol. 10: Agrobacterium tumefaciens and Es69-91 cherichia coli to form pseudonodules Guttersoa, N., Howie, W., Suslow, on alfalfa. J. Bacteriol. 158:1133-43 T. 199(I. Enhancing efficiencies of 84. Hodgson, A. L. M., Stacey, G. 1986. biocontrol agents by use of biotechPotential for Rhizobium improvement. nology. See Ref. 9a, pp. 749-65 Crit. Rev. Biotechnol. 4:1-74 Gutterson, N. I., Layton, T. J., Ziegle, 85. Hofte, H., Whiteley, H. R. 1989. Insecticidal crystal proteins of Bacillus J. S., Warren, G. J. 1986. Molecular cloning of genetic determinants for thuringiensis. Microbiol. Rev. 53:242inhibition of fungal growth by a flu55 orescent pseudomonad.J. Bacteriol. 86. Hollander, A. K. 1991. Environmental 165:696-703 impacts of genetically engineered miHaas, D., Keel, C., Laville, J., crobial and viral biocontrol agents. See Maurhofi;r, M., Oberhansli, T., et al. Ref. l17b, pp. 251-66

Annual Reviews www.annualreviews.org/aronline

Annu. Rev. Microbiol. 1993.47:913-944. Downloaded from arjournals.annualreviews.org by Dr Mark Wilson on 01/12/06. For personal use only.

940

WILSON

& LINDOW

87. Howie, W. J., Suslow, T. V. 1991. Role of antibiotic biosynthesis in the inhibition of Pythium ultimum in the cotton spermosphere and rhizosphere by Pseudornonas fluorescens. Mol. Plant-Microbe Interact. 4:393-99 88. Jones, D. A., Ryder, M. H., Clare, B. G., Fan-and, S. K., Kerr, A. 1988. Construction of a Tra- deletion mutant of pAgK84to safeguard the biological control of crowngall. Mol. Gen. Genet. 212:207-14 89. Jones, J. D. G., Grady, K. L,, Suslow, T. V., Bedbrook, J. R. 1986. Isolation and characterization of genes encoding two chitinase enzymes from Serratia marcescens. EMBOJ. 5:467-73 90. Kearns, L. P., Mahanty, H. K. 1993. Identification and cloning of Erwinia herbicola DNAresponsible for suppression of Erwinia amylovora. Acta Hortic. In press 91. Keel, C., Schnider, U., Maurhofer, M., Voisard, C., Laville, J., et 1992. Suppression of root diseases by Pseudomonas fluorescens CHA0: importance of the bacterial secondarymetabolite 2,4-diacetylphloroglucinol. Mol. Plant-Microbe Interact. 5:z1-13 92. Kistler, H. C. 1991. Genetic manipulation of plant pathogenic fungi. See Ref. 164b, pp. 152-69 92a. Klement, Z., ed. 1990. Proceedings of the Seventh International Conference on Plant Pathogenic Bacteria. Budapest: Akedimia Kiado 93. Kluepfel, D. A., Kline, E. L., Hughes, T., Skipper, H., Gooden, D., et al. 1990. Field testing of a genetically engineered rhizosphere inhabiting pseudomonad:development of a model system. See Ref. l17a, pp. 189-99 94. Kluepfel, D. A., Kline, E. J., ~Skipper, H. D., Hughes, T. A., Gooden, D. T., et al. 1991. Therelease and tracking of genetically engineered bacteria in the environment. Phytopathology 81: 348-52 95. Kluepfel, D. A., Tonkyn, D. W. 1990. Release of soil-borne genetically modified bacteria: biosafety implications from contained experiments. See Ref. l17a, pp. 55-65 96. Kluepfel, D. A., Tonkyn, D. W. 1992. The ecology of genetically altered bacteria in the rbizospbere. See Ref. 171a, pp. 407-13 97. Kostka, S. J. 1990. The design and execution of successive field releases of genetically engineered microorganisms. See Ref. ll7a, pp. 167-76 98. Kostka, S. J., Reeser, P. W., Miller,

99.

100.

101.

102.

103.

104. 105.

106.

107.

108.

109.

D. P. 1988. Experimental host range of Clavibacter xyli subsp, cynodontis (CXC)and CXC/Bacillus th uringiensis recombinant (CXC/BT). Phytopathology 78:1540 Kostka, S. J., Tomasino,S. F., Turner, J. T., Reeser, P. W. 1988. Field release of a transformed strain of Clavibacter xyli subsp, cynodontis containing a delta-endotoxin gene from Bacillus thuringiensis subsp, kurstaki (BT). Phytopathology 78:1540 Leahy, J. G,, Colwell, R. R. 1990. Microbial degradation of hydrocarbons in the environment. Microbiol. Rev. 54:305-15 Lehrbach, P. R., Zeyer, J., Reineke, W., Knackmuss, H.-J., Timmis, K. N. 1984. Enzymerecruitment in vitro: use of cloned genes to extend the range of haloaromatics degraded by Pseudomonassp. strain B 13. J. Bacteriol. 158:1025--32 Lereclus, D., Vallade, M., Chafaux, J., Arantes, O., Rambaud, S. 1992. Expansionof insecticidal host range of Bacillus thuringiensis by in vivo genetic recombination. Bio/Technology 10:41821 Levy, E., Gough, F. J., Berlin, K. D., Guiana, P. W., Smith, T. J. 1992. Inhibition of Septoria tritici and other phytopathogenic fungi and bacteria by Pseudornonasfluorescens and its antibiotics. Plant Pathol. 41:335-41 Deleted in proof Lindemann, J., Suslow, T. V. 1987. Competition between ice nucleationactive wild type and ice nucleationdeficient deletion mutant strains of Pseudomonas syringae and Pseudomonasfluorescens biovar I and biological control of frost injury on strawberry blossoms. Phytopathology 77:882-86 Lindemann, J., Suslow, T. V. 1987. Characteristics relevmxt to the question of environmental fate of generically engineered INA-deletion mutant strains of Pseudomonas, Curr. Plant. Sci. Biotechnol. Agric. 4:1005--12 Lindow, S. E. 1983. The role of bacterial ice nucleation in frost injury to plants. Annu. Rev. Phytopathol. 21: 363-84 Lindow, S. E.-19857Ecology of Pseudomonassyringae relevant to the field use of Ice- deletion mutants constructed in vitro for plant frost control. See Ref. 76a, pp. 23-35 Lindow,S. E. 1986. In vitro construction of biological control agents. In BCPCMonographNo. 34. Biotechnology and Crop Improvement and Pro-

Annual Reviews www.annualreviews.org/aronline RELEASE

110. 11I.

112.

Annu. Rev. Microbiol. 1993.47:913-944. Downloaded from arjournals.annualreviews.org by Dr Mark Wilson on 01/12/06. For personal use only.

113.

114.

115.

116.

117.

l17a.

l17b. 118.

119.

120. 121.

OF RECOMBINANT

tection,, pp. 185-98. London:Br. Crop Prot. Council Lindow, S. E. 1989. Genetic engineering: of bacteria from managedand natural habitats. Science 244:1300-7 Lindow,S. E. 1990. Design and results of field tests of recombinantIce- Pseudomonas syringae strains. See Ref. 118, pp. 61-69 Lindow,S. E. 1990. Use of genetically altered bacteria to achieve plant frost control. See Ref. 124a, pp. 85-110 Lindow, S. E. 1990. Environmental use of genetically engineered organisms. ]in Advances in Biotechnology, ed. E. Heseltine, pp. 101-14. Stockholm: AB Boktryck HBG Lindow, S. E. 1992. Ice- strains of Pseudomonas syringae introduced to control ice nucleation active strains on potato. See Ref. 171a, pp. 169-74 Lindow, S. E. 1992. Environmental release of pseudomonads: potential benefits mad risks. See Ref. 62a, pp. 399-407 Lindow, S. E., Knudsen, G. R., Seidler, R. J,, Walter, M. V., Lambou, V. W., et al. 1988. Aerial dispersal and epiphytic survival of Pseudomonas syringae duringa pretest for the release of genetically engineered strains into the environment. Appl. Environ. Microb~’ol. 54:1557-63 Lindow, S. E., Panopoulos, N. J. 1988. Field test of recombinant IcePseudomonassyringae for biological frost control in potato. See Ref. 164a, pp. 121-38 MacKenzie,D. R., Henry, S. C., eds. 1990. Biological Monitoring of Genetically EngineeredPlants and Microbes. Bethesda: Agricultural ResearchInstitute Maramorosch,K., ed. 1991. Biotechnology for Biological Control of Pests and Vectors. Boca Raton: CRCPress Marois, J., Bruening, J. 1990. Proceeding:~ of the International Conference on Risk Assessmentin Agricultural Biotechnology. Oakland: Univ. Calif. Press Maurhofer, M., Keel, C., Schnider, U., Voiisard, C., Haas, D., Defago, G. 1992. Influence of enhanced antibiotic production in Pseudomonas fluorescens strain CHA0on its disease suppressive capacity. Phytopathology 82:190-’95 McCoy, C. W. 1990. Entomogenous fungi as microbial pesticides. See Ref. 9a, pp. 139-59 Mclver, J., Djordevic, M. A., Weinman, J. J., Bender, G. L., Rolfe, B.

MICROORGANISMS

941

G. 1989. Extension of host range of Rhizobium leguminosarumby. trifolii caused by point mutations in nodD that result in alterations in regulatory function and recognition of inducer molecules. Mol. Plant-Microbe Interact. 2:97-106 122. Miller, R. V., Levy, S. B. 1989. Horizontal gene transfer in relation to environmental release of genetically engineered microorganisms. In Gene Transfer in the Environment, ed. S. B. Levy, R. V. Miller, pp. 405-20. New York: McGraw-Hill 123. Millis, N. F. 1992. Australian experience with the release of microorganisms. In Proc. Int. Syrup. on the Biosafety Results of Field Tests of Genetically Modified Plants and Microorganisms, 2nd, Goslar, Germany. (Abstr.) 124. Mulbry, W., Keamey, P. C. 1991. Degradation of pesticides by microorganisms and the potential for genetic manipulation. Crop Prot. 10: 334-46 124a. Nakas, J. P., Hagedorn,C., eds. 1990. Biotechnology of Plant-Microbe Interactions. NewYork: McGraw-Hill 125. Norelli, J. L., Gilbert, M, T., AIdwinckle, H, S., Zumoff, C. H., Beer, S. V. 1990. Population dynamics of nonpathogenic mutants of Erwinia amylovora in apple host tissue. Acta Hortic. 273:239-40 126. Nusslein, K., Maris, D., Timmis, K., Dwyer, D. F. 1992. Expression and transfer of engineered catabolic pathways harbored by Pseudomonas spp. introduced into activated sludge microcosms. Appl. Environ. Microbiol. 58: 3380-86 127. Obukowicz, M. G., Perlak, F. L, Kusano-Kretzmer, K., Mayer, E. J., Bolten, S. L., Watrud, L. S. 1986. Tn5-mediatedintegration of the deltaendotoxin gene from Bacillus thuringiensis into the chromosome of root-colonizing pseudomonads.J. Bacteriol. 168:982-89 128. Olson, B. H., Ogunseitan, O. A., Rochelle, P. A., Tebbe, C. C., Tsai, Y. L. 1991. The implications of horizontal gene transfer for the environmental impact of genetically engineered microorganisms. In Risk Assessment in Genetic Engineering, ed. M. A. Levin, H. S. Strauss, pp. 163-88. NewYork: McGraw-Hill 129. Oppenheim, A. B., Chet, I. 1992. Clonedchitinases in fungal plant-pathogen control strategies. Trends Biotechnol. 10:392-94

Annual Reviews www.annualreviews.org/aronline

Annu. Rev. Microbiol. 1993.47:913-944. Downloaded from arjournals.annualreviews.org by Dr Mark Wilson on 01/12/06. For personal use only.

942

WILSON

& LINDOW

130. Orser, C. S., Staskawicz, B. J., Panopoulos, N. J., Dahlbeck, D., Lindow, S. E. 1985. Cloning and expression of bacterial ice nucleation genes in Escherichia coil J. Bacteriol. 164:359-66 131. Paau, A. S. 1991. Improvement of Rhizobiuminoculants by mutation, genetic engineering and formulation. Biotechnol. Adv. 9:173-84 132. Panopoulos, N. J., Peet, R. C. 1985. The molecular genetics of plant pathogenic bacteria. Annu.Rev. Phytopathol. 23:381--419 133. Pantuwatana, S. 1991. Control of mosquito vectors by genetically engineered Bacillus thuringiensis and B. sphaericus in the tropics. See Ref. l17b, pp.14% 61 134. Papavizas, G. C., Lewis, J. A., Abd-El Moity, T. H. 1982. Evaluation of new biotypes of Trichodermaharzianumtolerant to benomyl and enhanced biocontrol capabilities. Phytopathology 72:126-32 135. Parke, J. L., Zablotowicz, R. M., Rand, R. E. 1992. Tracking a lacZYmarked strain of Pseudomonasfluorescens in a Wisconsin field trial. Phytopathology 82:1177-78 136. Prunier, J. P., Kostka, S. J., Labonne, G. 1990. Clavibacter xyli subsp, cynodontis: an endophytic bacterium from com. See Ref. 92a, pp. 305-10 137. Reeser, P. W., Kostka, S. J. 1988. Population dynamicsof Clavibacter xyli subsp, cynodontis (Cxc) and Cxc/Bacillus thuringiensis subsp, kurstaki (Bt) recombinant in corn (Zea mays). Phytopathology 78:1540 138. Reineke, W., Knackmuss, H.-J. 1988. Microbial degradationof haloaromatics. Annu. Rev. Microbiol. 42:263-87 139. Robbins, P. W., Albright, C., Benfield, B. 1988. Cloning and expression of a Streptomyces plicatus chitinase (chitinase-63) in Escherichia coli. J. Biol. Chem. 263:443-47 140. Roberts, D. P., Berman, P. M., Allen, C., Stromberg, V. K., Lacy, G. H., Mount, M. S. 1986. Erwinia carotovora: molecular cloning of a 3.4 kilobase DNAfragment mediating production of pectate lyases. Can. J. Plant Pathol. 8:17-27 141. Ronson, C. W., Bosworth, A., Genova, M., Gudbransen, S., Hankinson, T., et al. 1990. Field release of genetically-engineered Rhizobiummeliloti and Bradyrhizobium japonicum strain. See Ref. 70a, pp. 397-403 142. Sands, D. C., Ford, E. J., Miller, R. V. 1990. Genetic manipulation of broad

143.

144.

145.

146.

147.

148.

149.

150.

151.

152.

153.

154.

host-range fungi for biological control of weeds. Weed Technology 4:471-74 Sayler, G. S., Hooper, S. W., Layton, A. C., King, J. M. H. 1990. Catabolic plasmids of environmental and ecological significance. Microb. Ecol. 19:120 Sayre, P. G. 1990. Assessment of genetically engineered microorganisms under the Toxic Substances Control Act: considerations prior to small-scale release. See Ref. 70a, pp. 405-15 Sayre, P., Miller, R. V. 1991. Bacterial mobile genetic elements: importance in assessing the environmental fate of genetically engineered sequences. Plasmid 26:151-71 Seidler, R, 1988. EPASpecial Report: release of Ice-minus recombinant bacteria at California test sites. Environmental Research Laboratory, Corvallis, Oregon Seidler, R. J. 1992. Evaluation of methodsfor detecting ecological effects from genetically engineered microorganisms and microbial pest control agents in terrestrial ecosystems. Biotechnol. Adv. 10:14%78 Shapira, R., Ordentlich, A., Chet, I., Oppenheim, A. B. 1989. Control of plant diseases by chitinases expressed from cloned DNAin Escherichia coli. Phytopathology 79:1246-49 Shaw, J. J., Beauchamp, C., Dane, F., Kriel, R. J. 1992. Securing a permit from the United States Departmentof Agriculture for field workwith genetically engineered microbes: a nonprohibitory process. Microb. Releases 1:51-53 Shaw, J. J., Dane, F., Geiger, D., Kloepper, J. W. 1992. Use of bioluminescencefor detection of genetically engineered microorganisms released into the environment. Appl. Environ. Microbiol. 58:267-73 Shields, M. S., Reagin, M. J. 1992. Selection of a Pseudomonas cepacia strain constitutive for the degradation of trichloroethylene. Appl. Environ. Microbiol. 58:3977-83 Shim, J.-S., Farrand, S. K., Kerr, A. 1987. Biological control of crowngall: construction and testing of new biocontrol agents. Phytopathology77:46366 Sivan, A., Stasz, T. E., Hemmat,M., Hayes, C. K., Harman, G. E. 1992. Transformation of Trichoderma spp. with plasmids conferring hygromycin B resistance. Mycologia 84:687-94 Skot, L., Harrison, S. P., Nath, A., Mytton, L. R., Clifford, B. C. 1990.

Annual Reviews www.annualreviews.org/aronline

Annu. Rev. Microbiol. 1993.47:913-944. Downloaded from arjournals.annualreviews.org by Dr Mark Wilson on 01/12/06. For personal use only.

RELEASE

OF RECOMBINANT

Expression of insecticidal in Rhizobium contain,ing the ~-endotoxingene cloned from Bacillus thuringiensis subsp, tenebrionis. Plant Soil 27:285-95 155. Smit, E., van Elsas, J. D., van Veen, J. A. ]t992. Risks associated with the application of genetically modifiedmicroorganismsin terrestrial ecosystems. FEMSMicrobiol. Rev. 88:263-78 156. Spaink, H. P., Okker,R. J. H., Wijffelman, C. A., Tak, T:, Goosen-de Roo, L., et al. 1989. Symbiotic properties of Rhizobia containing a flavonoid-independent hybrid nodD product. J. Bacteriol. 171:4045-53 157. Stacey, G. 1985, The Rhizobium experience. See Ref. 76a, pp. 109-21 158. Stacey, G., Upchurch, R. G. 1984. Rhizobium inoculation of legumes. Trends Biotechnol. 2:65-70 159. Stasz, T. E. 1990. Genetic improvementof fungi by protoplast fusion for biological control of plant pathogens. Can. J. Plant Pathol. 12:322-27 159a. Stewart-Tull, D. E. S., Sussman, M., eds. 19’92. The Release of Genetically Modified Microorganisms--REGEM2. New York: Plenum 160. Stock, C. A., McLoughlin, T. J., Klein, J. A., Adang, M. J. 1990. Expression of a Bacillus thuringiensis crystal protein gene in Pseudomonas cepacia 526. Can. J. Microbiol. 36: 879-84 161. Stromber, V. K., Orvos, D. R., Scanferlato, V. S. 1990. In planta competition amongcell-degrading enzymemutants and wild-type strains of Erwinia carotovora. See Ref. 92a, pp. 721-25 162. Sundheim, L. 1990. Biocontrol of Fusariurn oxysporumwith a chitinase encoding gene from Serratia marcescens on a stable plasmid in Pseudornonas. J. Cell. Biochern. Suppl. 13A:171 163. Sundheim,L. 1992. Effect of chitinase encoding genes in biocontrol Pseudomonas spp. See Ref. 171a, pp. 331-33 164. Sundin, G. W., Jones, A. L., Fulbright, D. W. 1989. Copper resistance in Pseudomonas syringae pv. syringae from cherry orchards and its associated transfer in vitro with a plasmid. Phytopathoh~gy 79:861-65 164a. Sussman, M., Collins, C. H., Skinner, F. A., Stewart-Tull, D. E., eds. 1988. The Release of Genetically-Engineered Micro-Organisms. London: Academic 164b. Te Beest, D. O., ed. 1991. Microbial Control of Weeds. NewYork: Chapman and Hall 165. Te Beest, D. O., Yang, X. B., Cisar, C. R. 1992. The status of biological

MICROORGANISMS

943

control of weedswith fungal pathogens. Annu. Rev. Phytopathol. 30:637-57 166. Templeton, G. E., Heiny, D. K. 1990. Mycoherbicides. In NewDirections in Biological Control: Alternatives for SuppressingAgricultural Pests and Diseases, pp. 279-86. NewYork: Liss 167. Thanabalu, T., Hindley, J., Brenner, S., Oei, C., Berry, C. 1992. Expression of the mosquitocidaltoxins of Bacillus sphaericus and Bacillus thuringiensis subsp, israelensis by recombinant Caulobacter crescentus, a vehicle for biological control of aquatic insect larvae. Appl. Environ. Microbiol. 58:90510 168. Thomashow, L. S., Weller, D. M. 1988. Role of a phenazine antibiotic from Pseudomonasfluorescens in biological control of Gauemannomyces graminis var. tritici. J. Bacteriol. 170:3499-3508 169. Thomashow, L. S., Weller, D. M. 1990. Application of fluorescent pseudomonads to control diseases of wheat and some mechanismsof disease suppression. In Biological Control of Soil-Borne Plant Pathogens, ed. D. Hornby, pp. 109-22. Wallingford, UK: CABInternational 170. Thomashow, L. S., Weller, D. M., Bonsall, R. F., Pierson, L. S. III. 1990. Productionof the antibiotic phenazine-l-carboxylic acid by fluorescent Pseudomonas species in the rhizosphere of wheat. Appl. Environ. Microbiol. 56:908-12 171. Tiedje, J. M., Colwell, R. K., Grossman, Y. L., Hodson, R. E., Lenski, R. E., et al. 1989. The planned introduction of genetically engineeredorganisms: ecological considerations and recommendations. Ecology 70:298-315 171a. Tjamos, E. C., Papavizas, G. C., Cook, R. J., eds. 1992. Biological Control of Plant Diseases: Progress and Challenges for the Future. New York: Plenum 172. Tolin, S. A., Vidaver, A. K. 1989. Guidelines and regulations for research with genetically modified organisms: a view from academe. Annu. Rev. Phytopathol. 27:551-81 173. Tomasino, S. F., Beach, R. M., Leister, R. J. 1992. Control of European corn borer in field corn using a genetically engineered endophyte. Phytopathology 82:1128 174. Trigalet, A., Demery,D. 1986. lnvasiveness in tomato plants of Trd-induced avirulent mutants of Pseudomonas solanacearum. Physiol. Mol. Plant Pathol. 28:423-30

Annual Reviews www.annualreviews.org/aronline

Annu. Rev. Microbiol. 1993.47:913-944. Downloaded from arjournals.annualreviews.org by Dr Mark Wilson on 01/12/06. For personal use only.

944

WILSON

& LINDOW

175. Tdgalet, A., Trigalet-Demery, D. 1990. Use of avirulent mutants of Pseudomonassolanacearum for the biological control of bacterial wilt of tomato plants. Physiol. Mol. Plant Pathol. 36:27-38 176. Triplett, E. W. 1990. Construction of a syrobiotically effective strain of Rhizobiumlegurninosarumbv. trifolii with increased nodulation competitiveness. Appl. Environ. Microbiol. 56:98-103 177. Triplett, E. W., Barta, T. M. 1987. Trifolitoxin production and nodulation are necessary for the expression of superior nodulation competitiveness by Rhizobium leguminosarumbv. trifolii strain T24 on clover. Plant Physiol. 85:335-42 178. Triplett, E. W., Sadowsky, M. J. 1992. Genetics of competition for nodulation of legumes. Annu. Rev. Microbiol. 46:399-428 179. Tronsmo, A. 1992. Leaf and blossom epiphytes and endophytesas biological control agents. See Ref. 171a, pp. 43--54 180. Truchet, G., Rosenberg, G., Vasse, ,1., Julliot, J.-S., Camut,S., Denarie, J. 1984. Transfer of Rhizobiummeliloti pSym genes into Agrobacterium tumefaciens host-specific nodulation by atypical infection. J. Bacteriol. 157: 134-42 181. Turner, J. T., Lampel, J. S., Stearman, R. S., Sundin, G. W., Gunyuzlu, P., Anderson,I. J. 1991. Stability of the B-endotoxin gene from Bacillus tharingiensis subsp, kurstaki in a recombinant strain of Clavibacter xyli subsp, cynodontis. Appl. Environ. Microbiol. 57:3522-28 182. van Elsas, J. D., Heijnen, C. E., van Veen, J. A. 1990. The fate of introduced genetically engineered microorganisms(GEM’s)in soil, in microcosm, andfield: impactof soil textural aspects. See Ref. 117a, pp. 67-79 183. Vanneste, J. L., Yu, J., Beer, S. V. 1992. Role of antibiotic production by Erwinia herbicola Eh252in biological control of Erwinia amylovora. J. Bacteriol. 174:2785-96 184. Vicedo, B., Penalver, R., Asins, M. J., Lopez, M. M. 1993. Biological control of Agrobacteriumtumefaciens, colonization, and pAgK84transfer with Agrobacterium radiobacter and the Tra- mutant strain K1026. Appl. Environ. Microbiol. 59:309-15 185, Vincent, M. N., Harrison, L. A.,

186.

187.

188.

189. 190.

191.

192.

193.

194.

195.

196.

Brackin, J. M., Kovacevich, P. A., Mukerji, P., et al. 1991. Genetic analysis of the antifungal activity of a soilborne Pseudornonas aureofaciens strain. Appl. Environ. MicrobioL 57:2928-34 Waalwijk, C., Dullemans, A., Maat, C. 1991. Construction of a bioinsecticidal rhizosphere isolate of Pseudomonas fluorescens. FEMSMicrobiol. Lett. 77:257-64 Weller, D. M. 1988. Biological control of soilborne plant pathogens in the rhizosphere with bacteria. Annu. Rev. Phytopathol. 26:379-407 Weller, D. M., Saettler, A. W. 1978. Rifampin-resistant Xanthomonas phaseoli var. fuscans and Xanthomonas phaseoli: tools for study of bean blight bacteria. Phytopathology 68:778-81 Wilcox, H. E. 1990. Mycorrhizal associations. See Ref. 124a, pp. 227-55 Williamson, M. 1992. Environmental risks from the release of genetically modified organisms (GMOs)---the need for molecular ecology. Mol. Ecol. 1:3-8 Willis, D. K., Rich, J. R., Hrabak, E. M. 1991. hrp genes of phytopathogenic bacteria. Mol. Plant,Microbe Interact. 4:132-38 Winter, R. B., Yen, K., Ensley, B. D. 1989. Efficient degradation of trichlorethylene by a recombinant Escherichia coll. Bio/Technology 7:28285 Wortman, A. T., Somerville, C. C., Colwell, R. R. 1986. Chitinase determinants of Vibrio vulnificus: gene cloning and applications of a chitinase probe. Appl. Environ. Microbiol. 52: 142-45 Wynne,E. C., Pemberton, J. M. 1986. Cloning of a gene cluster from Cellvibrio mixtus which encodes for cellulase, chitinase, amylase, and pectinase. Appl. Environ. Microbiol. 52:1362-67 Yamamoto, T., Powell, G. K. 1993. Bacillus thuringiensis crystal protein: recent advancements in understanding the insecticidal activity. In Advanced Engineered Pesticides, ed. L. Kim. NewYork: Marcel Dekker In press Yoder, O. C., Weltring, K., Turgeon, B. G., Garber, R. C., Van Etten, H. D. 1986. Technology for molecular cloning of fungal virulence genes. In Biology and Molecular Biology of Plant-PathogenInteractions, ed. J. Bailey, pp. 371-84. NewYork: Plenum

Annu. Rev. Microbiol. 1993.47:913-944. Downloaded from arjournals.annualreviews.org by Dr Mark Wilson on 01/12/06. For personal use only.

Annu. Rev. Microbiol. 1993.47:913-944. Downloaded from arjournals.annualreviews.org by Dr Mark Wilson on 01/12/06. For personal use only.