If, .2"- .v L—Ma! '. .- ' -i .0 .'_ --4 l ._.. _ _ O V ‘e.’ .14 A a g "a . I.’I-'!Al‘_fl'l. 1:15-- I . L... v infant-v This is to certify that the dissertation entitled Molecular Cloning of the Polysaccharide Depolymerase Gene of Bacteriophage PEal(h) and Expression in Erwinia amylovora presented by John Stephen Hartung has been accepted towards fulfillment of the requirements for Doctorate Plant Pathology degree in Botany and Plant Pathology fl\\ tat V‘- ( Major professor and J. Klos Date June 25, 1985 MS U it an Affirmative Action/Equal Opportunity Institution 0-12771 MSU LIBRARIES .-;:—. RETURNING MATERIALS: Place in book drop to remove this checkout from your record. FINES will be charged if book is returned after the date stamped below. CLONING OF THE POLYSACCHARIDE DEPOLYMERASE GENE OF BACTERIOPHAGE PEa1(h) AND ITS EXPRESSION IN ERWINIA AMYLOVORA BY John Stephen Hartung A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Botany and Plant Pathology 1985 ABSTRACT CLONING OF THE POLYSACCHARIDE DEPOLYMERASE GENE FROM BACTERIOPHAGE PEal(H) AND ITS EXPRESSION IN ERWINIA AMYLOVORA BY John Stephen Hartung A bacteriophage gene which encoded a polysaccharide depolymerase (PD) specific for the surface polysaccharides of Erwinia amylovora was cloned and expressed in E2991}: The cloned gene was also used to study the controversial role of these surface polysaccharides in pathogenesis of g; amylovora. Bacteriophage PEa1(h) produced clear plaques surrounded by translucent haloes when it infected encapsulated strains of Erwinia amylovora. The haloes were caused by a soluble polysaccharide depolymerase (PD) associated with phage in- fection” The PD gene was cloned in Escherichia coli from the ds DNA phage genome using the JM105(pUC8) system. A new 5.85 kbp plasmid, pJH94, was found in JMlOS clone 94. The production of PD by strain JM105(pJH94) was confirmed with an in gitrg assay which used purified extracellular polysac- charides (EPS) prepared from EallOR as substrate. Southern blotting experiments confirmed that the cloned PD gene was of phage, not bacterial, origin. Purified plasmid pJH94 transformed EallOR to ampicillin resistance, with a concomitant loss of fluidal colony mor- phology; Production of PD by Ea110R(pJH94) was confirmed John Stephen Hartung with the in zitgg assay. Chemical analysis showed that EallORlpJH94) produced less slime and capsular polysaccha- rides, and that the polysaccharides were of lower molecular weight than the polysaccharides produced by EallOR or EallORlpUCB). The necrotic lesions incited by EallORlpJH94) in immature pear fruits did not produce ooze as did lesions incited by EallOR or Ea110R(pUC8). The results suggest that the disease, but not the ooze production characteristic of the disease, can occur in the presence of an enzyme which can depolymerize the EPS. It must be emphasized however that we do not know if the extracellular polysaccharides are totally degraded in 3112, nor do we know whether or not the oligosaccharide products of the depolymerase enzyme have biological activity. These questions must be addressed be- fore firm conclusions regarding the role of extracellular polysaccharides in pathogenicity and virulence can be made. ACKNOWLEDGEMENTS I would like to thank Dr.ILJ} Klos for providing the financial support and the freedom which allowed me to pursue my researtfin I would also like to express my appreciation for Dr. ELW. Fulbright and the members of his laboratory for their useful suggestions and for providing me with a great place to work. Drs. F. B. Dazzo and C.R. Somerville deserve special acknowledgement for the several useful suggestions which facilitated the research.Ihu Dazzo should also be commended for constructive criticism in the course of my research, and in the preparation of this dissertation. Lastly, and most importantly, I thank Anne and Chloe for putting up with me, and for providing essential emotional support, which I'm sure has not been an easy task. ii TABLE OF CONTENTS LIST OF TABLES . . . . . . . . . . . . . . . . . . LIST OF FIGURES . . . . . . . . . . . . . . . . . . LITERATURE REVIEW . . . . . . . . . . . . . . . . . . CHAPTER I MOLECULAR CLONING OF THE POLYSACCHARIDE DEPOLYMERASE GENE OF BACTERIOPHAGE PEa1(h) IN ESCHERICHIA COLI INTRODUCTION 0 O O O O O O O O O O O O O O O O O O 0 MATERIALS AND METHODS I O O O O O O C O O O O O O O 0 Bacterial and phage cultures .. . .. . .. . Media . .... . . . .. . . .. . . .. . . . Sources of enzymes and reagents .. .. . .. Purification of phage and phage DNA .. .. . Purification of plasmid pUC8 DNA .. .. .. . Restriction endonuclease digestion of phage PEa1(h) DNA .. . . . . . . . . 5' Terminal dephosphorylation of pUC8 DNA . . Ligation of pUC8 and PEa1(h) fragments .. .. . Transformation of bacterial strains with plasmid DNA .. . .. . .. . .. . .. . . Screening of E3911 transformants . . . . . . . Effect of chloroform vapor on the detection of haloes surrounding JMlOS c10ne94 ... .. Rapid small scale isolation of plasmid DNA and detection of plasmid pJH94 in JMlOS clone 94 .. . .. . .. . .. . . Preparation of exopolysaccharides produced in vitro, and in vitro assay for polysaccharide depolymerase .. . .. . .. . .. . .. . Preparation of EPS from immature pear fruits infected with EallOR .. . .. . .. . .. Preparation of soluble polysaccharide depolymer- ase from PEa1(h) lysates of EallOR and from HBlOl(pJH94) .. . . .. . . .. . . .. . iii 15 19 19 19 21 21 22 23 25 26 26 27 28 28 29 3O 31 Page Effect of pH on the activity of soluble polysac- charide depolymerases . .. . .. . .. .. 32 Purification of pJH94 .. . .. . .. . .. . . 33 Extraction of total DNA from EallOR .. .. .. 34 Southeff blotting .. . .. . . . .. . . .. 35 Alpha -P- Labeling of DNA by nick translation . 37 Hybridization of Southern blots .. .. .. .. 39 Autoradiography of nitrocellulose filters .. . 40 RESULTS . . . . . . . . . . . . . . . . . . . . . . . 42 Restriction endonuclease digestion of PEa1(h). . 42 Screening of §_.gg__1__i_ transformants . . . . . . . 42 Effect of chloroform vapor on the detection of haloes surrounding JM105 clone 94. .. .. 46 Detection and size estimation of pJH94 in JMIOS clone 94 .. . .. . .. . .. . .. . .. 46 Detection of polysaccharide depolymerase activity in culture supernatants .. . .. . .. . . 46 Preparation of polysaccharide depolymerases from PEa1(h) lysates of EallOR and from HBlOl(pJH94) o o o o o o o o o o o o o o 50 Effect of pH on the activity of soluble polysaccharide depolymeran . . . . . . .. 50 Hybridization of 2-P-Labled pUC8 and pJH94 to EallOR and PEa1(h) genomes .. . .. . .. 52 DISCUSSION 0 O O O O O O O O O O O O O O O O O O O O 58 CHAPTER II EXPRESSION OF THE CLONED POLYSACCHARIDE DEPOLYMERASE GENE OF BACTERIOPHAGE PEal(h) ERWINIA AMYLOVORA INTRODUCTION . . . . . . . . . . . . . . . . . . . . 64 MATERIALS AND METHODS . . . . . . . . . . . . . . . . 67 Transformation of _E_. amylovora EallOR with plasmids pJH94 and pUC8 . .. . .. . .. . 67 Physical detection of plasmids in El amylovora strains . .... . . .. . .. . .. . .. . 67 Production of polysaccharide by'gh amylovora strains in xitro .. . .. . .. . 68 Chemical analysis of polysaccharides and oligosaccharides . .. . .. . .. . .. . 69 Preparation of polysaccharide depolymerase from Ea110R(pJH94) culture supernatants . . . . 69 Comparison of polysaccharides produced by E; iv Page amylovora strains as substrates for polysac- charide depolymerase .. . .. . .. . .. 70 Test of polysaccharide depolymerase from Ea110R(pJH94) for lyase activity .. .. . 70 Quantification of polysaccharide depolymerase retained by Ea110R(pJH94) and found in LB medium 0 I O O O O O O O O O O O O C C O O 71 Analysis of proteins in culture supernatants of IL amylovora strains . .. .. 72 Determination of the minimum inhibitory concen- trations of CuSOé and antibiotics for strains Ea110R(pUC8) an EallOR (pJH94) .. .. .. 73 Pathogenicity study . .. . .. . .. . . .. . 74 RESULTS 0 O O O O O I O O O O O O O O O O O O O O O O 76 Strains of Erwinia amylovora .. .. .... .. . 76 Physical detection of plasmids in E amylovora strains 0 O O I O O O O O O O O O O O O O O 80 Chemical analysis of polysaccharides and oligosaccharides .. . .. . .. . .. . . 80 Comparison of polysaccharides produced by E; amylovora strains as substrates for polysaccharide depolymerase .. .. .. .. 83 Test of polysaccharide depolymerase from Ea110R(pJH94) for lyase activity.. .. .. 86 Quantification of polysaccharide depolymerase retained by Ea110R(pJH94) and found in LB medium 0 O O O O O O O O O I O O O C I O O 88 Analysis of proteins found in culture superna- tants of E.amylovora strains.... .. .. 88 Determination df-the minimum inhibitory concen- trations of CuSOé and antibiotics on strains Ea110R(pUC8) an Ea110R(pJH94) .. .. .. 91 Pathogenicity study . .. . .. . .. . . .. . 91 DISCUSSION 0 O O O O O O O O O O O O O O O O O O O O 95 APPENDIX INTERACTION OF BACTERIOPHAGE PEal(h) AND ERWINIA AMYLOVORA EallOR IN LIQUID MEDIA INTRODUCTION 0 I O O I O O O O O O O O O O O O O O O 1 03 MATERIALS AND METHODS . . . . . . . . . . . . . . . . 105 Media 0 O O O C O O O O O O O O O O O O O O O 105 Purification of phage . .. . .. . .. . .. 105 Optical density measurements .. .... .. . .. 105 Growth of PEa1(h) in a rich medium .. . .. . . 107 Page Interaction of PEa1(h) and EallOR in defined liquid media .. . .. . .. . .. . .. . 107 Lysis of EallOR by PEa1(h) in DM-l broth . . . 107 Effect of divalent cations on lysis of EallOR by PEa1(h) . . .. . .. . .. . .. . . . 108 RESULTS 0 O O O O O O O O O O O O O O O O O O O O O O 110 Growth of PEalUU in a rich medium .. .. . .. 110 Growth of PEa1(h)iJ1defined liquid media .. . 110 Effect of divalent cations on growth of PEa1(h) I O O O O O O O O O O O O O I O O O 115 DISCUSSION 0 O O O O O O O O O O O O O O O O O O O O 121 BIBLIOGRAPHY O O O O O O O O O O O O O O O C O O O O 124 10. LIST OF TABLES E: 011 strains used in this study .. .. .. Polysaccharide depolymerase activityl recovered from phage PEa1(h) lysates of EallOR and from Sarkosyl lysates (112) of HBlOl(pJH94) . . . . Strains of E: amylovora used in this study . . Quantification of ethanol precipitable polysaccharides produced by gt amylovora strains 0 O O O O O O O O O O O O O O O O O O O Quantification of reducing sugars in ethanol precipitable polysaccharides produced by E; amylovora strains .. . .. . .. . . .. . . . Quantification of uronic acids in ethanol precipitable polysaccharides produced byEamylovora ............... Quantification.of ethanol soluble carbo- hydrates recovered from & amylovora strains . Comparisons of polysaccharides produced by strains of E. amylovora as substrates for polysacchaEide depolymerase .. . .. . .. . . Quantification of PD retained by Ea110R(pJH94) and found in LB medium .. . .. . .. . . . . Media used to study the interaction of PEa1(h) and Ea110R in broth .. . .. . .. . . .. . Page 20 51 77 82 82 84 85 87 89 106 10. LIST OF FIGURES Page PEa1(h) DNA digested by restriction endonuclease Sau3A . . . . . . . . . . . . . 43 PEal(h)[§guBA fragments which were larger than 2kb,pooled after fraction- ation over an.alkaline sucrose density gradient . . . . . . . . . . . . . . . . . . 44 Screening of recombinants for polysaccharide depolymerase activity . .. . .. . .. . . . 45 Effect of chloroform vapor onthe appearance of haloes around JM105(pJH94) . . . .... . . 47 Plasmid pJH94 detected in JMlOS clknue 94 .. 48 Estimation of the size of pJH94 . .. . .. 49 The effect of pH on the activity of solubleepolysaccharide depolymerase isolated from PEa1(h) lysates of Eh amylovora BalloR O O I O O O O O O O O O O O O O O I O O 53 The effect of pH on the activity of polysac- charide depolymerase isolated from sarkosyl lysates of El 3911 HB101(pJH94L.. .. . .. 54 Agarose electrophoresis gel stained with ethidium bromide prior to Southern blotting . . . . . . . . . . . . . . . . . . 55 Autoradiogram of Southern blot which confirms that polysaccharide depolymerase is phage encoded .. .. .. .. .. .. .. 56 \dii Figure Page 11. Plaque morphology of PEa1(h) on E; amylovora strains 0 O O O O O I O O O O O O O O O O C O 78 12. Colonial morphology of E; amylovora strains . 79 13. The physical detection of plasmids in alkaline lysates of E; amylovora strains . . . . . . . 81 14. Analysis of proteins found in culture supernatants of E; amylovora strains . . . . . 9O 15. Symptoms caused by E; amylovora strains on immature pear fruits. . . . . . . . . . . . 93 16. Interaction of E; amylovora strain EallOR and phagePEa1(h)in nutrient broth 0 O O O O O I I O O O O O O O O 111 17. Interaction of strain EallOR and phage PEa1(h) in YureWicz broth 0 O O C I O O O O O O C O O 112 18. Interaction of EallOR and phage PEa1(h) at dif- ferent concentrations of glucose in Yurewicz broth . . .. . .. . .. . .. . 113 19. Interaction of EallOR and phage PEa1(h) at different concentrations of glucose in DM-17 broth 0 o o o o o o o o o o o o o o o 115 20. Effect of growth stage on lysis of EallOR by PEa1(h) in DM-17 medium . . . . . . . . . . 116 21. Effect of growth stage on lysis of Ea11OR by PEa1(h) in DM-17 medium supplemented With lmM caC12 O O O O O O O O I O O O 0 O O 0 117 22. Effect of ZnSO on the lysis of EallOR by PEa1(h) in DM-l medium. UH. DM-17 supplemented with 100 uM ZnSO . (B). DM-17 not supplemente with ZnSO4 .. . .. . .. . 119 23. Apparent synergistic effect of CaC12 and ZnSO on lysis of EallOR by PEa1(h) in D —17 medium a o o o o o o o o o o o o o o 120 LI TERATURE REVI EW Literature Review The phytopathogenic bacteria represent a diverse group of prokaryotes which includes the genera Agrobacterium, Corynebacterium, Erwinia, Pseudomonas, Xanthomonas and Streptomyces. A common feature of these genera is the production of polysaccharide capsules or extracellular poly- saccharides i_n £1132 (21,50,78,85,99) and the appearance of bacteria in a polysaccharide matrix i_n 1139 (56,85,103,125). The role(s) of bacterial surface polysaccharide in plant disease is an example of a subject much studied, yet poorly understood. When Duguid (51) suspended cultured bacterial cells in water and negatively stained them with India ink, two types of extracellular polysaccharides were observed: Those which remained tightly adherent to the cells (capsular polysac- charides or CPS) and those which were dispersed into the water (slime polysaccharides or EPS). The terminology of Duguid will be followed in this work except where explicitly noted. The presence of capsules and slime polysaccharides on phytopathogenic bacteria is not unique since capsules were found surrounding bacteria in such diverse habitats as ani- mal blood (6), bovine rumen fluid (44) and freshwater 2 streams (44). It has been proposed that these polysaccharide capsules aid the bacteria in adhering to surfaces and form- ing colonies (44) and act as ion exchangers to concentrate nutrients near the bacterial surface (44.) Capsular polysac- charides could also protect the bacteria from infection by bacteriophages which require specific membrane proteins or lipopolysaccharides for recognition (89,135). The slime polysaccharide of Xanthomonas phaseolicola produced in vitro was shown to prevent dessication, absorb U;V. light, and increase bacterial viability at high temper- atures (85). These results reinforced conclusions of earlier researchersxflu>studied Erwinia amylovora polysaccharides produced in xizg_(70,110). It is evident therefore that capsular and extracellular polysaccharides can perform a great variety of ecologically useful and adaptive functions. Avery and his colleagues (6) showed that the capsule of Pneumococcus was required for pathogenicity in mice be- cause the presence of the capsule prevented phagocytosis of the bacteria by the white blood cel ls. There has also been a long association of encapsulation with pathogenicity in phytopathogenic genera, particularly Pseudomonas (50,75,78) and Erwinia (7,21) although the nature of the association has remained controversial. A very strong correlation between virulence and the presence of EPS exists in Pseudomonas solanacearum (75,78). Sequeira and his colleagues reported that avirulent, acapsu- lar strains of g; solanacearum were rapidly attached to and enveloped by cell walls in tobacco leaf mesophyll while virulent, encapsulated strains were not attached or envel- oped. The attachment was proposed as a trigger for the induction of host resistance (115). It was also reported that all avirulent, acapsular strains were bound via their lipopolysaccharide (LPS) to a potato lectin. This in vitrg agglutination was blocked by the presence of EPS on virulent isolates, or when EPS was added to suspensions of avirulent cells prior to the potato lectin (116). It was proposed that resistance could be induced in the host after binding of a this binding could be prevented by the presence of bacterial EPS (116). However, when the bacterial agglutinin was puri- fied to apparent homogeneity and chemically characterized, the red blood cell agglutination activity characteristic of potato lectin was lost (83Lmhis work demonstrated that the bacterial agglutinin present in potato tubers was not potato lectin, but rather a hydroxyproline-rich glycoprotein simi- lar to the cell wall protein extensin (83,84L. A different role for capsular polysaccharides has been established for Rhizobium trifolii. Studies have demon- strated an early step in the establishment of a symbiotic infection of the root hair by & trifolii. Trifoliin A, a lectin present on the root hair surface, specifically bound the capsular polysaccharides of Rhizobium trifolii (42) or its oligosaccharide fragments (2), as well as its lipopoly- saccharide (73). In the R; trifollii/clover system however, recognition of the potential symbiont leads to infection; recognition of avirulent _P_. solanacearum via its LPS induces resistance to the bacterium.]31the models proposed for g; solanacearum and R: trifolii, EPS blocks recognition (116) and participates in the recognition event (46,2), respec— tively. The specificity of the attachment and envelopment of heterologous or saprophytic as opposed to homologous, viru- lent bacteria in the leaf meSOphyll was challenged by Hilde- brand et. al. (69) who reported that the observed "entrap- ment films” were not the result of a specific binding event. These workers studied the fate of various Pseudomonas syrin- gae pathovars and Pseudomonas fluorescens in bean, Phaseolus vulgaris, leaf mesophyll. They concluded that the "entrap- ment filnm" were formed non-specifically at air-water inter- faces which resulted from dessication following the intro- duction of bacteria into the mesophyll by infiltration. If the mesophyll remained watersoaked until fixation for elec- tron microscopy, no differential binding between homologous and heterologous bacteria was observed (69). They also ob- served larger entrapment films in younger leaves and pro- posed that the presence of such entrapment films in younger leaves could be correlated to increased susceptibility to bacterial pathogens, by providing a water soaked environment conducive to bacterial multiplication (69,76). El-Banoby and Rudolph (54) reported that EPS produced in vitro by several Pseudomonas syringae pathovars and from Xanthomonas campestris pv;malvacearum induced water soaking when infiltrated into leaves of homologous host plants. No water soaking was induced when EPS was infiltrated into leaf mesophyll of heterologous host plants (54L.The EPS acted as a host selective agent which induced water soaking. These results were later extended.to explain.host resistancelat the cultivar level with & syringae pv.phaseolicola (55,56). It was concluded that the resistance of bean leaves to the persistant water soaking effect of EPS was due to the enzymatic breakdown of EPS infiltrated into resistant leaves (56). Susceptible leaves were unable to enzymatically degrade EPS from a compatible bacterium, and the region of infiltration remained watersoaked because the EPS in the intercellular spaces captured water carried there via the transpiration stream (56). The possible involvement of polysaccharides produced by Erwinia amylovora in the symptomology of fire blight has been controversial. Hildebrand (70) studied the polysaccha- rides found in fire blight ooze and found that a non- specific vascular plugging mechanism induced wilt in pear shoots. When this topic was re-investigated, it was con- cluded that ooze polysaccharide induced wilt in rosaceous but not non-rosaceous plants in the manner of a host-selec- tive toxin (61). The ooze polysaccharide was given the trivial name of "amylovorin’. The role of "amylovorin" as a host-selective toxin has been disputed by others who showed that it acted as a non-specific vascular plugging agent as was originally suggested (69,119). Furthermore, it was re- ported that sensitivity of several plant species to wilting in solutions of “amylovorin” was correlated to shoot flexi- bility, but not to the susceptibility or resistance of the cultivars tested t01§£_amylovora (18). EPS produced in 31339 by numerous isolates was quan- tified and positively correlated with the virulence of the isolates on various susceptible hosts (D, but the quantifi- cation of the EPS was later partially'retracted.(118L.EPS has been reported to prevent agglutination of virulent isol- ates by a factor which was isolated from apple seeds and stems (109). There was no evidence that the factor was a lectin but rather a very basic protein which was postulated to interact with acidic polysaccharides based on an electro- static interaction as has recently been shown to be the case in the g; solanacearum/potato interaction (83,84). An aviru- lent, acapsular strain was more strongly agglutinated in vitrg than was a virulent, encapsulated strain (109). The authors proposed that this was due to the EPS of the‘viru- lent strain binding the agglutination factor and removing it from solution, leaving the encapsulated bacteria in solu- tion. The avirulent strain, producing no EPS, was agglu- tinated via its LPS and removed from solution. This model is basically the same as proposed by Sequei- ra for P; solanacearum, in which agglutination of cel ls is prevented by agglutination of EPS. It was said to be consis- tent with earlier work from the same laboratory which showed that the avirulent strain, but not the virulent strain, was agglutinated in xylem vessels after artificial inoculation (124,125). However, the same lab reported earlier that after natural infection, the pathogen moved exclusively in the phloem (63,87), which would make the significance of the xylem agglutination in these experiments moot. The "anti- agglutination" role for EPS proposed above is also proposed for E; stewartii EPS in corn plants (30). However, the factor which agglutinated avirulent isolates §_._ stewartii was isolated from seed of a corn variety which was very susceptible to the pathogen. Encapsulation can not be the sole virulence determi- nant for §_._ amylovora since an avirulent, fully encapsulated strain has been reported (19,20,25). When this strain was co-inoculated into apple shoots with an avirulent, acapsular strain, typical disease symptoms were observed (19). Similar results were obtained using an in yitrg pear fruit assay in which the acapsular, avirulent strain induced cell leakage, but not ooze (25). The authors concluded that virulence in §h_amylovora was associated with at least two factors: a cell leakage inducing agent and EPS (19,25). The research on bacterial slime and capsular polysac- charides in plant disease reviewed above is an example of a subject much studied, but surrounded with controversy. In the case of §k_amylovora, EPS has been reported to function as a toxin (61), vascular plugging agent (70,119) and anti- agglutination factor (109L.The role of capsular polysaccha- rides in pathogenesis of Pneumococcus was established, in part, by using an enzyme to remove the capsular polysac- charides from a virulent strain prior to a virulence assay (6,49). Analogous experiments performed with plant pathogens could be equally rewarding. It is also interesting to note that resistance to P; syringae pv. phaseolicola.has been associated with the presence of polysaccharide degrading enzymes in the leaf mesophyll (56). Enzymes which remove the capsular and slime poly- saccharides of bacteria have been isolated from heterologous bacteria (6,47). They are frequently associated with bacter- iophage infection of encapsulated bacteria (4,8,10,53,71, 121,137). The presence of such a polysaccharide depolymerase (PD) typically results in the appearance of expanding, translucent haloes surrounding true plaques when bacterio- phage are grown in soft agar overlays. The halo has been observed to result from the action of a_._aeruginosa (10),and &trifolii (8,67,71). Early workers observed that when PD containing lysates were spotted on mature lawns of susceptible bacteria, haloes or shallow "craters” developed. An assay which was used in purification of these enzymes consisted of spotting aliquots of such lysates on lawns in dilution series and observing the development of craters (4,74). Other workers developed in zitrg assays based on the liberation of reducing sugars (10,137), or hexoseamines (10) from polysaccharide substrate upon incubation with the depolymerases. In cases where in- creases in reducing sugar or hexoseamines were not detecta- ble, purification was achieved by quantifying the rate at which.enzyme containing aliquots reduced the viscosity of polysaccharide substrate solutions (8,67,71). The enzymes have been shown to be endo-glycanases (which cleave the glycosidic bond by the addition of a molecule of water) in the majority of cases Ulh22,23,106,137),2but a lyase (which cleaved the glycosid- ic bond by the removal of a molecule of water) has recently been reported (In. The gndg cleavage catalysed by these enzymes has been generally shown to procede until the resul- ting oligosaccharides consisted of 1-3 repeating unit oligo- saccharides (106,137L.This explained the concominant in- crease in reducing sugars and decrease in viscosity observed when polysaccharides were incubated with such enzymes. Enzy- ll matic activity was stimulated by the presence of divalent cations in some studies (67,71). The enzymes have been shown to be very substrate specific, generally hydrolysing only polysaccharide from strains on which the phage associated with the enzyme could reproduce (106,127)aalthough excep— tions have been noted in the Rhizobiaceae (126). In some cases complete hydrolysis of homologous polysaccharide re- quired prior deacetylation (71). The enzymes which have been purified and characterized had molecular weights ranging from 155,000 to 550,000 (10,23,71,105,137) and were composed of several subunits (23,105,137). Enzymatic activity with identical characteris- tics has been found bound to the phage particles in many systems (4,16,74,13?) and genetic linkage data indicated that one depolymerase was encoded by the bacteriophage (33). Stirm and his colleagues have shown that the soluble depol- ymerase associated with several Eh 9211 K phages consisted of phage tail spikes which were not assembled into the phage particles themselves (23,105). The same conclusion was reached by Sutherland who studied a Klebsiella/phage system (127L.He observed polysaccharide depolymerase activity with purified phage particles which made plaques without haloes when grown on an encapsulated strain of'gg aerogenes (127). Polysaccharide depolymerases have been exploited for structural determinations of their substrate polysaccharides (71,132,136L.Tme presence of polysaccharide depolymerase 12 activity bound to tail spikes of E_._ggli K phage particles resulted in "tunnels" through the capsule (16). "Tunneling" through the bacterial capsule allowed phage particles to attach and inject their genomes at the sites of membrane adhesion to the cell wall of E 2911. (13,14,16). These membrane adhesion sites have been shown to be the site of capsular polysaccharide synthesis (15). The role of the polysaccharide depolymerases in nature was therefore pro- posed to be binding to the polysaccharide capsule and diges- ting it to reach the membrane adhesion sites where the phage genome was injected (16). The gt 9911 K phages were shown to only infect encapsulated bacteria; spontaneous acapsular mutants were not infected (121). Thus the capsule, which prevents infection by phages which require receptors located on the outer membrane (44,89,135), is a required receptor for this group of phage. Bacteriophage which attack phytopathogenic Pseudomonads have been used to select mutants with altered surface com- ponents which were related to virulence (60),lor to identify surface components related to virulence (68). Bacteriophage which attacked phytopathogenic bacteria have been isolated from rice paddy water (62), orchard soils (57), and apple foliage which had been killed by E_._ amylovora (108). Lyso- genic strains of P; syringae pv. morsprunorum have been observed (59), but little is known about hOW’(Or whether) bacteriophage overwinter in orchards (57,107). Civerolo l3 explored the possibility of using a bacteriophage to control 3; pruni damage to peach foliage, but was only able to achieve: limited success (36,37,38,39,40). Ritchie (108) isolated from Erwinia amylovora infected apple tissue sever- al phage which produced clear plaques surrounded by translu- cent haloes in 2139. The haloes observed by Ritchie (108) resembled those produced by a polysaccharide depolymerase associated with phage infection. One of these phage, PEa1(h), was studied further. PEa1(h) lysates of’§h_amylovora were shown to contain an enzymatic activity in vitro which degraded slime polysaccha- rides extracted from E; amylovora grown on pear fruits (64), thus confirming the analogy with phage associated polysac- charide depolymerases found in other systems. A gene was cloned from PEa1(h) which encoded a polysaccharide depolym- erase which was produced in both _E_. 211 and E; amylovora (64). The molecular cloning of the polysaccharide depolym- erase gene of PEa1(h) and preliminary experiments directed towards the use of the cloned polysaccharide depolymerase gene in the study of bacterial exopolysaccharides in plant disease developement are reported in this dissertation. CHAPTER I MOLECULAR CLONING OF THE POLYSACCHARIDE DEPOLYMERASE GENE OF BACTERIOPHAGE PEa1(h) IN ESCHERICHIA COLI 14 Introduction Ritchie (107,108) reported the isolation and charac- terization of a series of phage which produced clear plaques surrounded by translucent haloes when grown on encapsulated Erwinia amylovora. One of these phage, PEa1(h), was selected for further study, since the phage/bacterial interaction appeared entirely analogous to those previously character- ized in Escherichia coli, Klebsiella aerogenes, Pseudomonas aeruginosa and Aerobacter aerogenes (10,22,110,137). Molec- ular cloning of thelwolysaccharide depolymerase gene from purified PEa1(h) and expression of the gene in E15211; would directly demonstrate that the polysaccharide depolymerase is phage encoded as has been suggested in other systems UJJ4127). It would also allow production of large amounts of the enzyme for further characterization and study. There does not seem to be any insurmountable barrier to heterologous gene expression between E; coli and E; amylovora, since drug resistance plasmids transferred from one species to the other resulted in full expression of markers in each species ELM. Therefore expression of the polysaccharide depolymerase gene of PEa1(h) could be expec- ted in E_._ggli, which is not a host for the phage. The 15 l6 cloning vector pBR322 (28) contains a colEl replicon (66) and so replicates independently of the chromosome. This property is useful since it allows amplification of the plasmid with chloramphenicol prior to purification (42,94). The small size of the plasmid (4.3 kb) also facilitates purification. The plasmid contains two selectable antibiotic resistance genes, either of which can be inactivated by the insertion of foreign DNA in one of several unique restric- tion endonuclease cleavage sites. Derivitives of pBR322, pUC8 and pUC9, have been con- structed (95). In these plasmids the tetracycline resistance gene of pBR322 has been replaced by the 133 2 gene of E; coli, leaving the ampicillin resistance gene as the selecta- ble marker. The lag 2 gene encodes a B-galactosidase which cleaves lactose and which can be conveniently assayed be- cause it also cleaves the colorless substrate analog 5- bromo-4-chloro-indolyl-B-D-galactoside (X-Gal) to yield a bright blue pigment UN”. The lag_z gene through operator and promoter functions can be specifically induced by the addition of isopropylthiogalactoside (IPTG)tx>the growth medium (95,96). The lag 2 gene on pUC8 can trans-complement §E_ggli JM83 which contains a deletion in the chromosomal lag 2 gene. The presence of pUC8 in this strain results in ampicillin resistant colonies which are blue colored when grown in the presence of X-Gal and IPTG. Several unique restriction endonuclease cleavage sites have been introduced 17 into pUC8 at the junction of the transcriptional promoter and the structural lag 2 gene (95). Insertion of foreign DNA into these sites, followed by transformation into JM83 re- sults in<:lones which are ampicillin resistant but lag 2- due to insertional inactivation of the lag 2 gene. Such clones grow as white colonies in the presence of X-Gal and IPTG because the inserted DNA is transcribed instead of the _l__a_g 2 gene. The E; coli lac I gene encodes a specific repressor of the lactose operon which prevents transcription of the Oper- on (94). Strains of §;.Egll with a constitutively expressed mutation of lag I, lag Iq (94), are constiutively repressed and inducible only when IPTG is added exogenously. figgli strain JM105 was derived from strain JM83 by the introduc- tion of an F'episome which contains the lag Iq gene. There- fore transcription of the lag 2 gene of pUC8 or of any gene inserted distal to the lag z promoter of pUC8 is constitu- tively repressed in JM105, and specifically inducible with IPTG. Plasmid pUC9 completes the cloning system, and differs from pUC8 only in the reverse orientation of multiple clon- ing sites (95). This al lows transcription of genes encoded on either strand of a given fragment of DNA depending upon its insertion into pUC8 or pUC9. This cloning system was chosen for this study because expression of the phage polysaccharide depolymerase gene in 18 promoter region of the phage gene, which might not.be pre- sent on the DNA fragment which contained the PD gene, could be substituted for with the 139 z promoter region. Further- more, transcripticnxof the phage gene could be either con- stitutive (JM83) or repressed and specifically inducible (JM105). This latter property was desirable since uncon- trolled transcription of the phage gene in E: 9911 might be deleterious to the bacterium (32) or cause plasmid instabi lity (123) . Materials and Methods Bacterial and phage cultures- The bacterial strains used in this study are described in Table 1. After streak- ing to single colonies and genotyping on appropriate media the strains were stored at -20C in 20 mM phosphate buffer pH 6&3(PB) with 40% glycerol (w/v). Phage were stored over chloroform at 4C in 20 mM phosphate buffer, pH6.8 (96). Media- EallOR was grown in nutrient agar (Difco) sup- plemented.with.0.5%remove bacteria, 3 volumes of 95% ethanol were added to the supernatant to precipitate extracellular polysaccharides. The polysaccha— rides were quantified by the phenol/sulfuric acid method (48,65), and used as substrate for an ip 31339 assay for 30 polysaccharide depolymerase (137). EPS at a concentration of 1J5 mg/ml was incubated at 42(21n 100mM Acetate buffer pH 5.0 with a: polysaccharide depolymerase containing aliquot. An aliquot was removed and assayed for increased reducing sugars using a galactose standard with the bicinchoninate assay (98). Preparation pf EPS from immature pear fruits infected with E; amylovora strain EallOR - Immature pear fruits ap- proximately 2.5 cm in diameter were surface disinfested with 0.05 % sodium hypochlorite for ten minutes, rinsed with distilled water, sliced, and placed in sterile enamel pans lined with wet paper towelsh A 10 ul drop of a culture of EallOR grown in nutrient broth (Difco) supplemented with .05 % glucose was placed on each slice and the fruits were incubated at 27 C for 4 days. The bacteria and slime poly- saccharides were obtained by washing the fruit with 20 mM potassium phosphate buffer, pH 6.8 supplemented with 150 mM NaCl. The bacteria were removed by two cycles of centrifuga- tion and two volumes of 95 % ethanol were added to the supernatant which was stored at -20 C overnight. The EPS was collected by centrifugation at 16,000 g, redissolved in a minimal volume of distilled water, and quantified with the phenol/sulfuric acid method (60). Preparation pf soluble pglysaccharide depolymerase from phage lysates pf EallOR and from HBlOl(pJH94) - Soluble polysaccharide depolymerase was obtained from 16 overlays of 31 EallOR which had been confluently lysed by phage PEal(h). These lawns were prepared and the lysates processed as described above for the isolation of bacteriophage PEalun. After bacterial debris, agar and bacteriophage particles were removed from the suspension by centrifugation, the supernatant was dialysed against EH) volumes (HS 20 mM Tris/HCl, pH 8.0 at 4 C. The concentration of protein in this "crude polysaccharide depolymerase" was determined by the method of Lowry et. al. (43,90). Plasmid pJH94 was introduced into El 3911 strain HB101 (29) by calcium chloride transformation with "rapid mini- prep“ DNA prepared from JM105(pJH94). A single colony of E; 9911 HB101 was used to inoculate 250 ml of LB medium which contained ampicillin. After 24 hours growth at 27 C, the bacteria were pelleted by centrifugation at 10,000 g for 20 minutes and resuspended in 125 ml of T8 (50 mM Tris/HCL, pH 8.0, 50 mM NaCl). Sarkosyl was added to 0.1 % (w/v) and the suspensions were vortex mixed for 30 seconds (114). The bacteria were again pelleted by centrifugation. The pellet was resuspended in 16 ml of TES (TS + 10 mM EDTA) and then 14 ml of sucrose mix ( 1.6 M sucrose, 0.55 M Tris/HCl pH 8.0,(L1 M EDTA) was added to the mixture which was incu- bated at 5 C for 20 minutes before 6.0 ml of lysozyme “LS mg/ml, in 50 mM Tris/HCl, pH 84” was added. After incuba- tion at 5 C for 20 minutes, 100 ml of 2.5 % sarkosyl was added and the lysate was stirred on ice for 1 hour at 100 32 RPM . An aliquot from this lysate was set aside and ammonium sulfate was added to the remaining lysate to 30 % saturation with stirring on ice. The precipitate was collected by centrifugation (12,000 g for 30 minutes) and resuspended in 20 mM Tris/HCl, pH 8.0..Ammonium sulfate was added to the 30 % supernatant to 80 % saturation and the precipitate was collected and dissolved in Tris buffer.lfim2crude lysate, and the 30 % and 80 % ammonium sulfate supernatants as well as the 80 % precipitate were dialysed against 20 mM Tris/HCl for'2 days at 4 C before being assayed for polysaccharide depolymerase. Effect pp pp pp the activity pp soluble polysaccharide depolymerases - Serial dilutions of enzyme containing solu- tions (12.5 ul) were added to reaction mixes wich contained 200 mM buffer (50 ul) and §;.amylovora EPS obtained from infected pear fruits (25 ul of 12 mg carbohydrate/ml) and distilled water (lZJSul). The 100 ul reaction mixes were incubated at‘42C2for'3-5 hours before being diluted to 14) ml with distilled water. The concentration of reducing su- gars present in each.reacticn1was then determined with the bicinchoninnate assay (98) using a galactose standard. Reac- tion mixes which had distilled water substituted for either substrate or enzyme were also incubated and assayed as controls. Acetate buffer was used at pH 4.0 and 5.0, potas- sium phosphate buffer was used at pH 6.0 and 7.0, and 33 Tris/HCl buffer was used at pH 84L All assays were per- formed in triplicate. Purification pp pJHgg - HBlOl(pJH94) was grown in 500 ml of DM-17 containing ampicillin, and amplified as was pUC8 above. After centrifugation at 10,0009 for 10 min- utes the pellet was resuspended in 10 ml of a solution containing 25 mM Tris/HCl pH 8.0, 50 mM glucose, 10 mM EDTA, and freshly dissolved lysozyme HSmg/ml) (94L.The suspen- sion was transferred tx)aa Beckman SW-27 polycarbonate cen- trifuge tube and allowed to stand at 25 C for 5 minutes before the addition of 201nlcnfa freshly made solution of (L2 N NaOH and 1% SDS (w/v). The contents of the tube were mixed by gentle inversion and placed inaniice water bath for 10 minutes before the addition of 15 ml of an ice cold solution of 5M potassium acetate, pH 4.8, prepared as des- cribed (94L.The contents of the tube were mixed by sharp inversion and returned to the ice water bath for 10 minutes. The tube was then centrifuged at 72,0009 for 20 minutes and the supernatant decanted into two, 30 ml Corex tubes before 0.6 volumes of isopropanol were added to each tube. The contents of the tubes were mixed by inversion and allowed to stand at room temperature before being centrifuged at 12,0009 for 30 minutes at room temperature in a Sorvall SS- 34 rotor. The supernatant was discarded and the pellet was washed with 70% ethanol at room temperature before being dried briefly in a vacuum dessicator at room temperature. 34 The pellet was dissolved in a volume of 8.3 ml TE, purified through CsClZ/ethidium bromide density gradients and further processed as described above for pUC8. After the final ethanol precipitation, pJH94 was resuspended in TE at a concentration of 660 ug/ml. Extraction pp total DNA from E; amylovora Ea110R - Two 10 ml cultures of Ea110R were grown overnight at 31 C, pooled, then centrifuged at 30009 for 5 minutes in a Sorvall SS-34 rotor. The pellet was resuspended in 5 ml of 50 mM Tris/HCl, pH 8.0, containing 50 mM EDTA before lyso- zyme was added to a final concentration of 1 mg/ml (45). The suspension was incubated in an ice bath for 30 minutes before 1 ml of STEP (0.5% SDS (w/v), 50 mM Tris/HCl pH 7.5, 400mM EDTA, and proteinase K (Sigma) (1 mg/ml), added immed- iately before use) was added to the suspension. The lysate was incubated at 50(2ftu 15 minutes with gentle mixing by inversion before an equal volume of Tris/HCl buffered phenol (pH 7.8) was added and mixed for 5 minutes. After a 5 minute centrifugation in a clinical centrifuge the upper (aqueous) phase was transferred.to a clean teflon capped glass tube and 0.6 ml of 3 M sodium acetate, pH 5.2 was mixed into the lysate. Two volumes of 95% ethanol were added to the tube which was mixed by inversion. The precipitate was spooled out of the mixture with a glass micropipette and trans- ferred to a clean tube containing 5 ml of 50 mM Tris/HCl pH 7.5, 1 mM EDTA, and 50 ugRNAse/ml. The precipitate was 35 dissolved by gentle inversion for 30 minutes and then incu— bated at 37 C with occasional inversion for 30 minutes. Five ml of chloroform was added, mixed to an emulsion, and the tube centrifuged for 5 minutes as before» The upper layer was transferred to a clean tube to which 2 volumes of 95% ethanol were added and mixed. The DNA precipitate was spooled out of the tube and dissolved in 6 ml of 50 mM Tris/HCl pH 7£5containing 1 mM EDTA. This procedure (45) yielded 660 ug of DNA with an A260/280 of 1.6 which was stored at 4CL Southern Blotting - A 0.9% agarose (w/v) gel was cast in a 200 X 145 mm electrophoresis apparatus using a 15 well comb. Samples were loaded in duplicate at opposite sides of the gel. Samples included 70 ng of pUC8 which had been digested with _EppRl, 70 ng of pJH94 digested by EppRl, 2.25 ug of lambda DNA digested with Hidell (BRL), 940 ng of PEAl(h) DNA digested with BglII, and.3.6 ugconfirnlthat only fragments larger than 2 kb had been retained. Screening pp E; coli transformants - After 48 hours at 27 C, clone 94 was surrounded by a very large translucent halo which appeared similar to the haloes surrounding plaques of PEAl(h) except that it was much larger. The halo size was not effected by IPTG, but was much larger when the recombinant clones had been grown at 27 C than when they were grown at 37 C (Figure 3 A,B). Since 27 C is the optimal 42 43 Figure 1. PEal(h) DNA digested by restriction endonuclease Spp3A. Lane 1 — lambda/pldeII size standards, Lane 2 - plasmid pUC8/pppH1,Lanes 3-8 — PEal(h)/§§pBA at 1U/ug,0.SU/ug,0.25U/ug,0.1250/ug, 0.063U/ug and 0.031U/ug DNA.Reactions were incu- bated for 3 hrs, at 37 C, and 1 ug DNA per lane was electrophoresed in 0.8% agarose and stained with ethidium bromide. Figure 2. PEal(h)/_Sip3A fragments which were larger than 2kb, pooled after fractionation over an alkaline sucrose density gradient. Lane 1 - PEal(h)/fl3A fragments, Lane 2- lambda/H_indIII size standards. Fragments were electrophoresed through 0.8% agar- ose and stained with ethidium bromide. The arrow indicates a 2 kb fragment. Figure 3. Screening of recombinants for polysaccharide de- polymerase activity. JM105 clones containing pUC8 derivitives with PEa1(h) DNA inserts were grown 2 days at 27 C (A) and 37 C (B) prior to be- ing exposed to chloroform vapor and overlayed with E; am lovora Ea110R. No IPTG was included in this medium. Note the large halo which sur- rounds one clone. growth temperature for both E.amylovora and PEAl(h) , these observations suggested.that clone 94 contained the intact polysaccharide depolymerase gene as well as its promoter. Effect pp chloroform vapor pp the detection pp haloes surrounding JM105 clone 93 - Excision of colonies of JM105 clone 94 prior to exposing the plate to chloroform vapor did not prevent the appearance of haloes surrounding the site of the bacterial colony (Figure 4), demonstrating that membrane damage caused by chloroform vapor was not responsible for the liberation of PD from the colonies and the appearance of haloes. This suggested that PD was being either excreted into the growth medium or was appearing there as the result of lysis of a portion of the cel ls which contained pJH94. Detection and size estimation pp plasmid pJH94 pp 99105 plppp.9g - All five replicate cultures of JM105 clone 94 contained a new plasmid, designated pJH94, which was approx- imately twice the size of the vector pUC8 (Figure 5). Purified pJH94 was completely digested with restric- tion endonucleases 9391 and Epp R1 and sized by agarose electrophoresis against standards (Figure 6L.pJH94 had an apparent size of 5850 base pairs. The phage DNA insert is therefore 3150 base pairs and has a single EcoRl site. Detection p: polysaccharide depolymeraseaunfivityiip pulture supernatants - Culture supernatants from JM105(pJH94) and JM105(pUC8) were spotted in dilution series On mature lawns of Ea110R grown on DM-17 + 1% glucose. 47 Figure 4. Effect of chloroform vapor on the appearance of haloes around JM105(pJH94L Ten replicate colonies were grown for 2 days at 27 C. Even numbered colonies were excised before plate was exposed to chloroform vapor. Plate was overlayed with E; amylovora Ea110R and incubated for 2 days at 27 C. Figure 5. Plasmid pJH94 detected in JM105 clone 94. The lanes contained: "Mini-Prep" DNA from 5 replicate cultures digested with _Ec_oR1 (Lanes 2-6), pUC8 digested with EpRl (Lane 7), Lambda/fldIII size standards (Lanes 1,8). 49 2 3 4 5 6 7 8 9 Figure 6. Estimation of the size of pJH94. Purified pJH94 was digested with gppRl (Lanes 1-3), and 939 1 (Lanes 7-9).Lambda DNA digested with pldeII as size standards (Lanes 4-6). DNA was electro- phoresed through 0.8 % agarose and stained with ethidium bromide. 50 Culture supernatants from JM105(pJH94) made distinct craters where 10 ul droplets were applied. Such craters are typical of PD containing fluids and have been usedlasauiassay for studying PD (4,53/WM. No activity was found in the superna- tants from JM105(pUC8). Preparation pp polysaccharide depolymerase from PEa1(h) lysates pp Ea110R and from E; coli HBlOl(pJH94) - The 400 ml of nutrient medium used to prepare 16 lawns of E; amylovora confluently lysed by PEa1(h) produced 42.6 units of poly- saccharide depolymerase, which was in two forms CTable 2). About two thirds of the activity was in the soluble form with a specific activity of about 0.63 units/ ug protein. The remaining activity pelletted with the bacteriophage and was presumably bound to the phage particles, but possibly could have included enzyme bound to intact phage tail plates (16,122). The 250 ml of nutrient medium used to grow 9; ppll Imfl01(pJH94) produced 1560 units of soluble polysaccharide depolymerase which were assayed after the cells were lysed and the protein precipitated with ammonium sulfate (Table 2). When the data were expressed as units of polysaccharide depolymerase produced per 151mm: of growth medium, HBlOl(pJH94) produced 75 times as much soluble polysaccha- ride depolymerase as did Ea110R infected by phage PEa1(h) (Table 2L Effect pp p9 pp the activity pp soluble polysaccharide Table 2. Polysaccharide depolymerase activity 51 1 recovered from phage PEa1(h) lysates of Ea110R and from Sarkosyl lysates (112) of HBlOl(pJH94). Units/liter2 Specific Activity3 Source PEa1(h) lysate of Ea110R Phage Pellet 36.5 n.d.4 Soluble 81.5 0.63 Total 118.0 Sarkosyl lysate of HBlOl(pJH94) 80% (NH ) SO Pellet 4 2 4 6240 12.5 uMole galactose equivalents/hr, in a reaction containing 25 ul EPS (12 mg/ml), 12.5 ul distilled H20, 50 ul of acetate buffer, 200 mM, pH 5.0, and 12.5 ul of a dilution series of polysaccharide depolymerase containing solu- tion. The reaction was incubated at 42 C for 3 hours {HBlOl(pJH94)} and 5 hours [Ea110R/PEa1Ufi}. Units/liter of growth medium used to prepare the lysates. Units/mg protein (43,90) n.d. = not determined 52 depolymerases - No differences were observed in the effect of pH on the activity of polysaccharide depolymerase isol- ated from phage PEa1(h) lysates of Ea110R and from sarkosyl lysates of HBlOl(pJH94) (Figures 7 and 8). Enzymes from each source had optimal activity at pH 5.0 and no activity at pH 8.0. The activities were nearly equal for each enzyme at pH 4.0 and 7.0. Hybridization f 23:}: labled pUC8 and pJH94 £9 p3910R and PEa1(h) genomes - The 99911 digestion of PEa1(h) (Figure 9;lane 4) appeared complete and the bands were well sepa- rated, although some band smearing was observed. Summation of the sizes of the PEa1(h) fragments after 99999 digestion gave a genomic size estimate of 47ldnx A continuous size distribution of Ea110R genomic fragments was produced by EcoRl (Figure 9;lane 6). When 32-P labled pUC8 was used as the probe for a blot of this gel (Figure 10) it hybridized with itself, pJH94, and vdjfll a region (M3 the Ea110R/EcoR1 genomic digest. Hybridization with either PEa1(h) or lambda genomes were not detected even after prolonged exposure of the autoradiogram. When 32-P labled pJH94 was used as the probe for a blot of this gel (Figure 10), it hybridized with itself, pUC8, and the same region of the Ea110R/EcoR1 digest as pUC8. In contrast to pUC8, pJH94 strongly hybridized toearegion of the PEa1(h)/89111 genomic digest. No hybridization with lamda DNA was detected even after prolonged exposure of the 53 EallOR/PEal(h) UN ITS/ml Figune7. The effect of pH on the activity of soluble polysaccharide depolymerase isolated from PEa1(h) lysates of E_._ amylovora Ea110R. A unit of activity produced one uMole of galactose equivalents per hour. 54 4ol 1. HBlOl(pJH94) ; 30_ l — E E 3 Z Z a 20- 10- 1 J 1 _1 4 ‘ 5 pH 6 7 3 1 Figure 8. The effect of pH on the activity of polysaccharide depolymerase isolated from sarkosyl lysates of 99 pp99’HB101(pJH94). A unit produced one uMole of reducing equivalents per hour. 55 1 2 3 4 5 6 12 11; {3 2,9 ‘16)" ‘ 7 é‘L‘ "9). w; _ _ “I II “I 6 ' Ifl uh H 6 ut‘w c" Figure 9. Agarose electrophoresis gel stained with ethidium bromide prior to Southern blotting. The lanes contained pUC8/EcoR1,0.13 ug (Lane 1),pJH94/§ppR1, 0.13 ug (Lane—2), lambda/p9deII standards (Lane 3), PEa1(h)/99911 (Lane 4), Ea110R genome/EppRl (Lane 6). l 2 3 44 5 (6 l 2 344 5 6 Figure 10. Autoradiogram of Southern blot which confirms that3polysaccharide depolymerase is phage encod- 33. P-labled pJH94 was the probe (A), and P-labled pUC8 was the probe (BL.Iene 1, pUC8, Lane 2,pJH94, Lane 3, Lambda/p9deI, Lane 4, PEa1(h)/99911, Lane 5, blank, Lane 6, Ea110R genome/EcoRl. Autoradiogram exposed for 15 minutes at -80 C. autoradiogram. The results demonstrate that the 3150 BP insertiJIpJH94 encodingIHDactivity iscfifphage,run:bac- terial origin. Discussion SE:EE£E;JM105 was transformed with a recombinant DNA library of phage PEa1(h) in plasmid pUC8, replica plated and grown 48 hours on a medium containing X-Gal and IPTG. After the recombinant clones were killed with chloroform vapor and overlayed with a lawn of E amylovora, clone number 94 was surrounded by a translucent halo when viewed with transmit- ted light. The region of the halo appeared as a shallow crater when viewed with reflected light and continued to expand after growth of the lawn was complete. The bacteria within the halo were ful 1y viable. These characteristics are identical with thosecnfthe haloes produced by PEa1(h) infection of & amylovora (107,108) and of haloes produced by phage which infect other encapsulated bacteria (4,74). A dilution series of the culture filtrate of clone 94 spotted on mature lawns of 99 amylovora resulted in the appearance of craters, just like those used to assay for polysaccharide depolymerases in other systems (4,53,74) and identical to those produced by spotting dilutions of PEa1(h) lysates on E amylovora (107). When culture filtrates and cell lysates of JM105 clone 94 were incubated with purified EPS from Ea110R, the concentration of reducing sugars in- 58 59 creased with time. This was similar to the results obtained when PEa1(h)/Ea110R lysates were incubated with the same EPS. No craters or haloes were observed when culture fil- trates from JM105 or JM105(pUC8) were spotted on mature lawns of Ea110R, nor were reducing sugars increased when they were incubated with EPS. When the effects of pH on the relative activities of the soluble polysaccharide depolymer- ases found in phage PEa1(h) lysates of E_._ amylovora Ea110R and in sarkosyl lysates of 99pp99 HBlOl(pJH94) were com- pared, no differences were observed; 1flu3 curves which related {Hi to enzymatic activity were identical (Figures 7 and 8). The biological and biochemical data together demon- strated that ipp99 strains JM105(pJH94) and HBlOl(pJH94) produced an enzymatic activity not present in either JM105 or HB101. This enzyme was indistinguishable from the soluble form of the phage associated polysaccharide depolymerase which caused haloes to appear around plaques of PEa1(h) on Ea110R. A plasmid of about 5.8 kb was found in the polysaccha- ride depolymerase producing strains and designated pJH94. Plasmids pJH94 and pUC8 were 32P-labled and used as hybridi- zation probes to determine if the insert DNA was of phage or bacterial origin. Southern blots clearly showed that pJH94, but not pUC8, hybridized to the PEal (h) genome (Figure 10). Both plasmids hybridized.to the same lowrmblecular weight region of the 99 amylovora genomic digest, indicating a 60 region of homology between the vector, pUC8, and a region of the 99amylovora genome.fHu2results demonstrate that the insert DNA carried by pJH94 is of phage, not bacterial, origin. The hybridization of pJH94 to several bands in the PEal(h) genomic digest lane is unexplained. One explanation is that the multiple band hybridization pattern is due to incomplete digestion of the PEa1(h) genome by 99911. However this seems unlikely since the digestion was carried out with excess enzyme for a prolonged period of time. Furthermore, the ethidium bromide stained gel (Figure 9) showed no sign of any bands resulting from partial digestion which can be detected by apparent non-stoichiometric binding of ethidium bromide by the fragments. The size of the phage DNA insert in pJH94 was deter- mined tolxaEL15 kb by restriction endonuclease digestion fol lowed by agarose gel electrophoresis with standards of known size. The molecular masses of similar well character- ized polysaccharide depolymerase enzymes (10,137) eallowed a rough approximtion that about 2 kb of DNA would be required to encode the entire enzyme. Production of polysaccharide depolymerase by JM105(pJH94) did not require IPTG in the growth medium, as it would have if transcription of the gene was under 999 z operator control. It is apparent that the transcriptional promoter of the polysaccharide depolymerase gene was cloned as well as the polysaccharide depolymerase gene itself. This conclusion is reinforced by the observa- 61 tion that polysaccharide depolymerase production was en- hanced when gpp99 JM105(pJH94) was grown at 27 C as com- pared to 37 C (Figure 3), since 27 C is the optimum tempera- ture for PEa1(h) multiplication in §p_amylovora (107). It has been proposed (111,127) that soluble depol- ymerases found in phage lysates are due to overproduction of phage base plate spikes which contain the activity. Ritchie (107) observed that non-halo making mutant phage were always found when PEal(h) was grown on Ea110R in broth. Phage PEal(nh) may have polysaccharide depolymerase activity associated with its particles as was observed on a non-halo making Klebsiella phage (127). The cloning of the polysac- charide depolymerase gene with an intact transcriptional promoter may provide insight on of the regulation of this gene. Whenever JM105(pJH94) was grown on solid or in liquid mediunn polysaccharide depolymerase activity was found in the growth medium. It is unclear whether the enzyme is excreted by the bacteria into the medium or if a portion of the cellr; lyse and thereby release enzyme into the medium. Proteins naturally excreted from bacterial cells typically have 51 very hydrophobic NHZ-terminal sequence which assoc- iates with the cell membrane during translation (17). It is not known if this is the case with this protein. It is clear that the destruction of membranes with chloroform is not required for the enzyme to appear intfluagrowth.medium (Figure 4). It was not possible to amplify and purify pJH94 inggl pp99 strain JM105. Uncontrolled transcription of plasmid encoded genes has led to both plasmid instability and cell lysis in other systems (32,123). Uncontrolled transcription from the phage polysaccharide depolymerase gene promoter could explain both the plasmid instability observed in strain JM105(pJH94) and the appearance of polysaccharide depolymerase in the growth medium. The molecular cloning of the polysaccharide depolymer- ase gene of PEa1(h) and its expression by strains of £2.99li provides a direct demonstration that the gene is phage encoded, and not a bacterial gene which is induced by phage infection as may be the case in'other systems (67fFU. The appearance of polysaccharide depolymerase in very high con- centration in cell lysates of HBlOl(pJH94) could be a great aid in the purification of this enzyme. The same cloning strategy should be applicable in other phage/bacterial sys- tems which have been demonstrated to produce polysaccharide depolymerases (22,23,71,100,106,127).The presence of a phage promoter which controls transcription of this gene, may make possible its use to study the role of extracellular polysac- charides in the pathogenesis of Erwinia amylovora (Chapter 2). CHAPTER II EXPRESSION OF THE CLONED POLYSACCHARIDE DEPOLYMERASE GENE OF BACTERIOPHAGE PEal(h) IN ERWINIA AMYLOVORA 63 Introduction The roles of extracellular and capsular polysaccharides of phytopathogenic bacteria in disease development have been studied using two basic approaches. The first approach has been to compare disease development incited by an acapsular mutant strain to that incited by a fully encapsulated parent strain in a susceptible host. This approach has been pursued vigorously with 99 amylovora (7,19,20,25, 125), 99 stewartii (30) and _P_. solanacearum (115,116). A concensus has devel- oped that the primary role of the EPS in these systems is to prevent agglutination of the bacteria by the susceptible host (30,109,116,125). Mutants which lack EPS seem to be rapidly agglutinated by either lectins (30,116), or by basic proteins (83,84,109) which bind to a moiety in the bacterial LPS. The interpretation of electron micrographs which purport to demonstrate this agglutination 9p 993p has been vigorously challenged (69). A different approach has been to study the fate of EPS from several phytopathogenic bacterial species in leaf meso- phyll of plant species susceptible and resistant to the several pathogens (54). The EPS induced persistant water soaking only in plants susceptible to the bacteria from which it had been obtained (54). The experiments were exten- 64 65 ded to compare the fate of EPS from & syringae pv. phaseo- licola in bean cultivars susceptible and resistant to sever- al strains of the pathogen.'These workers concluded that the leaf mes0phyll (55,56L.The hydrated gel presumably enhanced bacterial growth (69,76). The workers further concluded that resistance of bean leaves to this persistant water soaking was due to enzymatic degradation of the EPS in an incompati- ble, but not in a compatible interaction (56). Still another approach was used by Avery and Dubos in a study of the role of EPS in pathogenicity of Pneumococcus in mice (6,49). These workers enzymatically removed the extracellular polysaccharides of a virulent strain of gppp: mococcus prior to injecting it, with enzyme, into the mouse peritoneum.lfimaenzymatically decapsulated bacteria were rendered avirulent, presumably because they were more sus- ceptible to phagocytosis H”. The EPS interfered with the recognition of the bacteria by the phagocytes, a model similar to that proposed for 99 amylovora, §p_stewartii and .P. solanacearum EPS today. The cloning of the polysaccharide depolymerase gene of phage PEa1(h) reported in chapter 1 of this work made possi- ble a fourth approach to the study of EPS and CPS in disease development. The intact polysaccharide depolymerase gene was introduced into a fully virulent, encapsulated strain of 99 amylovora by transformation with plasmid pJH94. The result- 06 ing strain, Ea110R(pJH94), produced polysaccharide depolym- erase and was characterized with respect to its surface polysaccharides and pathogenicity. Materials and Methods Transformation p9 99 amylovora Ea110R with plasmids pJH94 and pUC8 - The CaClz procedure (94) described in chapter 1 for & pp99 was used to transform 9._ amylovora Ea110R, except that bacterial growth was at 27 C..AmpiCillle resistant transformants, designated Ea110R(pJH94) and Ea110R(pUC8) were selected for further study. Physical detection p9 plasmids 9p 99 amylovora strains- Strain Ea110R(pJH94) was grown in DM-17 containing ampicil- lin. Amplification of the two 250 ml cultures with chloram- phenicol (42,94) and alkaline lysis (26,94)‘were carried out as described for HBlOl(pJH94) in chapter 1. The plasmid containing pellet obtained after alkaline lysis was used as substrate for restriction endonuclease 9991 after it was sequentially extracted with phenol, phenol:chloroform, and chloroform, precipitated with ethanol, and resuspended in TE buffer. The A260/280 of the plasmid preparation was 1.9. Plasmid preparations were made in the same manner from strain Ea110R and Ea110R(pUC8). Aliquots of 20 ul from these preparations were digested for 3 hours at 37 C with 30 U of 9991. An aliquot of pJH94 isolated from HBlOl(pJH94) and purified through cesium chloride density gradients was also 67 68 digested with Sall as a control. Samples were electrophor- esed through 0.9% agarose at 50 mA and 100V for 90 minutes before bands were visualized with ethidium bromide. Production p9 polysaccharide pyi99 amylovora 9p vitro - Plasmid containing strains were grown in medium containing ampicillin at 200 mg/ml. Overnight cultures of Ea110R, Ea110R(pUC8), and Ea110R(pJH94) in DM-17 were diluted 1/10 in fresh medium. After 1 hour incubation at 25 C the cul- tures were diluted 1/1000 in phosphate buffer and aliquots of 0.1 ml were spread on Petri plates containing DM-17 supplemented with 2% dextrose (w/v). The Petri plates were sealed in a plastic bag and incubated for 4 days at 27 C. The plates were washed twice with 2.5 ml of 10 mM potassium phosphate buffer pH 7.2 which contained 150 mM NaCl and 1 mM M9504 (PBSM) (In. The suspension was centri- fuged for 10 minutes at 7,7009 in a 38-34 rotor, the super- natant retained, and the pellet washed 2 times with the same buffer. The supernatants from the first two centrifugation steps were pooled, and centrifuged 2 times at 12,0009 for 15 minutes. Three volumes of 95% ethanol were added to the final supernatant which was then stored overnight at -20 C. The precipitate was considered to be slime polysaccharide or EPS (51,71). The cell pellets from the previous steps were pooled and resuspended in 10 mM potassium phosphate buffer pH 7.2 which contained 500 mM NaCl and 1 mM M9504 (PBSSM) (In. The 69 suspension was stirred vigorously for 1 lunn: at room temperature before the cells were removed by 2 cycles of centrifugation at 12,0009 for 20 minutes. Three volumes of 95% ethanol were added to the supernatants which were then stored overnight at -20 C. The precipitate was considered to be capsular polysaccharide (CPS) (51,71). The polysaccharide precipitates were collected by cen- trifugation at 16,0009 for 1 hour, resuspended in 20 ml of dHZO with rotary shaking for 22 hours. The supernatants were concentrated to 1/3 their original aqueous volume by rotary evaporation. The initial cell suspension was diluted 1/50, 1/75, and 1/100 in PBSM and the cell numbers were determined with a Petroff-Hauser counting chamber after staining with 1/10 volume of a solution of 0.005% crystal violet (w/v). Chemical analysis p9 polysaccharides and oligosaccha- 99999 -Carbohydrates which were precipitated with 3 volumes of ethanol and those which were not preciptated were stud- ied. Total carbohydrate content of the samples was quanti- fied using the phenol/sulfuric acid method (48,65). The concentration of reducing ends in the samples was quantified with the bicinchoninnate method (98). Uronic acids were quantified using the method of Blumenkrantz (27). Preparation p9 pp9ysaccharide depolymerase from Ea110R(pJH94) culture supernatants - A 10 ml overnight cul- ture of Ea110R(pJH94) in DM-17 containing ampicillin at 200 70 ug/ml was used to inoculate 500 ml of the same medium.lflua culture was shaken for 48 hours at 27 C before the cells were removed by two cycles of centrifugation. Ammonium sul- fate (enzyme grade) was added to 80% saturation and stirred at 0 C for 3 hours. The precipitate was collected by centri- fugation at 10,0009 in a GSA rotor and resuspended in 30 ml of phosphate buffer before dialysis against two 2 liter changes of phosphate buffer at 4 C. Comparison pp polysaccharides produced py 99 amylovora strains 99 substrates for polysaccharide depolymerase - EPS prepared from Ea110R, Ea110R(pUC8), and Ea110R(pJH94) were adjusted to concentrations of 0.3 mg carbohydrate/ml each and used as substrate in an 9p 993592 assay for polysaccha- ride depolymersase. Each reaction mix contained 100 ul of PD prepared from Ea110R(pJH94), 200 ul of EPS, and 200 ul of 200 mM acetate buffer pH 5.0, and was incubated at 42 C for 5 hours. Aliquots were removed and assayed for reducing sugars with the bicinchonninate assay. Assays were performed in triplicate. Reaction mixes which had dHZO substituted for the polysaccharide depolymerase or for the EPS substrate were used as controls. CPS from the same strains was adjus- ted to 0.2 mg carbohydrate/m1 and assayed as substrate for PD in the same manner. Test p9 polysaccharide depolymerase from Ea110R(pJH94) for lyase activity- Polysaccharide lyases cleave the glyco- sidic bond by an elimination reaction, which creates an 71 unsaturated carbon-carbon bond in the product which can be quantified spectr0photometrically if the product is an uronic acid (73,79). Depolymerases which cleave the glyco- sidic bond by hydrolysis do not create such a bond. Three tubes were prepared each containing 1 mg EPS from Ea110R, 500 ul buffer, and 500 ul dHZO. Two of the tubes received a 200 ul aliquot of crude PD and were incubated at 42 C..After 7 hours, PD was added to the third tube, and the A235nm of the incubated reaction mixes was determined using the unin- cubated reaction mix as the blank. As a control the incu- bated reaction mixes were assayed for reducing sugars by the bicinchonninate method using the unincubated reaction mix as the blank. Quantification o_f polysaccharide depolymerase retained py Ea110R(pJH94) and found 9p 99 medium - Three 10 m1 cul- tures of Ea110R(pJH94) were grown 20 hours at 27 C in LB containing ampicillin (200 mg/l). The cultures were pooled, an OJLGOO of 1.3 was measured, and three 10 ml aliquots were removed. The aliquots were centrifuged (10,0009, 5 minutes, SS-34 rotor), and the cell pellets resuspended in 10 ml each of ice cold TS (50 mM Tris/HCl, pH 8, 50 mM NaCl). Sarkosyl was added to 0.1% (w/v) final concentration (114), and the cultures were vortex mixed for 30 seconds and centrifuged as before. Each pellet was resuspended in 0.4 ml TES (TS + 10 mM EDTA) and then 0.35 ml of sucrose mix (1.6 M sucrose, 0.55 M Tris, pH 8, 10 mM EDTA) was added to 72 each tube. Tubes were incubated at 5 C for 20 minutes before 0.15 ml lysozyme (5 mg/ml in 50 mM Tris pH 8) was added and mixed by inversion. This was followed by the addition of 3.6 ml of 10 mM EDTA, pH 8. After the suspensions were mixed they were incubated at 5 C for 20 minutes. At this point one culture received 2.5 ml of 2.5% sarkosyl (w/v); one culture was vortex mixed at high speed for 30 seconds before receiv- ing 2.5 ml dHZO and one culture was sonicated and then received 2.5 ml dHZO. All cultures received 3.0 ml dHZO to bring their volumes to 10 ml each. The original culture supernatant and each lysate was assayed for PD 9p y9ppp in triplicate. Reducing sugars were quantified in each reaction mix with the bicinchon- ninate assay. Analysis p9 proteins 9p culture supernatants pp 99 amylovora strains - Single colonies of strains Ea110R, Ea110R(pUC8), and Ea110R(pJH94) were used to initiate 10 ml cultures in LB medium or in LB medium supplemented with ampicillin. Cultures were grown overnight at 24 C before they were diluted 1/50 in fresh medium and grown a further 24 hours. A fourth overnight culture, strain Ea110R, was inoculated with phage PEa1(h) at a MALI. of 1/1, 3 hours after it was diluted 1/50 into fresh LB medium. Bacteria were removed from all cultures by two cycles of centrifuga- tion at 8,000 g for 15 and 30 minutes. All culture superna- tants were spotted in 2-fold dilution series on mature lawns 73 of strain Ea110R and examined after 6 hours for the presence of craters typically produced by polysaccharide depolymer- ases (4,74,107). Ammonium sulfate was added to the superna- tants to 80 % saturation with stirring for 30 minutes at 0 (L A fifth culture, of Ea110R, was grown in the manner of the first three cultures and the cells collected by centri- fugation. These cells were used to prepare sarkosyl lysates (114 and above). After dialysis against 20 mM Tris/HCl, pH 8.0, proteins were precipitated by the addition of ammonium sulfate to 80 % saturation at 0 C. Precipitates were collec- ted by centrifugation at 12,000 g for 20 minutes and resus- pended in and dialysed against 20 mM Tris/ HCl, pH 8.0. The concentration of protein in each of the five preparations was quantified by the method of Lowry et. al. (43,90). A denaturing 7.5 % polyacrylamide gel was prepared according to the procedure of Laemmli (82L.Samples which contained 16.5 ug of protein of each of the five prepara- tions and a sixth sample which contained 3 ug of protein size standards were denatured by boiling for 3 minutes in the presence:of SDS and electrophoresed through the poly- acrylamide gel. Electrophoresis was for 7 hours at 45 mA and 110 V. After electrophoresis the protein bands were visual- ized by silver staining as described by Morrissey (97). Determination p9 the minimium inhibitory concentration p9 CuSO4 and antibiotics with strains Ea110R(pUC8) and Ea110R(pJH94) - The bacterial strains were grown overnight 74 at 27 C on LB agar which contained 0.5% dextrose to enhance encapsulation and ampicillin to maintain selection for the plasmids in the strains. The bacteria were suspended in 5 mls of phosphate buffer and diluted 1/100. Aliquots of 0.1 ml was used to inoculate a 3.0 ml top agar NGAYE overlay which was immediately poured over a NGA plate which con- tained ampicillin. After the top agar had solidified, 3 filter paper discs were put on the surface of each plate. The discs had previously received 5 ul each of CuSO4 solu- tions at the following concentrations (mM): 200, 100, 50, 25, 12.5, 6.25, 3.12, 1.56, 0.78, 0.39. The plates were incubated at 27 C for 40 hours before the greatest dilution at which the bacteria grew up to the edge of the discs was recorded. The sensitivities of the two strains to several anti- biotics were compared in the same manner. Strains were grown overnight at 25 C in DM-17 which contained 1 % dextrose to enhance encapsulation and ampicillin to maintain selection for the plasmids contained in the strains. The O.D.600 was adjusted to 0.24 with phosphate buffer. Aliquots of 0.1 ml were added to 3J)nfl.top agar overlays (DM-17) which were poured over plates of the same medium which contained ampi- cillin. The antibiotics naladixic acid, streptomycin sul- fate, and tetracycline hydrochloride were diluted with dis- tilled H20 to concentrations of (mg/ml): Zdhrlu25, 0.63, 0.31, 0.16, 0.08, 0.04, 0.02, 0.01 and 0.005. Aliquots of 75 2.5 ul were applied to each of three filter paper discs which were placed on the surface of the overlay after it had hardened. Plates were incubated for 40 hours at 27 C and the greatest dilution at which bacteria were able to grow to the edge of the disc was determined. Pathogenicity study- The pathogenicity of strain Ea110R(pJH94) was compared to that of strains Ea110R, Ea110R(pUC8), and Ea8 (61). Cultures were grown to late exponential phase in DM-17, pelleted at 12,0009 for 5 min- utes, and resuspended in phosphate buffer or in phosphate buffer with ampicillin to an O.D.600 of 0.05. Immature pear fruits (25,104) approximately 3 cm in diameter were surface disinfested in a solution of 5% bleach and 0.1% triton X-100 for 10 minutes, rinsed in distilled water and placed in styrofoam egg cartons. Three 10 ul drops of inoculum or buffer were placed on each pear fruit and the fruit were stabbed through the droplet with a 20 G syringe needle. The pears (4/treatment) were incubated at 27 C while being observed daily for symptom development. RESULTS Strains p9 Erwinia amylovora - The genotypes of the four E amylovora strains are given in Table 3. Plaques with haloes were observed on all four strains tested when PEa1(h) was streaked on basal DM-17 supplemented with 1 % dextrose and overlayed with bacteria. Even the "acapsular" strain Ea8(61) produced plaques surrounded by haloes in this exper- iment. Although strain Ea110R(pJH94) produced haloes when infected with PEa1(h), the haloes were very faint and indis— tinct as compared with either haloes produced on strains Ea110R, Ea110R(pUC8), or Ea8 (Figure 11). When grown on solid DM-17 containing 2% dextrose, strains Ea110R and Ea110R(pUC8) produced very fluidal colon- ies typical of E amylovora. Strain Ea8 also had a fluidal colonial morphology under these conditions, although not nearly as pronounced as the typical strains. In contrast to these strains, Ea110R(pJH94)‘was not at all fluidal (Figure 12). Polysaccharide depolymerase activity was found in the culture filtrates and cell lysates of Ea110R(pJH94) but not in the culture supernatants or lysates of strain 76 77 Table 3. Strains of 99 amylovora used in this study Strain Relevant Genotype Reference or Source Ea110R rifr 107,108 Ea110R(pUC8) rifr(ampr,lac z) This work (95) Ea110R(pJH94) rifr(ampr,lac z-) This work 78 Figure 11. Plaque morphology of PEa1(h) on 99 9py9pypp9. Ea110R (A),Ea110R(pUC8) (B), Ea110R(pJH94) (C), and Ea8 (D). Bacteria were grown in DM-17 overlays which contained 1% glucose for 5 days at 27 C. 79 Figure 12. Colonial morphology of 99 amylovora. Ea110R (A), Ea110R(pUC8) (B), Ea110R(pJH94) (C), and Ea8 (D). The bacteria were grown on DM-17 medium which contained 2 % glucose for 5 days at 27 C. 80 Ea110R(pUC9). Physical detection p9 plasmids 9p 99 amylovora strains Plasmid pJH94 was detected in lysates of Ea110R(pJH94) after digestion with 9991 and electrophoresis through 0.9% agarose (Figure 13). Plasmid pJH94 purified from E921; HB101 was completely digested by Sall, but pJH94 prepared from Ea110R was apparently only partially digested by Sall (Figure 13; lanes 1 vs 2). Plasmid pUC8 was detected in strain Ea110R(pUC8) (Figure 13; lane 3). No resident plasmids were detected in strain Ea110R (Figure 13; lane 4). The prepara- tion of pJH94 DNA isolated from Ea110R(pJH94) (Figure 13; lane 2) was used to transform fipp99 HB101 to ampicillin resistance. All transformants tested (80/80) produced poly- saccharide depolymerase 9p y9ppp, which indicated that the ampicillin resistance and polysaccharide depolymerase mar— kers remained linked after passage in 99 amylovora. Chemical analysis p9 polysaccharides and oligosaccha- 99999 -The quantification of total carbohydrates produced by the bacterial strains is presented in Table 4. Strain Ea110R(pJH94) produced less ethanol precipitable EPS and CPS than either Ea110R or Ea110R(pUC8) on a per cell basis. The ethanol precipitable EPS and CPS produced by Ea110R(pJH94) also differed qualitatively from that of the other two strains. The concentration of reducing ends pre- sent per m9 of carbohydrate was much higher with strain Ea110R(pJH94) than with either of the other strains (Table 81 Figure 13. The physical detection of plasmids in alkaline lysates of 9:9py9pyp99 strains. pJH94 purified from HBlOl(pJH94) lane 1,1ysates of Ea110R(pJH94),lane 2,Ea110R(pUC8),1ane 3,and Ea110R, lane 4, lambda/HindIII size standards, lane 5, Lanes 1-4 electrophoresed after digestion with Sall. 82 Table 41 Quantification of ethanol precipitable polysaccha— rides produced by 99 amylovora strains. mg EPSl'2/10ll cells mg CPS/1011 cells STRAIN Ea110R 0.79 0.30 Ea110R(pUC8) 0.68 0.31 Ea110R(pJH94) 0.41 0.24 1. Carbohydrates quantifiedvfidfllthe phenol sulfuric acid method (48,65) using a galactose standard. Cells enumer- ated with a Petroff-Hausser bacterial cell counter. 2. Mean of 3 experiments, 3 replicates each. Table 5. Quantification of reducing sugars in ethanol precipitable polysaccharides produced by 99 amylovora strains ug Reducing equiv./mg EPSl'2 ug Reducing Equiv./mg CPS Strain Ea110R 15.1 18.1 Ea110R(pUC8) 16.1 19.3 Ea110R(pJH94) 49.4 35.7 1. Reducing equivalents determined with the bicinchonninate assay (27), with a galactose standard. Carbohydrates quantified with the pmenol/sulfuric acid assay (48,65), with a galactose standard. Each assay used algalactose standard. Cells enumerated with a Petroff-Hauser bacter- ial cell counter. 2. Mean of 3 experiments, 3 replicates each. 83 5). This demonstrated that the average polymer length was shorter for EPS and CPS recovered from Ea110R(pJH94) than for the other strains. The EPS and CPS recovered from this strain also contained less uronic acid per weight of carbo- hydrate (Table 6). Thus both the ethanol precipitable EPS and CPS of strain Ea110R(pJH94) differed quantitatively and qualitatively from that of strains Ea110R and Ea110R(pUC8). Much larger amounts of carbohydrate soluble in three volumes of ethanol were recovered from the EPS of strain Ea110R(pJH94) than from the other strains (Table 7). This ethanol soluble carbohydrate probably included oligosaccha- ride products of the enzymatic hydrolysis of EPS andCPS of strain Ea110R(pJH94). However, the very high proportion of reducing equivalents to carbohydrate in this material as wel 1 as that from the other strains may suggest that it also contained a simple sugar, possibly glucose from the growth medium. No differences were observed among the ethanol solu- ble sugars recovered from the CPS of these strains (Table 7). Comparison p9 polysaccharides produced 9y 9; amylovora strains 99 substrates for pplysaccharide depolymerase - Incubation of EPS from both Ea110R and Ea110R(pUC8) with polysaccharide depolymerase resulted in a sharp increase in the concentration of reducing equivalents as compared to tubes incubated without EPS or without polysaccharide depol- ymerase. In contrast, incubation of EPS from Ea110R(pJH94) 84 Table 6. Quantification of uronic acids in ethanol precip- itable polysaccharides produced by 99 amylovora. Strain mguronicacidl/mg carbohydrate2 Ea110R EPS 0.23 Ea110R(pUC8) EPS 0.21 Ea110R(pJH94) EPS 0.12 Ea110R CPS 0.18 Ea110R(pUC8) CPS 0.18 Ea110R(pJH94) CPS 0.12 1. Determined as glucuronic acid equivalents (27). Mean of two experiments, three replicates each. 2. Total carbohydrate, determined as galactose equivalents (48,65). Mean of two experiments, three replicates each. .cumccmum omouomflmm m uncemmm Away cocuoe mum: Ificcococflofin ocu cues cocfleuouoo .mucoEwnmmxm m mo cowumfi>oc cumpcmuw + new: .m .cumecmum mmouomamo m umcfimmm oocuoe Amo.mv. cflom UHHSMHSm Hocmcm mcu zufl3 cmcflsumumo .mucoefluomxm m mo :oflumfl>mc cumccmum + :09: .H mma + Hum mm + hmm mm + mum moa + mmm omH + owe mv + omm 1) 8 oumucwconumo ms Nucoam>flsvo unfloscon m: No.o mo.o mo.o OOH m. mo. H. Homoaaoa\me .... mo.o mo.o oa.o ~¢.N mm.o mm.o mmU mmU mmU mmm mmm mmm Lemmnmvmoaamm Amoodvmoaflmm madame LemmedvaHHmm Amoademoaamm moaamm camupm .mcflmuum muo>oHNEm um Eoum coum>ooon moumucwconumo oHnsHom Hocmcuo mo cowumowmwucmso .5 oHnme 86 with polysaccharide depolymerase resulted in only a barely detectable increase in the concentration of reducing equiva- lents in the same assay (Table 8). The tube which contained EPS from Ea110R(pJH94)tnu:no added PD showed a high con- centration of reducing equivalents. These results suggested that EPS from Ea110R(pJH94) had few, if any, remaining sites for polysaccharide depolymerase cleavage. Similar results were observed with CPS obtained from the 3 strains (Table 8). When incubated with polysaccharide depolymerase, CPS from Ea110R and Ea110R(pUC8) produced an increase in the concentration of reducing sugars in the reaction mix, which indicated that they had been cleaved by polysaccharide depolymerase. CPS from Ea110R(pJH94) produced no increase in reducing sugars when incubated with polysac- charide depolymerase. As was the case with EPS from this strain, the concentration of reducing equivalents measured in the reaction mix containing only CPS of Ea110R(pJH94) was very much greater than in the reaction mixes containing only CPS from Ea110R or Ea110R(pJH94). The results suggested that CPS from Ea110R(pJH94) was not substrate for polysaccharide depolymerase. 9999 p9 99999 pp9ysaccharide depo9ymerase from Ea110R(pJH94) for lyase activity - When assayed spectropho- tometrically there was no increase in the A235 in the tubes incubated with polysaccharide depolymerase for 7 hours as compared to the tube which had not been incubated" Inlcon- 87 Table 8. Comparisons of polysaccharides produced by strains (nigh amylovora as substrates for polysaccharide depolymerase. Strain Polysaccharide1 A5652 ug Galactose Equiv.3 Released by PD Ea110R EPS .345+.008 7.1 Ea110R(pUC8) EPS .325+.008 6.8 Ea110R(pJH94) EPS .020+.009 0.4 Ea110R CPS .093+.001 1.9 Ea110R(pUC8) CPS .067+.004 1.3 Ea110R(pJH94) CPS —.003+.002 0 1. EPSwas used at 60 ug/SOOul reaction. CPS was used at 40 ug/SOO ul reaction. 2. Mean + standard deviation of three replicates. A565 of control reactions containing substrate or enzyme only have been subtracted from all values. 3. The bicinchonninate assay (98) was used to quantify reducing sugars released by PD from the polysaccharides using a galactose standard. 88 trast, there was a sharp increase in the concentration of reducing equivalents in the incubated as compared to the non-incubated reaction mixes. Quantification of polysaccharide depolymerase retained by Ea110R(pJH94) and found in EE medium - A unit of polysac- charide depolymerase activity was defined as the amount of polysaccharide depolymerase which would produce 1 umole of reducing equivalents per hour in a standard reaction mix. The majority of polysaccharide depolymerase activity was retained within Ea110R(pJH94) in this experiment but a sig- nificant amount appeared in the culture supernatant (Table 9). Analysis of proteins found in culture supernatants of gt amylovora strains - Craters were produced on mature lawns of g; amylovora when dilutions of culture supernatants ob- tained from Ea110R(pJH94) or from Ea110R/PEa1(h) were spot- ted on them. No craters were observed when supernatants from the other strains were spotted on them. Culture supernatants obtained from strains Ea110R or Ea110R(pUC8) contained few proteins which could be resolved in a 7k5 % polyacrylamide gel (Figure 14; lanes 2 and 3). This was in sharp contrast to the supernatant obtained from strain Ea110R(pJH94) which produced a very complex pattern which was virtually indis- tinguishable from the patterns produced by either phage or sarkosyl lysis of strain Ea110R (Figure 14; lanes 4-6). The results suggest that strain Ea110R(pJH94) partially lysed 89 Table 9. Quantification of PD retained by Ea110R(pJH94) and found in LB medium. Units1 PD/ml Total Units Source Culture Supernatants 1.1 32 Sphaeroplasts lysed with 4.7 142 sarkosyl Sphaeroplasts lysed with 4.7 142 sonication Sphaeroplasts lysed with 1.6 p 50 vortex mixing 1.1 Unit = luMole Galactose Equivalent/ml hr. The following is a sample unit calculation: A5 5=a195 = 4 ug gal eq./50 ul aliquot assayed. 4 ug f 8 = 32 ug/400 ul reaction. 32ug/180 ug/uMole gal =.177 uMole Gal. eq5/3 hr = .059 uMole/hr in 50 ul aliquot of PD assayed. .059 X 20 = 1.19 U/ml at 1/4 dilution or 4.74 U/ml undiluted. 4.74 U/ml X 30 ml = 142 U total. 90 1 2 3 i 5 6 7 Figure 14. Analysis of proteins found in culture superna- tants of E. amylovora strains. 16. 5 ug of protein was loaded per lane in a 7. 5 % polyacryl- amide gel which was electrophoresed under dena- turing conditions. Culture supernatants were from Ea110R (lane 2), Ea110R(pUC8) (lanes 3 and 7), and Ea110R(pJH94) (lane 4). Proteins from a PEa1(h) lysate of Ea110R and a sarkosyl lysate of Ea110R were in lanes 5 and 6. Protein size standards (3 ug) were loaded in lane 1. 91 either before or during the stationary phase of growth. Neither strain Ea110R or Ea110R(pUC8) seemed to have lysed under identical conditions. Polysaccharide depolymerase appeared in the growth medium most likely as the result of this partial lysis of the culture, either before or during the stationary growth phase, and probably not as the result of specific excretion of the enzyme by Ea110R(pJH94). Determination of the minimum inhibitory concentrations of gu§g4 and antibiotics on strains Ea110R(pUC8) and Ea110R(pJH94) - No differences in sensitivity to antibiotics or CuSO4 were observed for the two strains. The minimum inhibitory concentrations for naladixic acid, streptomycin sulfate and tetracycline hydrochloride were 0.16 mg/ml, 0.04 mg/ml, and 0.31 mg/ml respectively for each of the two strains tested. The minimum inhibitory concentration of CuSO4 was 50 mM for both strains. Pathogenicity study~ Symptoms developed rapidly in fruits inoculated with Ea110R, Ea110R(pUC8) or Ea110R(pJH94) (Figure 15). Symptoms caused by the first two strains in- cluded extensive necrosis and the copious production of ooze which are diagnostic of fireblight infections. Strain Ea110R(pJH94) caused necrosis but in marked contrast to the other isolates usually failed to produce any ooze. Strain Ea8 and the buffer controls produced only slight necrosis or no symptoms at all in this experiment which was repeated several times. Occasionally ooze was produced by pear fruits 92 .Snmmuimvlomvmofimm can .Cmucmoiemmnfimoimm .Smm: mofimm mcamfim muo>oHNEm .m :uflz umwfiumo mmmo m>fim owumHooosm ouwwumufloum umom .mufisum ummm musumeefi co mcflmuum muo>o~mem .m an «5950 meoumewm .mH 93mg 93 94 inoculated with Ea110R(pJH94), particularly after prolonged incubation.]31these instances the bacteria present.in the ooze were sensitive to ampicillin, indicating that they had lost pJH94. DISCUSSION The introduction of pJH94 into E; amylovora strain Ea110R by transformation dramatically altered the colonial morphology of Ea110R (Figure 12). On rich medium, or on defined medium containing excess glucose, the slimy or flu- idal morphology typical of E_._ amylovora isolates was ob- served with strains EallOR and Ea110R(pUC8) but not with strain Ea110R(pJH94). Evidence which shows that this al- tered morphology is due to the presence of the cloned poly- saccharide depolymerase gene in Ea110R(pJH94) includes: i. The isolation of pJH94 from Ea110R(pJH94). ii. The presence of polysaccharide depolymerase activity in culture supernatants and«cell lysates of this strain but not in culture supernatants or cell lysates from Ea110R or Ea110R(pUC8). iii. The cleavage of both EPS and CPS from Ea110R by the polysaccharide depolymerease in yitrg. iv. The observation that EPS and CPS from Ea110R(pJH94) were not substrates for polysaccharide depolymerase in vitro 95 96 v. The recovery of quantitatively less ethanol precipitable EPS and CPS from Ea110R(pJH94). vi.The increased concentration of reducing ends inboththe EPS and CPSisolatedfrom Ea110R(pJH94). vii. The sharply reduced intensity of haloes surroun- ding plaques made by PEa1(h) on Ea110R(pJH94). If Ea110R(pJH94) had excreted polysaccharide depolymer- ase into the growth medium, the complexity of the protein mixture found in its culture supernatant.would.have dif- fered from that of Ea110R and Ea110R(pUC8) only by the addition of a band or bands corresponding to the polysaccha- ride depolymerase. If,cn1the other hand, polysaccharide depolymerase entered the growth medium as a result of lysis of Ea110R(pJH94), then the complexity of the protein mixture in the Ea110R(pJH94) supernatant would be similar to that observed for a phage infected culture or for a sarkosykl lysate of Ea110R. Culture supernatants of Ea110R(pJH94) contained a very complex mixture of proteins as revealed by SDS/acrylamide electrophoresis (Figure 14; lane 4). Superna- tants of PEa1(h) infected Ea110R contained an extremely similar mixture of proteins, as did a sarkosyl lysate of Ea110R (Figure 14; lanes 5,6). Such a complex mixture of proteins was not observed in supernatants of Ea110R or Ea110R(pUC8) (Figure 14; lanes 2,3). It is most likely that polysaccharide depolymerase appeared in the growth medium at or before stationary phase apparently due to lysis of a 97 portion of the cultures of Ea110R(pJH94). There was noevi- dence that polysaccharide depolymerase was specifically excreted by Ea110R(pJH94). The sensitivities of strain Ea110R(pUC8) and Ea110R(pJH94) to antibiotics and CuSO4 were compared because it had been reported that treatment of Ea110R with a crude preparation of polysaccharide depolymerase resulted in in- creased sensitivity to streptomycin sulfate (107). Ea110R(pJH94) was no more sensitive than Ea110R(pUC8) to any of three antibiotics or to CuSO4. Perhaps the earlier result (107) was due to the presence of another, unidentified factor in the crude enzyme prep used in that study. If so, identification of that factor could prove to be rewarding. The development of extensive necrosis when pear fruits were inoculated with Ea110R(pJH94) suggests that intact extracellular polysaccharides may not be required for patho- genesis by _E_:_._ amylovora. The lack of ooze produced by pear fruits infected with Ea110R(pJH94) suggests that ooze is composed largely of intact bacterial polysaccharides. This is in agreement with a chemical analysis of ooze polysac- charide and EPS which has shown very similar molar ratios of several sugars found in each polysaccharide (21), chromato- graphic evidence (21,118) and serological data (118). The pathogenicity study should be repeated using intact plants and using a wide range of inoculum concentrations on imma- ture pear fruits to gain a better understanding of the 98 effect of the lower molecular weight extracellular and cap- sular polysaccharides on disease development. While care must be taken when interpreting the results of the pathogenicity study which compared Ea110R, Ea110R- (pUC8), Ea110R(pJH94) and Ea8, some tentative conclusions can be made. Strains Ea110R.and Ea10R(pUC8) did not differ quantitatively or qualitatively in the production of EPS or CPS in litre. The pathogenicity of Ea110R was also not affected by pUC8, nor was the production of the ooze charac- teristic of fire blight disease. Plasmid pJH94 did alter colony morphology (Figure 12) and ethanol precipitable poly— saccharides, both quantitatively and qualitatively in yitrg (Tables 4-6). Neither EPS nor CPS isolated from Ea110R(pJH94) were good substrates for polysaccharide depo- lymerase, as compared to EPS and CPS isolated from Ea110R and Ea110R(pUC8) (Table 8, Figure 1). Plasmid pJH94 also prevented the production of ooze characteristic of fire blight disease although it did not prevent infection of, and multiplication in, pear fruit. The results of this preliminary pathogenicity study can be reconciled with the proposed role of EPS and CPS as anti- agglutination factors (109,116), since polysaccharide frag- ments could still retain this activity. Experiments with intact plants will be required to determine how'(or whether) the polysaccharide fragments influence the vascular plugging symptom associated with fire blight disease (70,119). Slime 99 polysaccharides of different molecular weights naturally produced by St michiganense pvzinsidiosum have been observed to accumulate at different locations in the transpiration stream of plants dipped in solutions of labled polysaccha- rides (99). Lower molecular weight polysaccharides moved further in the transpiration stream, to vessels of smaller capillary diameter, than did larger'ones,:o.2 as DEX'rnose 1.4L A9: no PHAGE A9: PHAGE ADDED AT 6 AT M.O.I. OF 111 . A 12, 010. .8 ' 9 o 0.3 . A 0.6 . o A 04. f‘i%’/ o / I. 0.2 1 n 1 I l j 3 5 7 9 11 13 Figure 18. Interaction of Ea110R and phage PEa1(h) at dif- ferent concentrations of glucose in Yurewicz broth. 114 like those from similar experiments carried out in nutrient broth (Figure 16 and (107)). Phage PEAl(h) was better able to lyse Ea110R when grown in DM-17 containing 0.2% or 1% glucose at all )LILI. tested. Figure 19 shows the results of the experiment with a MJlJ. of 1/1 with DM-17 containing 0.2 % and l % dextrose. Phage PEAl(h) was able to lyse Ea110R in DM-17 liquid medium but not in NBGYE or Yurewicz broth. PEa1(h) was able to completely lyse Ea110R grown in DM- 17 broth when added to cultures at several points during exponential growth. The ability of PEAl(h) to lyse Ea110R in DM-17 was not influenced by the growth stage of the culture at the time the phage were added, within the range tested (Figure 20). Effect bf Divalent Cations 9b Growth b£_PEa1(h) - CaClz added to Davis broth at concentrations of 0.1 - 2.5 mM had no effect on the lysis of Ea110R by PEaIUfi (Figure 21). The results are indistinguishable from the results of the same experiment carried out in medium lacking CaC12 (Figure 20). When MgSO4 was added at concentrations greater than 0.4 mM, the concentration used in Davis broth, no effect on lysis of Ea110R by PEa1(h) was observed. When no MgSO4 was included in Davis broth the growth of Ea110R was very restricted although the culture was apparently lysed by PEa1(h). The growth of Ea110R and its lysis by PEa1(h) was 115 A A: 1 % DEXTROSE 9,0 : 0.2% DEXTROSE .08 AOzNOPHAGE _ A,<>: PHAGE ADDED AT M.O.I. 1/1 . 0.2 O'D'eoo Edgun319. Interaction of Ea110R and phage PEa1(h) at different concentrations of glucose in DM-17 broth. 116 1L8 / _o.5 O,A,l PHAGEADDEDAT ' 5 {AT M.O.I. OF 1/3 8 O NO PHAGE 9 q _o.4 o A i Q 1... 1 \ __o.1 b ). I F L- HOURS Figure 20. Effect of growth stage on lysis of Ea110R by PEa1(h) in DM-17 medium. OJ%OO 117 0.8 0.7 0.6 O,A,l PHAGE ADDED AT M.O.I. or 1/3 O NO PHAGE 0.5 I O V 0.4 A 0.3 9 \ 0.2 . 9 0.1 l | I A o 2 4 6 8 10 HOURS ' Figure 21. Effect of growth stage on lysis of Ea110R by PEa1(h) in DM-17 medium supplemented with 1mM CaC12 118 unaffected by the addition of ZnSO4 to concentrations rang- ing from 2-100 uM (Figure 22, A&B). Although neither CaClZ nor ZnSO4 had any effect on lysis of Ea110R by PEa1(h) when added to Davis broth separately, when present together in Davis broth they seemed to greatly enhance lysis. In this experiment the culture grown in Davis broth without added CaC12 or ZnSO4 only lysed partially'and then recovered as was the case for cultures grown in Yurewicz broth or in rich media (Figure 23). The culture grown in the presence of 1 mM CaC12 and 100 uM ZnSO4 lysed completely. 119 6 _ AHOODM 211H 5 _ 0 PHAGE ADDED ATQ 4 ATMDI or 1/2 0 no PHASE 3 _ 1 g 2 _ O O P o d o o _l 6- Brno Zn” 5. A PHAGE ADDED 4 _ AT u.o.1.or 1/2 3 A G 2_ Figure 22. Effect of ZnSO on the lysis of Ea110R by PEa1(h) in DM-17 medium. (A). DM-17 supplemented with 100 uM ZnSO4. (B). DM-17 not supplemented with ZnSO4. 120 0.9 _o.a O,A PHAGE ADDED ATQ AT no.1. 112 0,A NO PHAGE A,A WITH Zn“and Ca” 0.0 WITHOUT Zn“and c.” _0.7 O) _O.6 0.5 0'0' 600 n m 0.4 _O.3 HOURS Figure 23. Apparent synergistic effect of CaCl2 and ZnSO4 on lys1s of Ea110R by PEa1(h) in DM-17 medium. DISCUSSION The results confirm Ritchiefls observation that PEAl(h) does not completely lyse Ea110R when grown in NBGYE (Figure 15 and 105). Yurewicz et. al. (137) purified a polysaccharide depo- lymerase from bacteriophage infected Aerobacter aerogenes grown in a liquid mineral salts medium supplemented with casamino acids and glucose..Their phage completely lysed the bacterial culture within three hours after inoculation with the phage at a M.O.I. of 1/25. When Ea110R was grown in Yurewicz broth and inoculated with PEa1(h), the phage failed to lyse the bacterial culture (Figure 16), even at a MJLI. of 1/1, just as was the case when NBGYE was the growth medium. In contrast to the situation observed with other li- quid media, phage PEa1(h) was able to completely lyse a culture of Ea110R growing in D-17 medium (Figure 18). Lysis ++ ++ was not influenced by either Ca or Zn ions when they were added singly to D-17 broth (Figures 20 and 21). However the presence of Ca++ and Zn++ ions together in D—17 did promote lysis in one experiment (Figure 22). The reason for 121 122 this apparent synergistic interaction of the ions without any effect when the ions were present singly is not under- stood. As can be seen in Table 1, the composition of Yurewicz broth and DM-17 are quite similar, yet PEAl(h) can lyse Ea110R in DM-17 but not in Yurewicz's broth. The difference in glucose concentration was shown not to be significant since more complete lysis occured in DM-17 containing either 0.2% or 1.0% glucose than in Yurewicz's broth at either glucose concentration (Figures 17-19). The very slight dif- ference in pH between the two media appears insignificant, but the difference in total ionic strength of the two media may be significant (3). Since phage PEa1(h) lyses Ea110R much better in DM-17 broth than in other liquid broths, this broth should be useful in producing PD for purification and characterization. The conclusion of Ritchie (107), that PEa1(h) requires intact capsular polysaccharides for infection of E_._ amylovo- £2! is similar to that of Stirm (121) and Bayer (16). This model is supported by the observation reported in chapter 2 that a strain Ea110R which contained and expressed the cloned polysaccharide depolymerase gene of PEa1(h) produced turbid, not clear plaques on DM-17 agar. This result mimics in a solid medium the rapid growth of phage resistant bac- teria observed in NBGYE or Yurewicz broth which resulted in very turbid cultures after phage infection, presumably due 123 to the enzymatic removal of capsular polysaccharides re- quired for phage binding in each case. BI BLIOGRAPHY 10. B I BLI OGRAPHY Abe, M. and S. Higashi. 1982. B-Glucosidase and B- Galactosidase from the periplasmic space of Rhizobium trifolii cells. J. Gen. Appl. Microbiol. 28:551-562. Abe,M., Sherwood, J.E.,Hollingsworth, R.I., and Dazzo, 52B" 1984. Stimulation of clover root hair infection by lectin-binding oligosaccharides frtmTthe capsular and extracellular polysaccharides of Rhizobium japonicum.;L Bacteriol. 160:517-520. Adams, ILH. 1959. BacteriOphages. Interscience Publish- ers, Inc. New York. Adams, M.H. and B.H. Park. 1956. An enzyme produced by a phage-host cell system 11: The properties of the polysaccharide depolymerase. Virology 2: 719-736. Anraku, Y. and LJL Heppel. 1967. On the nature of the changes induced in Escherichia coli by osmotic shock. J. Biol. Chem. 242:2561-2569. Avery, CLT. and R. Dubos. 1931. The protective action of a specific enzyme against type III Pneumococcus infection in mice. J. Exp. Medicine 54:73-89. Ayers, AHR., S.B. Ayers, and RJN. Goodman. 1979. Extracellular polysaccharide of Erwinia amlovora: a correlation with virulence..App1. Environ. Microbiol 38: 659-666. Barnet, Y.M. and B. Humphrey. 1975. Ex0polysaccharide depolymerases induced by Rhizobium bacteriophages. Can J. Microbiol. 21:1647-1650. Bartell, In, T. Orr, and C. Lam. 1966. Polysaccharide depolymerase associated with bacteriophage infection. J. Bacteriol. 92:56-62. Bartel, P., G.K.H. Lam, and T.E. Orr. 1968. Purification and properties of a polysaccharide depolymerase associated with phage-infected Pseudomonas aeruginosa. J. Biol. Chem. 243:2077-2080. 124 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 125 Bartell, P.F. and T.E. Orr. 1969. Origin of polysaccha- ride depolymerase associated with bacteriophage infec- tion.£L Virol.3:290-296. Bartell, IhF., T.E. Orr, J}F. Reese, and T. Imaeda. 1971. Interaction of Pseudomonas bacteriophage 2 with the slime polysaccharide and lipopolysaccharide of Pseudomonas aeruginosa strain B1. J. Virol. 8: 311-317. Bayer, M.E. 1968. Adsorption of bacteriophage to adhesions wall and membrane of Escherichia coli. J. Virology 2: 346-356. Bayer, M.E. 1968. Areas of adhesion between wall and membrane of Escherichia coli. J. Gen. Microbiol. 53:395- 404. Bayer, M.E. and H. Thurow. 1977. Polysaccharide capsule of Escherichia coli: Microscope study of its size, structure, and sites of synthesis. J. Bacteriol. 130:911-936. Bayer, M.E., H. Thurow and M.E. Bayer. 1979. Penetration of the polysaccharide capsule of Escherichia coli (B1161/42) by bacteriophage K29. Virology 94:95-118. Beckwith, J., T. Silhavy, H. Inouye, H. Shuman, M. Schwartz, 8. Emr, P. Bassford and E. Brickman. 1978. The Mechanisms of localization of proteins in Escherichia ggli. ib Transport of macromolecules in cellular sys- tems, eda SEC. Silverstein. pp.299-314. Berlin:Dahlem Konferenzen. Beer, S.V., T.M. Sjulin and H.S.1Aldwinckle. 1983. Amylovorin-induced shoot wilting: lack of correlation with susceptibility to Erwinia amylovora. Phytopathology 73:1328-1333. Bennett, RNA. 1980. Evidence for two virulence determin- ants in the fire blight pathogen Erwinia amylovora. J. Gen. Microbiol. 116:351-356. Bennett, RJL, and E. Billing. 1978. Capsulation and virulence in Erwinia amylovora. Ann. Appl. Biol. 89:41— 45. Bennett, R.A. and E. Billing. 1980. Origin of the polysaccharide component of ooze from plants infected with Erwinia amylovora. J. Gen. Microbiol. 116:341-349. Bessler, W., E. Freund-Molbert, H. Knufermann, C. Rudolph, H. Thurow and S. Stirm. A bacteriOphage-induced 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 126 depolymerase active on Klebsiella K11 capsular polysaccharide. Virology 56:134-151. Bessler, VL, F. Fehmel, E. Freund-Molbert, IL Knufermann and S. Stirm. 1975. Escherichia coli capsule bacteriophages IV: free capsule depolymerase 29. J. Virol. 15:976-984. Billing, E. 1960. An association between capsulation and phage sensitivity in Erwinia amylovora. Nature 186:819- 820. Billing, E. 1984. Studies on avirulent Ebwibia amylovora. Acta Horticulturae 151:249-253. Birnboim, PLC., and J. Doly. 1979. A rapid alkaline extraction procedure for screening recombinant plasmid DNA. Nucleic Acids Research 7:1513-1522. Blumenkrantz, N. and G. Asboe-Hansen. 1973. New method for quantitative determination of uronic acids. Anal. Biochem 54:484-489. Bolivar, F., R.L. Rodriguez, P.J. Green, M.C. Betlach, ILL. Heyneker, and HJW. Boyer. 1977. Construction and characterization of new cloning vehicles, II. A multi- purpose cloning system. Gene 2:95-113. Boyer, ELW. and D. Roullard-Dussoix. 1969. A complemen- tation analysis of the resriction and modification of DNA in Escherichia coli. J. Mol. Biol. 41:459-472. Bradshaw-Rouse, CLJ., M.H. Whatley, DAL. COplin, A. Woods, L. Sequeira and A. Kelman. 1981. Agglutination of Erwinia stewartii strains with a: corn agglutinin: correlation with extracellular polysaccharide production and pathogenicity. Appl. Envir. Microbiol. 42:344-350. Brosius, J. 1984. Plasmid vectors for the selection of promoters. Gene 27: 151-160. Brosius, J. 1984. Toxicity of an overproduced foreign gene product in Escherichia coli and its' use in plasmid vectors for the selection of transcription terminators. Gene 27: 161-172. 33.Chakrabarty, A.M., J.F. Niblack and LC. Gunsalus. 34. 1967. A phage initiated polysaccharide depolymerase in Pseudomonas putida. Virology 32:532-534. Chatterjee, A.K. amd M.P. Starr. 1972. Transfer among Erwinia spp. and other enterbacteriaceae of antibiotic 35. 36. 37. 38. 39. 40. 41. 42. 43. 44. 45. 46. 47. 127 resistance carried on R factors. J. Bacteriol. 112:576- 584. Chatterjee, A.K”, and M.P. Starr. 1980. Genetics of Erwinia species. Ann. Rev. Microbiol. 34: 645-676. Civerolo, ELL., and ELL. Keil. 1969. Inhibition of bacterial spot of peach foliage by Xanthomonas pruni bacteriophage. PhytOpathology 59:1966-1967. Civerolo, ELL. 1970. Comparative relationships between two Xanthomonas pruni bacteriophages and their bacterial host. Phytopathology 60:1385-1388. Civerolo, EgL. 1973. Relationships of Xanthomonas pruni bacteriophages to bacterial spot in Prunus Civerolo, E.L. 1974. Temperature effects on the relationships between Xanthomonas pruni and its' virulent phages. Phytopathology 64:1248-1255. Civerolo, EJL 1976. Phage sensitivity and virulence relationships in Xanthomonas pruni. Physiol. Pl. Path. 9:67-75. Clark, AmJ. and A.D. Margulies. 1965. Isolation and characterization of recombination-deficient mutants of Escherichia coli. Proc. N.A.S. USA 53:451-459. Clewell, ILB. 1972. Nature of ColEl plasmid replication in Escherichia coli in the presence of chloramphenicol. J. Bacteriol. 110:667-676. Cooper, TLG. 1977. The Tools of Biochemistry. John Wiley and Sons. New York. Costerton, J;W., and R.T. Irvin. 1981. The bacterial glycocalyx in nature and disease. Ann. Rev. Microbiol. 35:299-324. Davis,R.W., D. Botstein and J.R.Roth. 1980. Advanced Bacterial Genetics: A Manual for Genetic Engineering. Cold Spring Harbor Laboratories, Cold Spring Harbor, New York. Dazzo, F.B., and W.J. Brill. 1979. Bacterial polysaccharide which binds Rhizobium trifolii to clover root hairs. J. Bacteriol. 137:1362-1373. Doubet, ILS., and R.S. Quatrano. 1984. Properties of alginate lyases from marine bacteria. Appl. and Environ. Microbiol. 47:699-703. 48. 49. 50. 51. 52. 53. 54. 55. 56. 57. 58. 59. 128 Dubois, M.K., KAA.