.—__._.,_.._... _, _ ,, ’k-"‘""—“l‘~“-'-"" "a.“ "I‘Il‘l'—' -. 1 . -~. ~1 . 1 .1- 21 ‘ . "' .. I I "m "I II ‘. w). 'I:. :II' '1: .I'I 2; I1.. 'N’ s . ‘ . I' I‘V - {'7’ 'I .. _ III III 1 0 . “I 1 "'11 I. .1 . II II '1 11 1 )5. . I ' I IIII ' ' IIIII ' I‘" ”F; '1 :I III 'III, I ‘II1 ‘-'-1 ”‘2”; III II 1‘ ‘ I; “1"" 1‘ I I‘II I . "1 ' - '1'." ' ‘1'. ' '. 2/122 1' ' ~ - ”I ' d'2l 1" ' - 'l ‘. '1" -. ’11 1 IIIJII III.“ 3“? .IIIII ‘1‘ it“... II.II ‘II -!'-I.IIIIIII1I"E .11 'I'-'.‘I 'IQIJI ,IIII I-IIII'I II} n 391;,” 5* Iii] Wil' 11+. . I'I' I1} '11:)". '1'. ...' 1II'II‘III1I2 I‘VI' .EI:\"1I m1)”; I'I'I'II‘ fxflénw I|"‘1/I.||)1’urII-IIII‘ "I1IIII‘I'1I1'I‘I it“! MW £11... 11".I"”’vlfi1..1"'IJA'JJ'I' 11‘ I'I"" I. 9"“! III II ”LEW I }. ,I‘II'I g} II :1. 1’111"?IJ:I‘VIIII'IIIIIII“! I ll II'1I(I' “I I 3: 1:‘II:sI:II:I::I‘ T’F'I'ULIIII Iiml'l‘ll.“ IKII I 1 1 $9 ’I11III'H'EII‘I‘ III3IIII.IIII1I1:II.II'I IIIIIIIIIII . IIIIIIIIII I ' ' . I1 I1I" ' 1 1 I ' NW '1'. 11 I IIIII I I ‘4Ill:|\0..f I. l 1"I ~21 “‘ 1121111. ..'."1:.,;“".'1"‘ P1141 ' .‘ ' * ‘ 1' I )IIIII' III ”'3 I 1‘.."l .4$j.,‘"1:1"'\:I II I 1 .1. .- . 1211 1.4;211 1' , 1,,iI1‘1',1;II1I'-I~.'..:1 13‘ H .‘I‘I'f, ‘1 ' " 31:191. ' . 5 1:1: .I.‘. II, \l'l:'"!tIII .1” I In'I 2"1 .I I ' . 1 I 1 ' 1;' 1 1. ' . . l | 1- ‘1 ‘1' ’_ ':_ ! -‘|l I 1 '1’1"'1‘}‘" ' 3'3] 3;”. ”(MW ll "" ' 1:053" '."1.| 1' . . I ‘5 \ . . ' ' I "1 4.1! ,I III-‘IIIIJ IJII’I ."l ' 1 ,1'II'HII‘I'1II.II.I I IIllII‘I: ' I‘I.A1IIIII 11 'I‘ .I ,1- -.‘ ‘ I ll:II H1. tvdu I I ‘ ‘. I .' . I ‘ V ‘ 'V‘I‘ I‘I‘III‘ .11. 1|,.I'-‘I'I':IIIIIJ ".I'II' " IMP“ " III.I‘ ' IIj’l'I‘F ”R ‘v"11.f".":'5. I; i " I. ‘ ' ‘ .IIflIr 'D‘II I’IIJ:I.II . . I. :‘I [IIILIIII III. III .IIrII'IILI II-' ‘I I 1| '4 ' '1': --I '1 . 1 I.I I | IV: I‘ . 1"~ it ‘ '2: \1‘19 1 2 '.' "I‘ .: III l‘I .'3... '21-: 4| ". r 1'- 1 I v I "1:..1 . 'I’1‘11‘1' HHIII‘II IerngI'IIIfi 1M ~I,I'I,I‘”.l II‘LN' ‘IIl‘ II‘IizI H4042.» . . MI .1 7*.11'11111'131‘ ' . 1' '~'1 ‘ ”“13! "115.1% ,.'.1Q.'"':"'.2'1 "1’; '14' I .1'- ' I ‘IlIIIIIII‘l. II . - .."- .I‘ 1 . .. 2 15111 1111.. 111$ 111111311" 112113 ' 1 1 III II'.I__. #1-. This is to certify that the dissertation entitled ISOLATION AND CHARACTERIZATION OF SYMBIOTICALLY DEFECTIVE MUTANT STRAINS 0F RHIZOBIUM TRIFOLII AND RHIZOBIUM MELILOTI presentedby Alicia E. Gardiol has been accepted towards fulfillment of the requirements for Ph.D. degreein Microbiology E. Public Health $4,45W Major professorv 0 Date Nov. 15,1285 ”(III-n. Am..— -' . - '- Ill . n . , 042771 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. ISOLATION AND CHARACTERIZATION OF SYMBIOTICALLY DEFECTIVE MUTANT STRAINS OF RHIZOBIUM TRIFOLII AND RHIZOBIUM MELILOTI BY Alicia E. Gardiol AN ABSTRACT OF A DISSERATION Submitted to Michigan State University in partial fulfillment of the requirements for degree of DOCTOR OF PHILOSOPHY Department of Microbiology and Public Health 1985 ABSTRACT ISOLATION AND CHARACTERIZATION OF SYMBIOTICALLY DEFECTIVE MUTANT STRAINS OF RHIZOBIUM TRIFOLII AND RHIZOBIUM MELILOTI by Alicia E. Gardiol Three classes of mutants were obtained after TnS mutagenesis of Rhizobium trifolii 0403 rif. Strains 738 and 755 (defective in. nitrogen fixation) and 251 (partially defective in nitrogen fixation) produced the same plasmid pattern as wild type upon electrophoresis whereas 43 and 308 (unable to nodulate) contained a deletion in the gym plasmid. 251 attached better and 43 and 308 attached less to clover root hairs than did the wild type. Strain 251 with a single Tn5 insertion in the gym plasmid was agglutinated better and 43 and 308 less by trifoliin A than wild type. Strain 251 capsular polysaccharide (CPS) differed in depolymerization rate and non-carbohydrate substitutions from wild type CPS. These results indicate that CPS non- carbohydrate substitutions are important in g; trifolii 0403 El: attachment to clover root hairs and lectin-binding ability. CPS was isolated from 3; trifolii 843 and TnS mutants in modulation genes of the symbiotic plasmid. CPS from mutants in region I(Hac) differed from wild type CPS in depolymerization rates and levels of pyruvic and acetic acid Alicia E. Gardiol substitutions and mutant strains had a significantly lower ability to bind the clover lectin. An _i_n 11—152 assay to measure CPS pyruvyl transferase activity (CPTO :hi 3; trifolii was developed. Pyruvylation occurred at the lipid-bound oligosaccharide intermediate stage. CPT was measured for two wild type and symbiotically defective mutant strains. CPT was affected by TnS mutations in the gm plasmid and by clover root exudate suggesting that functions related to CPS pyruvylation may be encoded in this plasmid. The effect of succinate metabolism on growth and bacteroid differentiation of Rhizobium meliloti was investigated with wild type (LS-30) and succinate dehydrogenase mutant (UR6). UR6 was defective in bacteroid differentiation (Bad-) in 3139. lg 31539 succinate effects were concentration dependent. At low concentration, succinate ‘was ‘utilized. preferentially' before glucose. At higher concentration, succinate decreased growth yield and induced bacteroid-like cell morphology in 15% of the cell population. These effects were observed for L5-30 but not for UR6 strain, suggesting that a functional TCA cycle is necessary for these in vivo and in vitro succinate effects. ACKNOWLEDGMENTS I wish to thank the Department of Microbiology for giving me the opportunity and support to study at Michigan State University. I would like to thank Dr. Frank B. Dazzo for continuous advice, encouragement and support. I especially wish to acknowledge my Guidance Committee, Drs. Barry Chelm, Kenneth Nadler, Harold Sadoff, and Loren Snyder. I 'would like tx> thank. the Center' for International Programs for the Thoman Fellowship award and for supporting my participation at the Advanced Bacterial Genetics Course at Cold Spring Harbor Laboratory. I gratefully acknowledge Drs. Barry Rolfe and Michael Djordjevic for providing* mutant strains, Dr. Rawle Hollingsworth for running NMR samples, and Dr. Georges Truchet for performing electron microscopy. I am grateful to Drs. P. T. Magee and Dr. R. Hausinger for providing laboratory facilities, Dr. K. Nadler for guidance on Tn5 mutagenesis, Dr. L. Snyder for directing my rotation project, Dr. R. Reusch for help on membrane proteins isolation, Dr. M. Howe for providing bacterial strains, and Dr. M. Clancy, B. B. Magee and D. Lehman for their help with 32F experiments. I especially thank Dr. Mikiko Abe for advice on oligosaccharide isolation, Dr. John Sherwood for providing clover lectin and Estelle Hrabak, Dave Gerhold, H. S. 11 Pankratz, Lori Herr, Kathy Smith, H. Yang, Jaime Maya- Flores, and Fatima Lima for their help. This research was supported by grants NSF-PCM80-21906, and NIH-GM 34331-01 to Dr. Frank Dazzo. I acknowledge the Fulbright Commission (Uruguay) for a travel fellowship. I feel this is Ruben and Natalia's work as well as mine. I could not have done it without them. I wish to dedicate this work with my deepest admiration to my parents. iii TABLE OF CONTENTS PAGE LIST OF TABLES ....................................... viii LIST OF FIGURES ...................................... x INTRODUCTION ......................................... 1 Stages in the symbiotic process ................. 1 Rhizobium — lectin interaction .............. (.... 2 Rhizobium polysaccharides that bind the clover lectin .............................. 3 Attachment to clover root hairs ................. 5 Rhizobium symbiotic genes ....................... 7 Polysaccharide synthesis ............... ......... 10 List of references ............. . ........ . ...... . 13 CHAPTER I. TNS INSERTION IN THE SYMBIOTIC PLASMID ALTERS THE SURFACE PROPERTIES OF RHIZOBIUM TRIFOLII 0403 RIF .................... . 20 Abstract ...................................... .. 20 Introduction ..................... . .............. 22 Materials and methods ........................... 25 Bacterial strains and phages ............... 25 Media and growth conditions ..... . ....... ... 25 Transposon (TnS) mutagenesis.. ...... ...... 26 Screening for symbiotically defective mutant strains ............ ... ......... 26 Rhizobium DNA Isolation ........... . ........ 27 iv Restriction endonucleases and DNA hybridization ......................... 27 Plasmid profile analysis and location of TnS ................................... 28 Root hair interactions ..................... .29 Protein determination ...................... 30 Bacterial agglutination assay .............. 31 Purification of phage Mu and preparation of Mu DNA probe ....................... 31 Nodulation and host range studies .......... 32 Isolation of polysaccharide depolymerase PD-I... ..... ............... ........... 33 Oligosaccharide isolation.. ........ . ....... 33 Kinetic study of CPS depolymerization rates................................. ‘34 1 H-NMR analysis of oligosaccharides ........ 34 Oligosaccharide glycosyl composition ....... 35 Results..... ............... .... ................. 36 Discussion... ................................... 53 List of references......................... ..... 59 CHAPTER II. ALTERATIONS IN CAPSULAR POLYSACCHARIDE OF TN5 - INDUCED MUTANTS IN THE RHIZOBIUM TRIFOLII NODULATION REGION.... ............... ... 65 Abstract........... .................... ......... 65 Introduction....... ............................. 67 Materials and methods...... ........ .. ........... 69 Bacterial strains........... ........ . ...... 69 V Media and growth conditions ................ 69 Deformation studies ........................ 69 Protein determination ................. ..... 69 Lectin binding studies .................. ... 69 Methods for isolation of polysaccharide depolymerase PD-I ..................... 73 Kinetic study of CPS depolymerization rates ........................... . ..... 73 lH-NMR analysis of oligosaccharides ........ 74 Oligosaccharide composition....... ......... 74 Results............................... ......... . 75 Discussion ...................................... 85 List of References..... ........ .. ............... 88 CHAPTER III. RHIZOBIUM TRIFOLII CAPSULAR POLYSACCHARIDE BIOSYNTHESIS. PYRUVYLATION OF LIPID-BOUND SACCHARIDES............... ...... . 90 Abstract... ...... .......... ....... .............. 90 Introduction.................................... 91 Materials and methods..... ....... . .............. 93 Bacterial strains ...... . .......... . ........ 93 CPS pyruvylation assay..................... 93 Incorporation of radioactivity into lipid-bound saccharides............... 95 Analysis of labeled product obtained from [1-14 C]PEP. Mild acid hydrolysis...... 95 Analysis of aqueous phase soluble moiety... 95 Preparation of root exudate ........... ..... 96 vi Results ......................................... 98 Discussion ...................................... 106 List of references ...... . ....................... 110 CHAPTER IV. SUCCINATE METABOLISM AS RELATED TO GROWTH, BACTEROID DIFFERENTIATION AND FUNCTION IN RHIZOBIUM MELILOTI .................. 112 Abstract ...................... . ........ .... ..... 112 Introduction.. .................... .. ..... ....... 114 Materials and methods ...... . ............... ..... 116 Bacterial strains... ............... ........ 116. Symbiotic phenotype. ....... ....... ...... ... 116 Electron and light micrOSCOpy of nodules... 116 Growth conditions .......................... 117 Cell-free extract preparation .............. 117 Protein determination ...................... 117 Glucose utilization ........................ 117 Succinate utilization ...................... 117 Enzyme activities.. ........................ 118 Electron microscopy of cells............... 118 Oxygen consumption ....... .................. 119 Results......................................... 120 Discussion.......... ...... ...................... 132 List of references.............................. 136 SUMMARY 140 List of references...................... ...... . 152 CHAPTER TABLE 1 1 LIST OF TABLES Attachment of wild type R; trifolii 0403 rif and mutant strains to white clover root hairs. Orientation of attachment of R; trifolii 0403 El: and mutant strains to clover root hairs. Shepherd crooks and infection threads induced by 3; trifolii 0403 rif and mutant strains. Trifoliin A - agglutinating activity of R; trifolii 0403 rif and mutant strains. Nodulation patterns of R. trifolii 0403 El: (wt) and R; trifolii 251 (ms) on Trifolium repens, Trifolium subterraneum, Trifolium fragiferum, and Trifolium pratense. Depolymerization rates of CPS from R; trifolii 0403 rif and R; trifolii 251. Non-carbohydrate composition of oligosaccharides obtained by PD-I depolymerization of CPS isolated from 3; trifolii 0403 El; and R; trifolii 251. List of bacterial strains used in this study. 'In situ binding of trifoliin A to R; EEifolii in clover slide cultures. Depolymerization rates of CPS from 3; trifolii 843 (wild type) and TnS mutant strains in three nodulation regions using enzyme PD-I. viii PAGE 40 42 43 44 45 50 52 71 77 81 Non-carbohydrate composition of oligosaccharides obtained by PD-I depolymerization of CPS from R; trifolii 843 and mutant strains in three nodulation regions. 83 Incorporation of radioactivity into lipid-bound saccharides in cells of R; trifolii 843 from labeled UDP-sugars. 99 Mild acid hydrolysis of labeled product. 101 Pyruvylation of lipid-bound saccharides in cells of R. trifolii 843 (wild type) and Tn5 mutant strains. 103 Pyruvylation of lipid-bound saccharides in cells of R. trifolii 0403 rif (wild type) and R; trifolii 251. 105 mannitol uptake and mannitol dehydrogenase activities in L5-30 and UR6 strains. 130 ix CHAPTER -FIGURE 1 1 LIST OF FIGURES PAGE Autoradio 32 - A-- gram of P labeled ..TnS hybridized to digested total DNA. Egg RI digestion. (A) Lanes: a, R; trifolii 738; b, R; trifolii 43; c, R; trifolii 0403 Eli; d, R; trifolii 308; e, R; trifolii 755; f, R; trifolii 251. Hind III digestion. (B) Lanes: a, R; trifolii 0403 rig; b, R; trifolii 251. Plasmid patterns by Eckhardt agarose gel electrophoresis. (A) Lanes: a, R; trifolii 308; b, R; trifolii 0403 rig; c, R. trifolii 43. (B) a, R. trifolii 738; b, R. trifolii 755; c, R. trifolii 251; d, R. trifolii 0403 rif. Plasmid profile and location of Tn5 insertion in R. trifolii 251. (A) Plasmid Eckhardt g l. (B) Autoradiogram of P-labeled A::Tn5 hybridized to plasmid Southern blot. Phage Mu purified by CsCl equilibrium gradient centrifugation. Depolymerization rates of CPS from R; trifolii 0403 Eli, R; trifolii 251 and depyruvylated CPS with enzyme PD-I followed by increase of absorbance at 235 nm under saturating conditions of substrate and identical protein concentration. Nodulation regions in the gym plasmid of R; trifolii 843. 14 Kb Hind III restriction fragment showing location of Tn5 insertions. The regions were established based on the symbiotic phenotype of the mutant strains (M. A. Djordjevic et al.,Mol. Gen. Genet., submitted). 39 47 48 49 7O Typical root hair response induced by wild type and mutant strains. Strains are R; trifolii 843 (wild type), R; trifolii 851 (39g I::Tn5), R. trifolii 262 (nod II::Tn5), R. trifolii 297 (nod III::Tn5), and R. trifolii 845 (psym ) . 76 Ex-planta binding of trifoliin A to wild type and symbiotically defective mutants of R; trifolii 843. Cells were grown on 8111 agar plates for 3 to 10 days and heat fixed to slides. Binding to trifoliin A was examined by indirect immunofluorescence and recorded as the percentage of cells showing fluorescence. R; trifolii 843 (wild type), R; trifolii 851 (39g (1)::Tn5), and R; trifolii 845 (pSym ). 78 Kinetic study of the relative rates of depolymerization of CPS from R; trifolii 843 and mutant strains in region I, by PD-I enzyme under identical conditions. Strains are R; trifolii 843, R; trifolii 252, and R; trifolii 851. 80 A 1H-NMR spectrum of the oligosaccharides produced by depolymerization of CPS from a 5-day— old culture of R; trifolii 843, using PD-I enzyme. Peaks represent (A) H-4 of 4-deoxy-L-threo hex-4- enopyranosyluronic acid, which results from /3-elimination of glucuronic acid catalyzed by PD-I, (B) methylene and (B ) methyl of 3-hydroxybutanoic acid, (C acetate, and (D) Pyruvate. 82 Gel filtration of the radioactive moiety soluble in aqueous phase using Bio-Gel P6. Fractions were collected and aliquots monitored for radioactivity. Unlabeled saccharides were determined by the phenolsulfuric acid method. 102 Ultrastructure of root nodules on alfalfa 6 weeks after inoculation with UR? strain. (A)(B) Enlargement of plant cells in the central zone filled with bacteroids. Results with wild type L5-30 were identical. 121 Ultrastructure of root nodules on alfalfa induced 3 week after inoculation with UR6 strain. (A) Some degenerated bacteria in the proximal infection zone. (B) Enlarged profiles of rough endoplasmic reticulum in the distal infection zone. (C) Lysosomes (arrows) in distal infection zone. (D) Electron-dense granules (arrow) associated with bacteria in distal infection zone. (B) Bacteria undergoing degeneration in host cells in distal central zone. Note the ultrastructurally well preserved host cytoplasm. 122 Growth of R; meliloti L5-30 (A) and UR6 (B) in minimal medium containing 10 mM glucose plus 10 mM succinate; 10 mM glucose; or 10 mM succinate. Klett units; glucose concentration in the culture medium (mM). 123 Growth of R; meliloti L5-30 (A) and UR6 (B) in minimal medium containing 10 mM mannitol plus 10 mM succinate; 10 mM mannitol; or 10 mM succinate. Klett units; succinate concentration in the culture medium (mM). ~125 Growth of R; meliloti L5-30 in mannitol minimal medium containing (A) succinate or (B) malate. In (A), labels are 27 mM mannitol only; 27 mM mannitol plus 20 mM succinate; 20 mM succinate only; in (B), they are 27 mM mannitol only; 27 mM mannitol plus 20 mM malate; 20 mM malate only. 126 Cell morphology of L5-30 strain grown in different media. Culture samples removed at indicated times (arrows) in Figure 5, were observed by TEM. L5- 30 grown in mannitol (a); mannitol plus succinate (b)(c); mannitol plus malate (d). 127 Growth of R; meliloti UR6 in mannitol minimal medium containing (A) succinate or (B) malate. (A) 27 mM mannitol only; 27 mM mannitol plus 20 xii mM succinate; 20 mM succinate only. (B) 27 mM mannitol; 27 mM mannitol plus 20 mM; 20 mM malate. 128 Cell morphology of UR6 strain grown in different media. Culture samples removed at indicated times (arrows) in Figure 7, were observed by TEM. UR6 grown in mannitol (a); mannitol plus succinate (b); mannitol plus malate (c). 129 Decrease of dissolved oxygen in cell suspensions of L5-30 and UR6 strains. Cells were grown as for Figures 5 and 7. At indicated times (arrows) shaking was stopped, a NBS oxygen electrode was introduced into the cultures and the decrease of dissolved oxygen was followed as a function of time. The oxygen electrode was calibrated with 5% NaZSO (0%) and distilled water saturated with air (100%) at 25°C. 131 xiii INTRODUCTION Stages in the symbiotic process Rhizobium is a genus of gram-negative soil bacteria that can infect and nodulate legumes. The establishment of a functional symbiosis is a multi-step process. Rhizobium motility' and chemotaxis may be important in interstrain competition for nodule sites on the root (5). Plant root hairs are deformed early in the infection process by unknown substances made by the bacteria (110). After recognition and attachment of the rhizobial symbiont to root hairs (28,26) the bacteria specifically infect the host cells, presumably by enzymatic degradation of the root hair cell wall (19). A tubular infection thread confines the invading bacteria as they enter the host cell (24). The infectibility of legume rOOt hairs is transient and maturing root hairs are important infection points in clover while in soybean, cowpea or alfalfa infections occur more frequently in newly emergent root hairs (13). The infection thread branches at the base of the root hair, penetrates the outer cortex of the root and stimulate the proliferation and enlargement of the inner cortex cells (66). This results in the formation of differentiated nodule tissue infected intracellularly by rhizobia released from the infection threads and surrounded by a peribacteroid membrane of plant origin. The bacteria 1 then differentiate into bacteroids and fix nitrogen (transform atmospheric nitrogen into ammonia) which is excreted into the cytosol of the host cells to be assimilated by the plant. In this symbiosis the bacteria receives photosyntates from the plant (24). The phenotypic code used for the different stages of this developmental process are: root colonization (Roc), attachment to root (Roa), root hair curling (Hac), infection thread formation (Inf), nodule initiation (Noi), bacterial release (Bar), bacteroid.(differentiation (Bad), nitrogen. fixation. (Nif), and nodule persistence (Nop) (108). Rhizobium - lectin interaction Lectins are sugar-binding proteins or glycoproteins of non-immune origin which agglutinate cells and/or precipitate glycoconjugates (51). It has been hypothesized that lectins are regulatory molecules, being receptors for a signal molecule that then elicits a response in different systems (28,73,76,79,82). Trifoliin A is a clover lectin which specifically binds to Rhizobium trifolii (33). This glycoprotein contains approximately 6 mol reducing sugar/mg protein, has a subunit molecular weight of 53,000 and an isoelectric point of 7.3 (33). The interaction between lectin and sugars has been studied by different methods including agglutination of bacteria (90), affinity chromatography (78), and binding to radioactively labeled saccharides (74). Trifoliin A agglutinates & trifolii by interacting with the capsular polysaccharide (CPS) (25) and lipopolysaccharide (LPS) (58). 2—deoxy-D-glucose and the - isomer of quinovosamine are haptens (27,58) of this lectin. Rhizobium polysaccharides that bind the clover lectin Isolation, purification and characterization of Rhizobium polysaccharides have recently been reviewed (20,42). There are several published structures of acidic heteropolysaccharides and homoglucans secreted by strains of Rhizobium trifolii and other rhizobia. Complete structures of Rhizobium lipopolysaccharies are run: yet known, mostly because of their exceedingly complex sugar composition. The first structure proposed for a rhizobial acidic extracellular polysaccharide (EPS) was a repeating octasaccharide of R; meliloti (60). Subsequently, saccharide structures for EPS of g; trifolii U226 and 5; trifolii 0403 were proposed (61,81). The same saccharide structure was also proposed for an EPS from two strains of R. leguminosarum and R; trifolii NA-30 (81), and also for one strain of R; phaseoli (43). A slightly different saccharide structure has been proposed for an EPS of g; trifolii 4S (4). Glycosyl composition and non-carbohydrate substitutions of R; trifolii 0403 oligosaccharide repeating unit obtained by cleavage of CPS with a polysaccharide depolymerase system has recently been reported (56, R. Hollingsworth, manuscript in preparation). CPS and EPS of plate-grown cultures of R; trifolii 0403 are not identical since they differ in degree of non-carbohydrate substitution (2). Rhizobium polysaccharides have different non- carbohydrate substitutions including pyruvate (41), acetate (41), 3-hydroxybutanoate (55), methyl (74), and succinate (54). These substitutions can change with culture age thereby changing the ability to be bound by lectins. The methylation of galactosyl residues of R; japonicum CPS during stationary phase of growth results in a reduction of its interaction with the soybean lectin (74). The levels of non-carbohydrate substitutions of the CPS of R; trifolii 0403 change with culture age and the lectin binding ability corresponds ix) these changes (2,91). The neutral glycosyl composition does not vary with culture age (91). Binding of trifoliin A to R; trifolii LPS is also culture age- dependent, being optimal in early stationary phase (58). A compound which increased when the LPS was able to bind trifoliin It was 2-amino-2,6-dideoxyglycose (quinovosamine) (58) . Pyruvate and acetate substitutions in the EPS of R; trifolii L158 and other fast-growing rhizobia have also been shown to change with culture age (18). These culture age- dependent changes in the ability to bind the host lectin have been shown for _R_._ japonicum (74), _R_. leguminosarum (107), and g; trifolii (32,58). The Rhizobium-lectin interaction can also be affected at the plant level (by addition of fixed nitrogen to the plant growth media), and by interactions between the plant and bacteria (12,31). LPS of R; trifolii 0403 (58) activates a host response in white clover seedlings which triggers root hair infection (29). At l./kg/seedling, these lectin-binding polymers bind to clover root hair tips and significantly enhance infection thread formation by R; trifolii 0403 when applied to seedlings. At ll)/ug/seedling, the LPS suppressed root hair infection (29). LPS from a serologically unrelated strain of R; trifolii (strain 28-2) also stimulated root hair infection of white clover by R; trifolii 0403 at l flg/seedling, whereas no stimulation of infection was obtained by pretreatment with LPS from R; meliloti F28, or from E; 391;. CPS from R; trifolii 0403 at 2.5 [Ag/seedling stimulated the formation of infection threads in white clover using a similar assay, and inhibited formation of infection threads in clover root hairs at higher concentrations (2). The EPS from the same culture of R; trifolii 0403 which was unable to bind the lectin did not display this biological activity on the infection process (2). Cyclic/@(l-2)glucan from the periplasmic space of R; trifolii 48 also was able to increase the number of infections (1). Attachment to clover root hairs The time-course and the orientation of attachment of R; trifolii to root hairs on white clover seedlings inoculated with encapsulated bacteria which bind trifoliin A uniformly has been examined. Specific reversible interactions involving trifoliin A (Phase I) were followed by irreversible interactions involving extracellular microfibrils (Phase II). A similar sequence of reversible and irreversible phases of attachment of Agrobacterium tumefaciens to plant cells has been prOposed (70). Phase I attachment can be subdivided into 3 steps (30). Phase 1A is the clumping of cells in random orientation at root hair tips. Phase IB involves erosion of the capsule of unattached cells by enzymes in root exudate. 1C attachment is initially randomly oriented and then predominantly polar. Granular, electron-dense aggregates and trifoliin A can be detected at the interface between the bacteria and the root hair surface in IA and 1C attachments. The enzyme(s) in clover root exudate which alter the capsular polysaccharide of R; trifolii during Phase 18 are antigenically unrelated to trifoliin A (31). Immunoelectrophoresis studies (31) suggest that the capsular polysaccharides of R; trifolii are cleaved into smaller, dialyzable fragments. Bhuvaneswari and Solheini (14) have proposed that these enzyme-mediated cleavages result in oligosaccharide products which induce clover root hair branching. Within hours 3; trifolii displays a pattern of attachment combining 1A + lC attachments on the same root hair. This attachment pattern is symbiont-specific and 2- deoxy-D-glucose inhibitable, and is present on approximately 93% of the infected root hairs examined 4 days after inoculation with R; trifolii 0403 (30). This orientation may provide the optimal distribution of bacteria for marked curling and successful infection of the root hair (44). The hapten 2-deoxy-D-glucose can inhibit the attachment of trifoliin A-binding' R; ‘trifolii cells to (clover. root hairs (27,30,112). Cells remain attached to clover root hairs when exposed to the shear forces of high-speed vortexing after 12h of incubation, what suggested a Phase II of firm attachment. At Phase II, extracellular microfibrils associated with the bacteria attached to the root hair surface can be seen by SEM. These are probably bundles of cellulose and/or fimbriae. Host-specific attachment has been demonstrated in R; trifolii - clover, B; japonicum - soybean, and R; leguminosarum - pea root systems reviewed in (28). Rhizobium symbiotic genes The development of a: nitrogen-fixing Rhizobium-legume symbiosis is as multi-step process that requires bacterial and plant gene products. Early genetic evidence showed that the chromosomes of R; trifolii, R; leguminosarum, and R_. phaseoli were essentially identical when in a series of genetic crosses the chromosome of R; leguminosarum was replaced section by section by the corresponding part of the genome from either 3; phaseoli or R; trifolii (16) and in no case was any of the interspecific recombinants altered in its host range. Several reports have shown that many of the bacterial symbiotic genes of fast-growing Rhizobium species are plasmid-borne and are located in one plasmid called for that reason gym plasmid which spans approximately 30 kilobase pairs (kb)(69). Root hair curling (Egg), nodulation (29g) (7,11,15,21,38,39,40,52,53,62,64,69,84,88,113) and nitrogen fixation (gig and fig) genes (l7,22,48,72,77,80,86,97,111) have been located in this megaplasmid. Egg genes for root hair attachment of R; trifolii to clover root hairs are also encoded on the gym plasmid (112) and the genetic information required for trifoliin A- binding, "Phase I" attachment, marked root hair curling (Shepherd crooks), root hair penetration and infection thread formation was expressed in pTi-cured A; tumefaciens containing the gym plasmid of R; trifolii 5035 (57, F. B. Dazzo, G. L. Truchet, and P. J. Hooykaas, Abstr. Annu Meet. Am. Soc. Microbiol. 1983, K9, p. 178). In R; meliloti, mgg genes are organized in two clusters. One appears to encode for conserved nodulation functions and the other affects host specificity (64). Other genetic models of nodulation also proposed that in R; leguminosarum, there is a core of conserved nodulation- specific genes and that the host range is determined by ancillary genes which are very closely linked (39). Three nodulation regions have been identified within a 14 kb Hind III fragment of the R; trifolii gym plasmid. Tn5 mutants in region I (Hac) are Hac-Nod- whereas mutants in region II (Superhac) are Hac++ and induce delayed and fewer nodules. Mutants in region III (Hsp) acquired the ability to nodulate peas and lost the ability to nodulate white clover efficiently (M. A. Djordjevic et al., Mol. Gen. Genet., in press). Functional conservation of n_od genes involved in root hair curling has been demonstrated in several species (37,64) and similarity at the DNA sequence level has also been shown (85,100). DNA sequence and complementation data have allowed the identification of four common n_og genes (nod A, B, C, and D) in By meliloti, Ry leguminosarum and R; trifolii (37,39,64,85,100, and Th (F. Egelhoff et al., in press). Mutant strains altered in the levels of polysaccharide synthesis and defective in nitrogen fixation have also been described (7,89). The reduction in EPS synthesis by mutants of R; japonicum (68) and R; leguminosarum (75) resulted in decreased infectibility and nodulation of their respective host plants. Genes necessary for EPS synthesis in R; trifolii have been cloned and the cloned DNA was able to restore both the ability to fix nitrogen and to synthesize EPS to a Fix- EPS- mutant strain (89). It was not reported at which level the polysaccharide synthesis was blocked nor the location of the genes involved. Recently, EPS- mutant strains of R_._ meliloti have been described in which the ability to induce nodules and infection thread formation was uncoupled (45). The nature of the mutation causing this phenotype was unknown and so was the location of the affected genes. In order to test if there is a direct or indirect relationship between the mutations and the 10 symbiotic phenotypes, more knowledge is necessary about the biosynthetic pathway of these polymers. One must establish at which step the txflysaccharide synthesis is affected in the different mutant strains as well as on the identification of the genes coding for these functions. Rhizobium mutant strains defective 1J1 organic acids uptake and metabolism are ineffective in nitrogen-fixing symbiosis 'with the host plant (47,49,83). Succinate and malate are abundant organic acids within legume root nodules (35,63,94) and succinate is transported and metabolized by free-living bacteria (46,50,71,83) and bacteroids (9,50,87,93) of different Rhizobium species. Organic acids support the highest rate of oxygen respiration by bacteroid suspensions (103) and succinate is a very effective substrate for supporting nitrogen fixation by both free- living Rhizobium species (10,109) and bacteroids (101). Utilization of tricarboxylic acid intermediates are related to symbiotic effectiveness (6). "In 'vitro" induction of bacteroid - like cell morphology by succinate in R; trifolii (104,105,106) and induction of sphere to rod morphogenesis in Arthrobacter crystallopoietes (65) by succinate have been described. Polysaccharide synthesis Microbial exopolysaccharide synthesis has recently been reviewed (95). Certain physiological conditions favor exopolysaccharide synthesis. For many microorganims, nutrient imbalance in the presence of large amounts of 11 utilizable carbohydrate leads to :anreased.jpolysaccharide production. Suboptimal incubation temperatures may also promote polysaccharide synthesis. In synthesizing exopolysaccharide, the water-soluble monosaccharide units have to be converted into high energy molecules and then must be passed through the lipophilic cell membrane and assembled into polysaccharide chains. These chains will fornu a capsule or slime at. the cell surface. Monosaccharides are activated through their incorporation into :nucleoside: diphosphate sugars and are transferred to lipid-linked derivatives, a process that leads to the assembly of oligosaccharides in a form soluble in organic solvent. Membrane-bound enzymes transfer a sugar l-phosphate to the isoprenoid lipid phoSphate acceptor. The key lipid carrier is a C55-polyisoprenoid called bactoprenol (67). Subsequently, further sugars are transferred from sugar nucleotides to form the oligosaccharide repeating units of the polymer. The lipid-linked oligosaccharides are polymerized in a bdock fashion to form the polysaccharide. Eventually, the polysaccharide is released from the bactoprenol and the lipid pyrophosphate converted back to the lipid phosphate. The biosynthesis of exopolysaccharide or capsular polysaccharide of different genera of bacteria including Klebsiella aerogenes (96), Xanthomonas campestris (59), Acetobacter xylinum (23), and .Aerobacter' aerogenes (102) have been studied and shown to involve lipid-bound oligosaccharide intermediates as precursors of the 12 polysaccharide. Synthesis of (l->2)/%-glucan from UDP-[14C] glucose by enzyme preparations of Rhizobium japonicum (34) and Rhizobium phaseoli (3) have been reported. Lipid-bound saccharides formed by incubation of UDP-glucose with a particulate enzyme of Rhizobium meliloti were also studied (98,99). Pyruvic acid ketal residues are present in many bacterial polysaccharides (95). However, there are very few reports (59) on the im yi_t_rg pyruvylation of the polysaccharides, and none for Rhizobium polysaccharides. 10. 11. 12. 13. 14. 15. 13 LIST OF REFERENCES Abe, M. A. Amemura, and S. Higashi. 1982. Plant and Soil 64:315-324. Abe, M., J. E. Sherwood, R. I. Hollingsworth, and F. B. Dazzo. 1984. J. Bacteriol. 160:517-520. Amemura, A. 1984. Agric. Biol. Chem. 48(7): 1809-1817. Amemura, A., T. Harada, M. Abe, and S. Higashi. 1983. Carbohydr. Res. 115:165-174. Ames, P., and K. Bergman, 1981. J. Bacteriol. 148:728- 729. Antoun, H., L. M. Bordeleau, and R. Sauvageau. 1984. Plant and Soil 77:29-38. Arias, A., A. Gardiol, and G. Martinez-Drets. 1982. J. Bacteriol. 151:1069-1072. Banfalvi, Z., Sakanyan, C. Koncz, A. Kiss, I. Dusha, and A. Kondorosi. 1981. Mol. Gen. Genet. 184:318-325. Bergersen, F. J. 1977. p. 519-555. Tm R. W. F. Hardy and W. S. Silver (ed.), A treatise on dinitrogen fixation, section III. John Wiley and Sons, New York. Bergersen, F. J., G. L. Turner, A. H. Gibson, and W. F. Dudman. 1976. Biochim. Biophys. Acta 444:164-174. Beynon, J. L., Beringer, J. E., and Johnston, A. W. B. 1980. J. Gen. Microbiol. 120:421-429. Bhuvaneswari, T. V., and. W. D. Bauer. 1978. Plant Physiol. 62:71-74. Bhuvaneswari, T. V., A. A. Bhagwat, and W. D. Bauer. 1981. Plant Physiol. 68:114-1149. Bhuvaneswari T. V. and B. Solheim. 1985. Physiol. Plant. 63, 25-34. Brewin, N. J., Beringer, J. E., Buchanan - Wollaston, A. V., Johnston, A. W. B., and Hirsch, P. R. 1980. J. Gen. Microbiol. 116:261-270. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 14 Brewin, N. J., J. L. Beynon, and A. w. B. Johnston. 1981. p. 65-77. In J. M. Lyons, R. C. Valentine, D. A. Phillips, D. W. Rains and R. C. Huftaker (ed.), Genetic Engineering of Symbiotic Nitrogen Fixation and Conservation of Fixed Nitrogen. Plenum Press. New York and London. Buikema, W. J., S. R. Long, S. E. Brown, R. C. van den Bos, C. Earl, and F. M. Ausubel. 1983. J. Mol. Appl. Genet. 2:249-260. Cadmus, M. C., K. A. Burton, and M. E. Slodki. 1982. Appl. Environ. Microbiol. 44:242-245. Callaham, D. A. and J. G. Torrey. 1981. Can. J. Bot. 59:1647-1664. Carlson, R. W. 1982. _I_n W. Broughton (ed), Nitrogen Fixation. II Rhizobium, Oxford University Press, UK. Christensen, A. H., and K. R. Schubert. 1983. J. Bacteriol. 156:592-599. Corbin, D., L. Barran, and G. Ditta. 1983. Proc. Natl. Acad. Sci. U.S.A. 80:3005-3007. Couso, R. O., L. Ielpi, R. C. Garcia, and M. A. Dankert. Eur. J. Biochem. 1982. 123:617-627. Dart, P. 1977. pp. 367-472. Im A treatise on nitrogen fixation. R. W. F. Hardy and W. S. Silver, eds. J. Wiley and Sons, New York. Dazzo, F. B. 1976. Appl. Environ. Microbiol. 32, 166- 177. Dazzo, F. B. 1984. p. 85-93. gm K. C. Marshall (ed.), Microbial Adhesion and Aggregation. Springer-Verlag, Berlin, Heidelberg, New York, Tokyo. Dazzo, F. B., C. A. Napoli, and D. H. Hubbell. 1976. Appl. Environ. Microbiol. 32:166-171. Dazzo, F. B., and G. L. Truchet. 1983. J. Membrane Biol. 73:1-16. Dazzo, F. B., G. Truchet, and E. Hrabak. 1984. lm C. Veeger and W. E. Newton, (eds.), Adv. in Nitrogen Fixation Res., Martinus Nijhoff/Dr W. Junk Pub., The Hague, p. 413. Dazzo, F. B., G. L. Truchet, J. E. Sherwood, E. M. Hrabak, M. Abe, and S. H. Pankratz. 1984. Appl. Environ. Microbiol. 48:1140-1150. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42. 43. 44. 45. 46. 15 Dazzo, F. B., G. L. Truchet, J. E. Sherwood, E. M. Hrabak, and A. E. Gardiol. 1982. Appl. Environ. Microbiol. 44:478-490. Dazzo, F. B., M. R. Urbano, and W. J. Brill. 1979. Curr. Microbiol. 2:15-20. Dazzo, F. B., W. E. Yanke, and W. J. Brill. 1978. Biochim. Biophys. Acta 539:276-286. Dedonder,R. A. and W. Z. Hassid, 1964. Biochim. Biophys. Acta 90:239-248. de Vries, G. E., P. In’t Veld, and J. w. Kijne. 1980. Plant Sci. Lett. 20:115-123. de Vries G. E., A. A. N. van Brussel, and A. Quispel. 1982. J. Bacteriol. 149:872-879. Djordjevic, M.A., P.R. Schofield, R. W. Ridge, N. A. Morrison, B. J. Bassam, J. Plazinski, J. M. Watson, and B. G. Rolfe. 1985. Plant Molec. Biol. 4:147-160. Downie, J. A., G. Hombrecher, Q-S. Ma, C. D. Knight, B. Wells, and A. W. B. Johnston. 1983. Mol Gen. Genet. 190:359-365. Downie, J. A., C. D. Knight, A. W. B. Johnston, and L. Rossen. 1985. Mol. Gen. Genet. 198: 255-262. Downie, J. A., Q.- S. Ma, C. D. Knight, G. Hombrecher, and A. W. B. Johnston. 1983. EMBO J. 2:947-952. Dudman, W. F., and M.Heidelberger. 1969.Science 164:954- 955. Dudman, W. 1984. pp. 397-404. lg C. Veegers & W. Newton (eds.), Advances in Nitrogen Fixation Research, Martinus Nijhoff/Junk, The Hague, Netherlands. Dudman, W., L. Franzen, J. Darvill, M. Mcneil, A. Darvill, and P. Albersheim. 1983. Carbohydr. Res. 117:141-156. Ervin, S. E. and. 13. H. Hubbell. 1985. Appl. Environ. Microbiol. 49,61-68. Finan, T. M., A. M. Hirsch, J. A. Leigh, e. Johansen, G. Kuldau, S. Deegan, G. C. Walker, and E. R. Signer. 1985. Cell 40:869-877. Finan, T. M., J. M. Wood, and D. C. Jordan. 1981. J. Bacteriol. 148:193-202. 47. 48. 49. 50. 51. 52. 53. 54. 55. 56. 57. 58. 59. 60. 61. 62. 63. 16 Finan, T. M., J. M. Wood, and D. C. Jordan. 1983. J. Bacteriol. 154:1403-1413. Forrai, T., E. Vincze, Z. Banfalvi, G. B. Kiss, G. S. Randhawa,and A. Kondorosi. 1983. J. Bacteriol. 153:635- 6430 Gardiol, A., A. Arias, C. Cervenansky, and G. Martinez- Drets. 1982. J. Bacteriol. 151:1621-1623. Glenn, A. R., P. S. Poole, and J. F. Hudman. 1980. J. Gen. Microbiol. 119:267-272. Goldstein, I. J., R. C. Hughes, M. Monsigny, T. Osawa, and N. Sharon. 1980. Nature 285:66. Higashi, S. 1967. J. Gen. Appl. Microbiol. 13:391-403. Hirsch, P. R., van Montagu, M., Johnston, A. W. B., Brewin, N. J. and Schell, J. 1980. J. Gen. Microbiol. 120:403-412. Hisamatsu, M., J. Abe, A. Amemura, and T. Harada. 1980. Agric. Biol. Chem. 44:1049-1055. Hollingsworth, R. I., M. Abe, and F. B. Dazzo. 1984. Carbohydr. Res. 133:C7-C11. Hollingsworth, R. I., M Abe, J. E. Sherwood, and F. B. Dazzo. 1984. J. Bacteriol. 160:510-516. Hooykaas, P. J., A. A. N. van Brussel, H. den Hulk-Ras, G. M. S. van Slogteren, and R. A. Schilperoort. 1981. Nature.(London) 291:357-353. Hrabak, E. M., M. R. Urbano, and F. B. Dazzo. 1981. J. Bacteriol. 148:697-711. Ielpi, L., R. O. Couso, and. M. A. Dankert. 1981. Biochem. Biophys. Res. Commun. 102:1400-1408. Jansson , P . , L. Kenne , B. Lindberg , I . Lonngren , V . Ruden, and S. Svensson. 1977. J. Amer. Chem. Soc. 99:3812-3815. Jansson, P., B. Lindberg, H. Ljunngren. 1979. Carbohydr. Res. 75:207-220. Johnston, A. W. B., Beynon, J. L., Buchanan - Wollaston, A. V., Setchell, S. M., Hirsch, P. R. and Beringer J. E. 1978. Nature 276:634-636. Kawai, S., and S. Mori. 1984. Soil Sci. Plant Nutr. 30:261-266. 64. 65. 66. 67. 68. 69. 70. 71. 72. 73. 74. 75. 76. 77. 78. 79. 80. 17 Kondorosi, E., Z. Banfalvi, and A. Kondorosi. 1984. Mol. Gen. Genet. 193:445-452. Krulwich, T. A., and J. C. Ensign. 1969. J. Bacteriol. 97:526-534. Law, I. J., Y. Yamamoto, A. J. Mort, and W. D. Bauer. 1982. Planta 154:100-109. Lennarz, 1%. J. and IL. G. Scher. 1972. Biochim. Biophys. Acta 265:417-441. Libbenga, K. R., and P. A. A. Harkes, 1973. Planta 114:17-21. Long, S. R., W. J. Buikema, and F. M. Ausubel. 1982. ”Nature (London) 298:485-488. Matthysse, A. G. 1981. J. Bacteriol. 145, 583-595. McAllister, (2. F., and J. E. Lepo. 1983. J. Bacteriol. 153:1155-1162. Meade, H. M., S. R. Long, G. B. Ruvkun, S. E. Brown, and F. M. Ausubel. 1982. J. Bacteriol. 149:114-122. Mellor, R. B., G. M. Gadd, P. Rowell, and W. D. P. Stewart. 1981. Biochem. Biophys. Res. Commun. 99:1348- 1353. Mort, A. J. and W. D. Bauer. 1980. Plant Physiol. 66: 158-163. Napoli, C. and P. Albersheim. 1980. J. Bacteriol. 141:1454-1456. Nordbring-Hertz, B., E. Friman, P. A. Johansson, and B. Mattiasson. 1981. pp. 43-50. 1m. Lectins-biology, biochemistry, clinical biochemistry, Vol. 1. T. C. Bog- Hansen, ed. W. de Gruyter, Berlin. Nuti, M. P., A. A. Lepidi,R. K. Prakash, R. A. Schilperoort, and F. C. Cannon, 1979. Nature. 282:533-535. Oda, Y., K-L Kassai, and S-L Ishii. 1981. J. Biochem. 89:285-296. Pistole, T. G. 1981. Ann. Rev. Microbiol. 35:85-112. Prakash, R. K., R. A. Schilperoort, and M. P. Nuti, 1981. J. Bacteriol. 145:1129-1136. 81. 82. 83. 84. 85. 86. 87. 88. 89. 90. 91. 92. 93. 94. 95. 96. 97. 98. 18 Robertsen, B., P. Aman, A. Darvill, M. Mcneil, and P. Albersheim. 1981. Plant Physiol. 67:389-400. Romeiro, R., A. Karr, and R. Goodman. 1981. Plant Physiol. 68:772-777. Ronson, C. W., P. Lyttleton, and J. G. Robertson. 1981. Proc. Natl. Acad. Sci. USA. 78:4284-4288. Rosenberg, C., P. Boistard, J. Denarie, and F. Casse- Delbart. 1981. Mol. Gen. Genet. 184:325-333. Rossen, L., A” IL. B. Johnston, .1. A. Downie. 1984 Nucl. Acids Res. 12:9497-9508. Ruvkun, G. B., and Ausubel, F. M. 1980. Proc. Nat. Acad. Sci. U.S.A. 77:191-195. Saroso, S., A. R. Glenn, and M. J. Dilworth. 1984. J. Gen Microbiol. 130:1809-1814. Schofield, P. R., M. A. Djordjevic, B. G. Rolfe, and J. Shine. 1983. M01. Gen. Genet. 192:459-465. Scott, K. F., J. E. Hughes, P. M. Gresshoff, J. E. Beringer, B. G. Rolfe, and J. Shine. 1982. J. Mol. Appl. Genet. 1:315-326. Sequiera, L. and T. C. Graham. 1977. Physiol. Plant Pathol. 11:43-54. Sherwood, J. E., J. M. Vasse, F. B. Dazzo, and G. L. Truchet. 1984. J. Bacteriol. 159:145-152. Solheim, B. J. and K. E. Fjelheim. 1984. Physiol. Plant. 62, 11-17. Stovall, I., and M. Cole. 1978. Plant Physiol. 61:787- 790. Stumpf, D. K., and R. H. Burris. 1979. Anal. Biochem. 95:311-315. Sutherland, I. W. 1979. Trends Biochem. Sci. 4:55- 59. Sutherland, I. W., and Norval, M. (1970). Biochemical Journal 120:567-576. Szeto, W. W., J. L. Zimmerman, V. Sundaresan, and F. M. Ausubel. 1984. Cell 36:1035-1043. Tolmasky, M. E., R. J. Staneloni, and L. F. Leloir, 1982. J. Biol. Chem. 257:6751-6757. 19 99. Tolmasky, M. E., R. J. Staneloni, R. A. Ugalde, and L. F. Leloir. 1980. Arch. Biochem. Biophys. 203:358-364. 100. Torok, I., II. Kondorosi, T. Stepkowski, J. Posfai, and A. Kondorosi. 1984. Nucl. Acids Res. 12:9509-9524. 101. Trinchant, (I. C., and.;1. Rigaud. 1979. L. Physiol. Veg. 17:547-556. 102. Troy, F. A., F. E. Frerman, and E. C. Heath. 1971. J. Biol. Chem. 246:118-133. 103. Tuzimura, K. and H. Meguro. 1960. J. Biochem. 47: 391-397. 104. Urban, J. E. 1979.. Appl. Environ. Microbiol. 38:1173- 1178. 105. Urban, J. E., and D. B. Bechtel. 1984. Appl. Environ. Microbiol. 47:1178-1181. 106. Urban, J. E., and F. B. Dazzo. 1982. Appl. Environ. Microbiol. 44:219-226. 107. van der Schaal, I., J. Kijne, C. Diaz, and F. van Iren. 1984. p. 531-538. Im Lectins, Vol III. T. C. Bog-Hansen and G. A. Spengler, eds., W. de Gruyter and Co., Berlin. 108. Vincent, J. M. 1980. p. 103-129. 1m W. E. Newton and W. H. Orme-Johnson (ed.), Nitrogen fixation, vol.2. Symbiotic associations and cyanobacteria. University Park Press, Baltimore. 109. Wilcockson, J., and D. Werner. 1979. Arch. Microbiol. 122:153-160. 110. Yao, It ‘Y. and LL. M. Vincent. 1976. Plant and Soil 45:1-16. 111. Zimmerman, J. L., W. W. Szeto, and F. M. Ausubel. 1983. J. Bacteriol. 156:1025-1034. 112. Zurkowski, W. 1980. Microbios 7:27-32. 113. Zurkowski, W. and Imukiewicz. 1979. Arch. Microbiol. 123:195-201. CHAPTER I SURFACE PROPERTIES WERE ALTERED IN E; TRIFOLII 0403 RIF WITH A TN5 INSERTION IN THE SYM PLASMID. ABSTRACT Rhizobium trifolii 0403 rif was mutagenized by transposon (Tn5) mutagenesis and the‘mutagenized pool was screened for mutant strains defective in root-nodule symbiosis with white clover. Eight hundred Kanr Rifr isolates were analyzed on the white clover host plant. Thirteen of these (1.6%) were symbiotically defective mutant strains. Three classes of mutant strains were obtained based on their symbiotic phenotype: unable to nodulate (Nod-), defective in nitrogen fixation (Fix-), and partially defective in nitrogen fixation (Fixi). Five strains were further studied: 5; trifolii 43 Nod-, g; trifolii 308 Nod", 3; trifolii 738 Fix', 3; trifolii 755 Fix- and R; trifolii 251 Fixi. Hybridization of total DNA restriction fragments with labeled Tn5 demonstrated the presence of a single TnS insertion in all the strains. Strains 251, 738 and 755 showed the same plasmid pattern as 20 21 the wild type strain, whereas strains 43 and 308 showed a deletion in the gym plasmid. Microscopic studies showed that strain 251 attached better and strains 43 and 308 attached less to clover root hairs than did the wild type strain. Strain 251 induced a similar number of clover root hair deformations and infection threads as did the wild type strain, whereas strains 308 and 43 were unable to deform or infect white clover root hairs. Quantitative agglutination assays with trifoliin A showed that strain 251 had a higher titer and strains 308 and 43 had lower titers than the wild type strain. Tn5 was located in the gym plasmid of strain 251 and presence of Mu DNA was not detected in this strain. Depolymerization of 251 CPS by a 9-1yase enzyme was slower than that of wild type CPS, and isolated oligosaccharide products from strain 251 CPS had less acetic and more pyruvic acid substitutions than wild type oligosaccharides. These results indicate that CPS non-carbohydrate substitutions are important in R; trifolii 0403 El: attachment to clover root hairs and lectin binding ability. 22 INTRODUCTI ON The Rhizobium trifolii - clover symbiosis involves a complex sequence of interactions (54) leading to the formation <1f 21 root nodule 'that fixes nitrogen. Several reports have shown that. many' of the bacterial symbiotic genes of fast-growing Rhizobium species are plasmid-borne and are located in a region of the gym plasmid which spans approximately 30 kilobase pairs (kb) (41). Genes encoding root hair curling (Egg), nodulation (mgd) (2, 6, 7, 10, 19, 20, 21, 28, 3o, 35, 39, 4o, 47, 49, 60) and nitrogen fixation (mi£_and film) genes (8, 11, 26, 41, 42, 45, 48, 53, 58) have been located in this megaplasmid. Studies of the R; trifolii - clover symbiosis (13) indicate that the specific ("Phase I") attachment process of the bacteria to root hairs is initiated by an interaction between the lectin, trifoliin A, associated with the root hair surface and carbohydrate receptors on the bacterial symbiont. Egg genes for root hair attachment of R; trifolii to clover root hairs are also encoded on the gym plasmid (59) and the genetic information required for trifoliin A- binding, "Phase I" attachment, marked root hair curling (shepherd crooks), root hair penetration and infection thread formation was expressed in pTi-cured g; tumefaciens containing the sym plasmid of _R_._ trifolii 5035 (32, F. B. 23 Dazzo, G. L. Truchet, and P. J. Hooykaas, Abstr. Annu. Meet. Am. Soc. Microbiol. 1983, K9, p. 178.). It has been hypothesized that the specific attachment of different Rhizobium species to host root hairs results from the interaction between the lectin on the host roots and Rhizobium CPS (15, 37, 44, 50, 52). Characterization of symbiotically defective mutant strains would be useful to identify the biochemical events leading to successful nodulation. In the present study, three different classes of mutant strains of R_. trifolii 0403 El: were obtained after Tn5 mutagenesis, based on their symbiotic phenotype with the white clover host plant. This wild type strain was selected because of the available information on its surface properties and its documented interaction with clover root hairs (16, 17). We used transposon (Tn5) mutagenesis as has been done for other Rhizobium species (4, 41), since Tn5 produces random insertion events at high frequency tagging the DNA lesion physically and genetically with a selectable resistance marker (3, 36). Once inserted, Tn5 has very low frequencies of transposition to a new site and of precise excision (41). These characteristics have made Tn5 the most widely used transposon in gene manipulations in different bacteria. We describe the isolation, symbiotic properties, and biochemical characterization of symbiotically defective mutant strains of R; trifolii 0403 rif. 24 A preliminary report of this work was presented at the Second International Symposium on the Molecular Genetics of the Bacteria-Plant Interaction, Cornell University, Ithaca, New York, June 1984. 25 MATERIALS AND METHODS Bacterial strains and phages. The following bacterial strains and bacteriophages were kindly provided to us. 3; trifolii 0403 and _E_; _cgl_i_ 1830 mg; m me_t (pJB4JI) were obtained from J. Beringer , Rothamsted Experimental Station, Harpenden, United Kingdom. 3; trifolii 0403 5;; was obtained from K. Nadler, Michigan State University, East Lansing. Ry trifolii 4S and bacteriophage 48 were from M. Abe and S. Higashi, Kagoshima University, Kagoshima, Japan. Bacteriophage A::Tn5 used for the Tn5 probe was from A. Christensen, Michigan. State University, East. Lansing. E; 99;; HM8305 F' E£Q+ £39 z 8305 :: Mu Egg 62/ (Egg — lgg) gig met tyr str was obtained from M. Howe, University of Wisconsin, Madison. Media and growth conditions. 3; trifolii strains were grown on either minimal Y (4), minimal BIII (12) or rich (38) media. Minimal Y contained succinate (YS) or glucose (YG) as C source. KNO3 was used as nitrogen source instead of sodium glutamate for growth. studies on. single carbon sources. Rifampicin (R) (20 lug/m1) or Kanamycin' (K) (80 /hg/ml) were added to YS or YG when indicated (YSRK, YGRK). R; trifolii strains were kept on BIII slants and R; trifolii Tn5 mutant strains were kept on YGRK slants or in stocks of YGRK-grown cells brought to 15% glycerol and kept 26 at -80°C in screw capped vials. E; coli 1830 was grown on rich TY medium (4) and E; coli HM8305 was grown on minimal Ozeki (43) or rich SB (33) media. Transposon (Tn5) mutagenesis. Escherichia coli 1830 mg; Egg mg; carrying the suicide plasmid pJB4JI (pPHlJI::Mu::Tn5) constructed In! Beringer et en” (5) was used. E; 99;; 1830 was grown on TY broth to stationary phase and R; EELIQLII 0403 Eli on BIII slants for 5 days. Mating was performed by mixing 1.2 x 1010 cells of donor with 5 x 109 cells of recipient. The cell suspensions were pipetted onto 0.451/un filters (Type HAWP; Millipore Corp., Bedford, Mass.). The filter was washed with water and incubated at 28°C on TY plates for 5h. Cells were resuspended in water, plated on selective YSRK medium, and incubated at 28°C. Transfer frequencies were calculated per number of recipient cells after the cross. Transconjugants were restreaked twice on YSRK to obtain single colonies. Screeninggfor symbiotically defective mutant strains. Surface-sterilized seeds (12) of Trifolium repens var. Ladino were germinated into humid air for 48h and transferred to the surface of (0.8%) agar slopes containing Fahraeus (N-free) medium (12). 5 day-old cultures of Rhizobium strains on BIII agar slants were suspended in Fahraeus medium and 5 x 106 cells were inoculated. per seedling: The tubes were incubated in a plant growth chamber with a 14h photoperiod at 22°C (26,900 lux) and 10h darkness at 20°C. Roots were scored for nodulation every week and 27 plants were assayed for nitrogen fixation by the acetylene reduction technique (12) after 6 weeks. Rhizobium DNA isolation. Cells grown iJI‘Y broth were harvested and lysed. Cells were suspended and incubated at 0°C for 30 min in 50 mM EDTA-Tris (pH 8.0) containing lysozyme (Sigma Chemical Co.) (1 mg/ml). The solution was brought to 0.1% SDS, 100 mM EDTA, and 1 mg/ml of proteinase K (Sigma Chemical Co.), and incubated at 50°C for 30 min. The lysate was extracted with the same volume of water- saturated phenol adjusted to pH 7.0 with l M Tris-HCl (pH 8.0) buffer. Aqueous phase was brought to 0.3 M sodium acetate (pH 7.0) and DNA was precipitated with 2 volumes of ethanol. The precipitate was wound out with a glass micropipette, dissolved :Ui 50 mM ‘Tris-HCl (pH 7.5) containing 1 mM EDTA, digested with RNAse (Type 1-A; Sigma Chemical Co.) (50 flg/ml) for 30 min at 37°C, extracted once with chloroform, and then ethanol-precipitated. The DNA was dissolved in 50 mM Tris-HCl (pH 7.0) containing 1 mM EDTA and stored at 4°C. The DNA concentration was determined by absorbance at 260 nm. Restriction endonucleases and DNA hybridization. Restriction endonucleases were purchased from Bethesda Reseaixfii Laboratories, Rockville, Md. and used as recommended by the manufacturer. DNA was digested for 8h (with Egg RI or Himg III and separated by electrophoresis in submerged horizontal 0.7% agarose gel in 40 mM Tris buffer (pH 7.5) containing 20 mM sodium acetate and 1 mM EDTA at 28 150 mA for 4h. The gel was treated with 0.4 M NaOH, 0.8 M NaCl for 30 min to denature the DNA, and neutralized with 0.5 M Tris-HCl (pH 7.5) containing 1.5 M NaCl. DNA transfer to nitrocellulose filters (Scheicher £1 Schuell) was performed for 24h with 10 x SSC (1 x SSC is 0.15 M NaCl plus 0.015 M sodium citrate) as described by Southern (51). Filters were rinsed with 5 x SSC, air dried and vacuum baked for 3h at 80°C. The DNA probe was obtained by labeling the DNA with 32P by the nick translation method (46). Hybridization and washing of the filters was performed as described by Engel and Dodgson (24). Autoradiography was performed for 8h at -80°C using XAR-S X-ray film (Eastman- Kodak, Rochester, N. Y., USA) with a Cronex Lightning-Plus intensifying screen (Du Pont, Co., Wilmington, Del.). Plasmid profile analysis and location of Tn5. Agarose gel electrophoresis was performed as described by Eckhardt (23) with slight modifications. Cultures were grown on Y broth to exponential phase (50 Klett units in a Klett- Summerson colorimeter with a No. 66 filter). Approximately 2 x 108 cells were pelleted for 5 min with an Eppendorf microfuge and the supernatant. was discarded. Cells ‘were washed with 1 ml of 0.1% sodium sarkosylate in 80 mM Tris- HCl (pH 8.0) containing 20 mM EDTA (A. Christensen, personal communication) and resuspended in 40 [“1 of the lysozyme (Sigma. Chemical Co.) mixture for' Gram negative bacteria (23). After 10 min incubation at room temperature, the lysates were added to the wells. 40 /u1 of the SDS mixture 29 were carefully layered on top of the bacteria-lysozyme mixture, and the slots were‘then filled with the overlay mixture and sealed with 0.7% agarose gel as described (23). 0.7% vertical agarose gels were run jjléi buffer containing 89 mM each Tris-boric acid (pH 8.2) at 8 mA for 1h and then' at 28 mA for 14h. Gels were stained for 30 min with ethidium bromide (l/ug/ml) in the same electrophoresis buffer, rinsed with distilled water for 10 min, and the DNA was visualized with a U.V. transilluminator and photographed. Agrobacterium tumefaciens C58 harboring three plasmids of known molecular weights: RP4 36 Md, Ti 130 Md, and pAtt58 214 Md was used for plasmid standard sizes. Plasmid DNA on Eckhardt agarose gels was partially depurinated by treatment with 0.25 M HCl for 15 min (55), and then was denatured, neutralized, transferred to nitrocellulose filters and hybridized to 32P A::Tn5 as described for restriction endonuclease digested DNA. Autoradiography was performed for 24h. Root hair interactions. (A) Attachment. Slide culture ggggy. Bacterial attachment to root hairs on primary seedling roots was examined by a previously described assay (12) in hydroponic, modified Fahraeus slide cultures without agar (25). 4 replicate seedlings of Louisiana Nolin var. of white clover were used. Rhizobium strains were grown for 5 days on BIII agar plates, centrifuged twice in 10 mM K phosphate buffer (pH 7.0) with 0.15 M NaCl (PBS) at 6,000 x 7 g and resuspended in Fahraeus medium. 2 x 10 bacteria were inoculated per seedling. After incubation in a plant growth "‘ 30 chamber for 12h, the slide cultures were disassambled. The roots were rinsed gently with a stream of Fahraeus medium while still on the slides, then covered with a glass cover slip and examined by phase contrast microscopy at 500 x along the optical median planes of the root. 15-20 root hairs (gg. ZOO/um in length) were counted per treatment. For orientation of attachment studies, 50 root hairs along the optical median planes were examined per strain 4h after inoculation of 4 x 107 cells per seedling in modified Fahraeus slide cultures. 1A attachments represent randomly oriented cells clumped to root hair tips and 1C attachments represent single ,cells polarly attached to root hairs. Beaker assgy. Approximately 10 root hairs (ca. 200 fem in length) were examined per strain after incubation of the seedlings for 2h in _a cell suspension containing 106 cells/m1 with slow shaking at 28°C. Seedlings were rinsed and observed as described before. (B) Deformation and infection studies. 10 replicate seedlings of Ty repens var. Louisiana Nolin were inoculated with 5 x 106 cells per seedling in Fahraeus slide cultures and incubated for 4 days. Seedlings were examined as described under (A) and the number of roots hairs with marked deformations (shepherd crooks) and of infected root hairs along the entire seedling root were counted. Protein determination. Protein concentration of lectin preparations was measured by the Bio-Rad protein assay. 5" 31 Bacterial agglutination assay. Harvesting and washing of cells, removal of nondispersible flocs, and quantitative bacterial agglutination assays were performed as described (14) with modifications by J. Sherwood (personal communication). Cells grown for 5d on BIII agar were removed from the plates with PBS and centrifuged at 12,000 x g for 10 min. The cell pellets were washed twice, resuspended in M buffer (pH 7.0) (0.2 mM K phosphate buffer with 0.15 M NaCl, 0.15 mM MnCl 0.5 mM CaCl and 0.5 mM M9804),' passed 2' 2' through glass wool in a Pasteur pipette to remove cell 8 cells/ml (30 Klett units aggregates, and adjusted to 3 x 10 in a Klett-Summerson colorimeter with a No. 66 filter). Cell suspensions (25 /k.l) were added to a two-fold dilution series of the lectin in M buffer (25 *1) in U-shaped polyvinylchloride microtiter plates (Dynatech Laboratories, Alexandria, VA.). Plates were sealed and incubated at room temperature for 14h. The maximum titer achieved. without autoagglutination was determined by hourly examination of the plates. Trifoliin A used was kindly provided by J. Sherwood and corresponded to the active fraction after the DEAE-Sephadex purification step of trifoliin A (J. Sherwood, ‘personal communication). Purification of phage Mu and preparation of Mu DNA Erobe. Mu lysate preparation and phage purification were Exerformed essentially as described (27). E; 99;; HM8305 (Mu Lysogen) was grown at 32°C in 4 flasks with 250 ml of SBM medium (34) to 4 x 108 cells/ml. Phage growth was heat- 32 induced by addition of an equal volume of 58°C medium and growth was continued at 42°C for 45 min. The culture was quickly centrifuged by allowing the rotor to accelerate to 4000 x g and then braking. The cell pellets were resuspended in one-twentieth vol. of 37°C SBM and slowly shaken at 37°C for 4h. The Mu lysate was centrifuged to remove debris. 22 ll PFU/ml) were layered per ml of lysate (1x10 ultracentrifuge tube over a cesium chloride step gradient consisting of 7 m1 of 1.4 g/ml CsCl, 6 m1 of 1.5 g/ml CsCl, and 3 ml of 1.7 g/ml CsCl, and centrifuged at 20,000 rpm (52,200 x g) in an SW27 rotor for 2h at 4°C. The resulting phage bands located at approximately 1.47 g/ml were pooled, approximately adjusted by addition of CsCl to a density of 1.50 g/ml and centrifuged to equilibrium at 22,000 rpm (62,700 xg) in an SW27 rotor at 4°C for 48h. Presence of the phage in each step was confirmed by negative staining with neutralized phosphotungstic acid followed by transmission electron microscopy using a Philips 300 TEM. Purified phage was dialyzed against 0.1 M Tris-HCl (pH 7.9) containing 0.3 M NaCl and 2.5 mM MgC12. DNA was isolated from phage particles by phenol extraction (56) and labeled lMith 32P by the nick translation method (46). Hybridization, ‘washing of the filters and autoradiography were performed as described above. . Nodulation and host range studies. Seedlings of Medicago sativa var. Veronal, Trifolium subterraneum var. (Clare, Trifolium pratense, Trifolium fragiferum and 33 Trifolium repens vars. Ladino, Louisiana Nolin and White Dutch were inoculated with 2 x 107 cells and incubated under the same conditions described for screening of mutant strains. In addition, surface-sterilized seeds of Phaseolus vulgaris var. Black Turtle Soup and line 21-58 (from F. Bliss, Univ. of Wisconsin) and of Pisum sativum lines 8221 and 9888 F were germinated into humid air for 3 days, planted on the surface of Smucker (A. Smucker, personal communication) or Fahraeus agar medium, respectively, and 8 cells per seedling. All plants were inoculated with 10 incubated under the same conditions described for screening of mutant strains and scored for nodulation every week. Isolation of polysaccharide depolymerase PD-I. The l;- lyase enzyme was isolated from the clarified phage lysate of R_. trifolii 4S by precipitation with (NH4)2SO4 (70% of saturation) and purified by DEAF-cellulose (DE52 Whatman, Inc., Clifton, N. J.) column chromatography (29). Oligosaccharide isolation. BIII plates were inoculated with a suspension of 107 cells previously grown for 5 days on BIII agar medium. Cells grown for 5 days were harvested by centrifugation in PBS (pH 7.2). CPS was extracted from pelleted cells with PBS (pH 7.2) containing 0.5 M NaCl, then precipitated with 2 vol. of cold ethanol, centrifuged, redissolved in water, dialyzed against water, and 1yophilized. CPS was depolymerized with polysaccharide depolymerase PD-I and the oligosaccharide products were purified by gel filtration chromatography through Bio-Gel 34 P10 (Bio-Rad Laboratories, Richmond, Ca.) in 20 mM Tris-HCl (pH 7.2), concentrated by flash evaporation, desalted by gel filtration through Bio-Gel P2 and lyophilized (31). Kinetic study of CPS depolymerization rates. Depolymerization rates of different CPS samples were measured with constant amount of PD—I enzyme under saturating conditions of CPS substrate. CPS solutions in 25 mM 'Tris-HCl .buffer' (pH 7.2) containing 2 mud CaCl2 were adjusted to a final concentration of 0.15 mg/ml after assaying for total carbohydrate using the phenol-sulfuric acid assay (22). One unit of PD-l depolymerase was defined as the amount of enzyme producing an increase of absorbance at 235 nm of 0.01 per minute. PD-I enzyme preparation (41kg protein, 0.5 Units) was added to 1-ml samples of each CPS solution. The depolymerization activity was measured by an increase in absorbance at 235 nm (due to the unsaturated sugar formed by PD-I cleavage of CPS at glucuronic acid) (31) over a 4h period using a Gilford Response spectrophotometer interfaced. ‘with 51 computer' kinetic program. The computer program calculated a quadratic equation of the data by the least-square fitting method. Initial rates were calculated from the value of the derivative of the quadratic expression at time zero. 1H-NMR analysis of oligosaccharides. Oligosaccharide (OS) samples were deuterium-exchanged and the spectra were obtained on a Bruker WM-250 instrument at room temperature 35 (31). The chemical shifts were measured relative to an external tetramethyl-silane standard. Oligosaccharide glycosyl composition. Glycosyl composition of the oligosaccharide repeating unit was analyzed by GLC as described (31). 36 RESULTS Symbiotic characterization of mutant strains. Nod and Fixgphenotype. Rhizobium trifolii 0403 rif was mutagenized by transposon (Tn5) mutagenesis and the mutagenized pool was screened for mutant strains defective in root nodule symbiosis with Ladino white clover. Eight hundred Kanr Rifr isolates were analyzed on the clover host plant. Thirteen of these (1.6 %) were symbiotically defective strains, falling into three different classes based on their symbiotic phenotype with the host plant: unable to nodulate (Nod-), defective in nitrogen fixation (Fix-), and partially defective in nitrogen fixation (Fixi). Nod- and Fix- phenotypes were assigned after five independent experiments with 5-10 replicates each. Five mutant strains: 3; trifolii 308 Nod", 13; trifolii 43 Nod-, & trifolii 251 Fixi, g; trifolii 738 Fix- and R_._ trifolii 755 Fix- were further characterized. Strains 308, 43, 251, and 755 grew like wild type in minimal medium containing different single carbon sources, whereas strain 738 had slower growth rate than that of the wild type strain. The average number of nodules per plant (5 replicates) induced by the different strains 40 days after inoculation of Ladino white clover was 6 for 0403 £23, 6.8 for 251, 19 for 755, and 12 for 738. Fix- phenotype was confirmed by acetylene reduction (13). Nodules with R; 37 trifolii 738 or E; trifolii 755 failed to reduce acetylene. E; trifolii 251 had 50% (8.1 nmol/h/plant) of the wild type activity (17.4 nmol/h/plant) on Trifolium repens var. Ladino. Physical analysis Ci" mutant strains for presence of Egg. Autoradiograms of 32P-labeled A::Tn5 DNA hybridized to Egg RI-digested or Eggg III-digested total DNA showed one and three bands respectively (Figure 1). This result demostrated the presence of a single Tn5 insertion since Tn5 lacks Egg RI restriction sites but has two Elmg III restriction sites. Plasmid patterns of mutant strains. R_. trifolii 0403 contains at least three plasmids with sizes of 235, 205, and 190 Mdal. The gym plasmid of this strain was identified as the smallest plasmid (190 Mdal) separated on Eckhardt agarose gels (R. Taylor, M. S. Thesis, University of Florida, Gainesville, 1981). The 205 Mdal plasmid appears as a partially resolved doublet and may contain two co- migrating plasmids. Strains 251, 738 and 755 showed the same plasmid profile as the wild type strain when analyzed by Eckhardt agarose gel electrophoresis, whereas strains 308 and 43 showed a deletion in the gym plasmid (Figure 2). Attachment” deformation. and infection studies. Quantitative attachment studies in Fahraeus slide cultures and in beaker assays showed that strain 251 attached better and strains 308 and 43 attached less to clover root hairs than did the wild type strain (Table 1). The percentage of 38 FIGURE 1. Autoradiogram of 32P-labeled A::Tn5 hybridized to digested total DNA. E_cg RI‘ digestion. (A) Lanes: a, R_._ trifolii 738; b, _R_._ trifolii 43: c, _R_. trifolii 0403 E_i_f: d, EL, trifolii 308; e, E_._ trifolii 755; f, R_. trifolii 251. M III digestion. (B) Lanes: a, E; trifolii 0403 gi_f_; b, R. trifolii 251. ' 'L- 39 FIGURE 2. Plasmid patterns by Eckhardt agarose gel electrophoresis. (A) Lanes: a, E; trifolii 308; b, R_ trifolii 0403 gig; c, E; trifolii 43. (B) a, E; trifolii 738; b, E; gglgglg; 755; c, E; trifolii 251; d, E; trifolii 0403 gig. 40 TABLE 1. Attachment of wild type E; trifolii 0403 rif and mutant strains to white clover root hairs. Attached cells per root hair (i : s.d) Strain Slide culture assaya Beaker assayb 0403 g}; (wild type) 36 i 3 25 i 2 251 65 i 2 41 i 3 308 15 i 3 12 i 2 43 16 i 3 14 i 3 755 29 i 5 31 i 4 738 NDC 24 + 2 a 15-20 root hairs ( a. 200 m in length) of Louisiana Nolin var. of white clover seedlings were examined per strain 12h after inoculation of 2 x 107 cells per seedling in Fahraeus slide cultures. b . . . . . . 10 root hairs were examined per strain after incubation of the seedlings for 2h with a cell suspension containing 106 cells/ml in beakers. C Not done. 41 root hairs with the combined 1A+1C pattern of attachment (clump at tip and polar attachments along the sides of the same root hair) was lower for Nod- strains 308 and 43 than for the wild type strain (Table 2). The wild type E; trifolii 0403 El: and mutant strains 251, 738 and 755 induced. similar number' of .marked root hair deformations (shepherd crooks) (Table 3). The number of infection threads in Fahraeus slide cultures induced by strains 251 and 738 were similar to those induced Iby the lflild 'type strain, whereas strain 755 induced fewer infection threads (Table 3). Neither shepherd crooks nor infected root hairs were detected on seedlings incubated with E; trifolii 308 or 43 (Table 3). anntitative agglutination with trifoliin A. Quantitative agglutination assays (M? the different mutant strains showed that trifoliin A had a significantly higher specific agglutinating activity using strain 251 and lower specific agglutinating activities with strains 308 and 43 than with wild type strain (Table 4). Nodulation and host ragge studies. The nodulation patterns (Table 5) of E; trifolii 251 were similar to those of the wild type 0403 g}: strain on E; pratense and Ladino var. of T_. repens. However, it induced more nodules on White Dutch and Louisiana Nolin var. of L repens, var. (Zlare of I; subterraneum, and I; fragiferum. Neither strain Inodulated. Veronal alfalfa or' pea lines 8221 and 9888F. Ihawever, both strains ineffectively nodulated bean varieties 42 TABLE 2. Orientation of attachment of E; trifolii 0403 rif . . a and mutant strains to clover root hairs. Percentage of root hairs with Strain bacterial attachments No attachment 1A only 1C only 1A + 1C 0403 rif 6 28 22 44 251 13 21 21 45 308 33 18 38 11 43 ' 21 7 54 18 755 7 25 23 45 738 4 30 21 45 a Approximately 50 root hairs (g. 200 lam in length) of Louisiana Nolin var. of white clover seedlings were examined per strain 4h after inoculation of 4 x 107 cells per seedling in Fahraeus slide cultures. 1A attachments represent randomly oriented cells clumped to root hair tips, and 1C attachments represent single cells polarly attached to root hairs. The lA+lC pattern has both on the same root hair. 43 TABEE 3. Shepherd crooks and infection threads induced by E; trifolii 0403 rif and mutant strains. # shepherd crooks # infected root hairs Strain per seedling per seedling (R i s.d.)a (R : s.d.)b 0403 ml: (wild type) 17 i 2 12 i 1 251 13 i 1 10 i 1 308 0 0 43 0 0 755 11 i 4 6 i 3 738 11 i 4 12 i 3 a,b Values are average of four replicate seedlings of Trifolium repens var. Louisiana Nolin inoculated with 5 )( 106 cells per seedling in Fahraeus slide cultures and incubated for 4 days. 44 TABLE 4. Trifoliin A - agglutinating activity of E; trifolii 0403 rif and mutant strains.a Specific agglutinating activityb Strain (agglutinating units/mg trifoliin A protein) 0403 El: 5,333 251 42,664 308 2,667 43 2,667 755 5,333 738 5,333 Bacterial agglutination was performed in microtiter plates. Trifoliin A (60 [Jog/ml) isolated from seeds of Trifolium repens var. Louisiana Nolin was used. Cells were grown for 5 days on BIII agar' medium and prepared as described in the text. b Cell suspensions (25,44) were added to a two-fold dilution series of the lectin (25 .lpl)° Specific agglutinating activity is the number of agglutinating units per mg of trifoliin A (18). 45 TABLE 5. Nodulation patterns of _R_. trifolii 0403 rif (wt) and. E; trifolii. 251 (ms) on 'Trifolium. repens, Trifolium subterraneum, Trifolium fragiferum, and Trifolium gratense. Average number of nodules per plant Clover host 10d 15d 20d 40d wt ms wt ms wt ms wt ms 2g repens vars. White Dutch 1.0 1.4 2.5 4.0 3.0 5.4 3.0 11.0 Louisiana Nolin 1.0 1.4 1.0 3.0 3.5 4.0 4.5 8.4 Ladino 2.0 1.8 4.0 3.8 6.0 5.6 6.0 6.8 I; subterraneum 0.0 1.3 0.0 1.5 0.5 2.0 1.5 2.3 2g fragiferum 0.0 2.6 1.5 3.6 2.0 5.8 3.0 8.0 E; pratense 2.0 1.0 4.0 3.0 9.0 6.0 9.0 12.0 a Average number of nodules per plant (5 replicates) induced by strains 0403 rif and 251 were recorded at indicated times after inoculation. 46 (Black Turtle Soup or line 21-58 at 40 days). Nodulation of E; vulgaris by E; trifolii 0403 has been previously reported (57). Location of Tn5 insertion. in R. trifolii 251. The presence .of a 'Tn5 insertion in the gym plasmid of E; trifolii 251 was demonstrated by hybridization of 32P- labeled A::Tn5 DNA to plasmid DNA which had been separated by agarose gel electrophoresis (Figure 3). Physical analysis of R. trifolii 251 for the presence of Mu DNA. DNA was isolated from phage Mu which had been purified by CsCl equilibrium gradient centrifugation (Figure 4) and labeled with 32P by nick translation. Autoradiograms of 32P - labeled Mu DNA hybridized to Egg RI-digested total DNA from strain 251 did not show any hybridization bands, indicating that Mu sequences are not present in E; trifolii 251 total DNA. CPS depolymerization rates. Kinetic studies showed that the E; trifolii 0403 El: CPS substrate was depolymerized by PD-I at a faster rate than was CPS of E; trifolii 251 (Figure 5, Table 6). A one-way analysis of variance using time F .distribution indicated that these values were significantly different at a probability of chance (P<0.005) (Table 6). Depyruvylated CPS was not cleaved by the enzyme, and therefore, pyruvate substitutions were essential for PD- I lyase activity. 1 H-NMR analysis of CPS for non-carbohydrate components. CPS was converted into structurally analyzable 47 FIGURE 3. Plasmid profile and location of Tn5 insertion in E; trifolii 251. (A) Plasmid Eckhardt gel. (8) Autoradiogram 32 of P-labeled A::Tn5 hybridized to plasmid Southern blot. 48 FIGURE 4. Phage Mu purified by CsCl equilibrium gradient centrifugation. 49 ABSORBANCE at 235nm . o 10 20 ’ 3'0 4‘0 MINUTES FIGURE 5. Depolymerization rates of CPS from R_. trifolii 0403 EL: 4--), E; trifolii 251 Gun") and depyruvylated CPS (--) with enzyme PD-I followed by increase of absorbance at 235 nm under saturating conditions of substrate and identical protein concentration. 50 TABLE 6. Depolymerization rates of CPS from E; trifolii 0403 rif and E; trifolii 251. No. of Initial ratea Strain b replicates Mean : s.d. 0403 rif 3 44 i 4 251 3 27 i 2 Expressed as 104 x Abs 235 nm/min. b Standard deviation of the mean. 51 oligosaccharides using ea bacteriophage depolymerase system and examined by 250-MHZ Fourier-transform lH-NMR. The spectra showed the presence of acetate esters by resonances (relative to external .Me4Si) between 16 1J88 and 2.15. Signals assigned to ketal-linked pyruvate groups appeared between (6 1.23 and 1.43. Two other groups of resonance, one between 6.1.03 and 1.17 and signals between 3 2.36 and 2.58 were assigned to ether-linked 3-hydroxybutanoic acid. These were quantitated by comparing the integrals of the signals for the respective groups to the area of the signal of the C-4 proton of the unsaturated terminal sugar identified as 4-deoxy-L-ggggg-hex-4-enopyranosyluronic acid (DEPUA) resulting‘ from the enzymatic cleavage of the CPS at glucuronic acid (31). These NMR measurements indicated that the oligosaccharide fragments from CPS of strain 251 had less acetic and. more pyruvic acid substitutions per 08 repeating unit than wild type OS (Table 7). Oligosaccharide glycosyl composition. The glycosyl composition of 251 OS was the same as that of wild type OS (G1u+G1uUA: Gal: DEPUA) (6: l: l) (31). 52 TABLE 7. Non-carbohydrate composition of oligosaccharides obtained by PD-I depolymerization of CPS isolated from R_._ trifolii 0403 rif and g; trifolii 251 a Molar proportion of non—carbohydrate components per oligosaccharide repeating unit (: 0.05).b Strain Pyruvic Acetic 3-hydroxybutanoic acid acid acid 0403 rif 1.74 1.70 0.41 251 2.04 1.48 0.31 a CPS was isolated from cells grown for 5 days on BIII agar plates, depolymerized into its oligosaccharide repeating unit (OS) and purified through Bio-Gel P10 and P2. 08 were exchanged with deuterium oxide and the 1H-NMR spectra were recorded. 1) Values were obtained by comparing the integrals for the various groups with that of the single C-4 proton of the unsaturated terminal sugar in each oligosaccharide. 53 DISCUSSION Three classes of mutant strains were isolated after Tn5 mutagenesis of E; grifolii 0403 El; based on their symbiotic phenotype on white clover: unable to nodulate (Nod-), defective 1J1 nitrogen fixation (Fix-) anui partially defective in nitrogen fixation (Fixi). Five strains were further studied: Eg_ trifolii 308 (Nod-), Em. trifolii '43 (Nod'), 3; trifolii 755 (Fix'), 3; trifolii 738 (Fix‘), and 3; trifolii 251 (Fixi). In comparison with the wild type strain, mutant strains 308 (Nod’) and 43 (Nod') had a deletion in the gym plasmid (Figure 2), attached less to clover root hairs (Table 1) with lower percentage of the combined 1A+lC attachment to the same root hair (Table 2), were unable to deform or infect root hairs (Table 3), and had lower specific agglutinating activities with trifoliin 11 (Table 4). This similarity of results obtained with two strains having a deletion in the gym plasmid suggested that functions coded in this plasmid may have a positive effect on the ability of Rhizobium trifolii 0403 rif to attach to clover root hairs and to interact with the clover lectin. Complementation of these mutants with a cloned (nod-nif) region should restore wild type phenotype if the deletion of the plasmid is responsible of the phenotype. Analysis of CPS 54 from ‘strain 308 revealed the presence of a novel N- containing non-carbohydrate substitution which was significantly higher in the mutant than in wild type CPS (R. Hollingsworth and F. Dazzo, personal communication). Strains 755 (Fix-) and 738 (Fix-) with normal plasmid patterns (Figure 2), had the same attachment (Table l) and lectin binding (Table 4) ability as the wild type strain. Both strains induced. more nodules “than the wild type strain although 755 induced less and 738 the same number of root hair infections as wild type (Table 3). It is well known that many (but not all) Rhizobium ineffective strains induce numerous nodules on the host plants. Thus, the Tn5 insertion of strain 755 in the gym plasmid (data not shown) affected root hair infection, nodulation, and nitrogen fixation. Strain 251 with a single Tn5 insertion (Figure l) and a normal plasmid pattern when compared to the wild type (Figure 2) attached in higher numbers to clover root hairs (Table 1), had similar orientation of attachment (Table 2), induced similar number of shepherd crooks and root hair infections (Table 3), and was agglutinated better by trifoliin A than was the wild type strain (Table 4). The degree of distribution of lectin receptors on E; trifolii cells has been correlated with the degree and orientation of attachment (50). The fact that this mutant strain which binds the lectin better is also able to attach to clover 'root hairs better is consistent with the proposed importance of lectin binding in attachment of Rhizobium to clover root 55 hairs (50). "Phase I" loose attachment is followed by "Phase II" firm anchoring of the bacterial cell to the root hair surface (17). During Phase II, extracellular fibrillar materials are characteristically found associated with the adherent bacteria. It is not known whether these fibrils consist of bundles of cellulose, fimbriae, or some other fibrillar polymers made by the attached bacteria (17). Strain 251 also showed more Phase II (firm) attachments and fibrillar polymers on the root hair surface than did the wild type strain (K. Smith, H. Yang and F. Dazzo, personal communication). Although strain 251 induced more nodules on several clover hosts (Table 5), this property was not correlated with the induction of more root hair infections in Louisiana Nolin var. of E; repens (Table 3) and may be due to the fact that the strain is partially defective in nitrogen fixation. Interestingly, the white clover variety's response to nodulation (Table 5) by strain 251 matches the number of root hair infections induced by the wild type strain (White Dutch>Louisiana Nolin>Ladino) (D. Gerhold and F. Dazzo, personal communication). This strain would be ideal for studying the effect of improved attachment on the degree (M? the strain's success 111 interstrain competition (with a Fix+ strain) for nodule sites on the root. Strain 251 was selected for further genetic and biochemical characterization because it has a single Tn5 insertion and showed a significant increase in attachment and lectin binding ability as compared to the wild type 56 strain. Tn5 was located in the gym plasmid (Figure 3), and presence of Mu DNA was not detected in strain 251 DNA. The latter was investigated since Mu phage was originally present in the suicide plasmid (pPHlJI::Mu::Tn5) used as a vector for Tn5 mutagenesis. CPS isolated from strain 251 was analyzed since the specific attachment of different Rhizobium species to host root hairs has been hypothesized to result from the interaction between the lectin on the host root and Rhizobium CPS (15, 37, 44, 50, 52). In addition, lectin-binding ability of E; trifolii 0403 CPS has been reported to be age-dependent (50) and changes of CPS (1, 50) and EPS non-carbohydrate substitutions with culture age have been reported (9). CPS from strain 251 was different from wild type CPS as shown by kinetic study of depolymerization rates using PD-I ablyase (Table 6, Figure 5) and quantitative lH-NMR determination of non-carbohydrate substitutions (Table 7). The rate of depolymerization of 251 CPS by depolymerase PD-I was significantly less than that of wild type CPS and depyruvylated CPS of wild type was not depolymerized by the enzyme, indicating that pyruvate substitutions are essential for PD-I depolymerase activity. Previous reports showed that deacetylated CPS was depolymerized faster than wild type CPS by fl-lyase PD-II, a different polysaccharide depolymerase enzyme (31). This indicates that changes in the non-carbohydrate substitutions of the CPS may change the conformation of the substrate thereby altering the kinetics of the depolymerization 57 reaction. CPS from strain 251 had more pyruvate and less acetate non-carbohydrate substitutions than wild type CPS. Tn5 mutants in the nodulation region I (Hac) of E; trifolii 843 also had alterations in the levels of acetate and/or pyruvate in their CPS (Chapter II). However, strain 251 is Hac+, and therefore the Tn5 insertion in this strain is not in the Hac region of the gym plasmid. Structural analysis of E; trifolii 0403 oligosaccharides obtained by depolymerization of CPS with our polysaccharide depolymerase system showed that acetate is ester-linked and pyruvate is ketal-linked to the same nonreducing terminal galactose residue UR. I. Hollingsworth, inanuscript 1J1 preparation). The decrease in acetate levels and corresponding increase in pyruvate levels may indicate that the transferase enzymes adding' these substitutions ix) the glycosyl residues are competing for sites for substitution. Therefore, the alteration in the level of one of these substitutions may affect the. level of the other. Interestingly, CPS pyruvyl transferase activity was higher for E; trifolii 251 than for the wild type strain (Chapter III). These results also indicate that functions related to CPS synthesis may be encoded in the gym plasmid. It would be important to further define the location of the mutated gene relative to the nodulation and nitrogen fixation regions already identified 111 the gym plasmid of other strains of E; trifolii (49, M. A. Djordjevic, Mol. Gen. Genet., in press). This ‘would provi 58 provide important information as to the organization and function of the bacterial symbiotic genes. Results obtained indicate that CPS non-carbohydrate substitutions are important in E; grifolii 0403 rif attachment to clover root hairs and lectin binding ability. 10. 59 LIST OF REFERENCES Abe, M., J. E. Sherwood, R. I. Hollingsworth, and F. B. Dazzo. 1984. Stimulation of clover root hair infection by lectin binding oligosaccharides from the capsular and extracellular polysaccharides of Rhizobium trifolii. J. Bacteriol. 160:517-520. Banfalvi, Z., Sakanyan, C. Koncz, A. Kiss, I. Dusha, and A. Kondorosi. 1981. Location of nodulation and nitrogen fixation genes on a high molecular weight plasmid of Rhizobium meliloti. Mol. Gen. Genet. 184:318-325. Berg, D. E., R. Jorgensen, and J. Davies. 1978. Transposable kanamycin-neomycin resistance determinants, p. 13-15. In D. Schlessinger. (ed.), Microbiology.l978. American Society for Microbiology, Washington, D.C. Beringer, J. E. 1974. R factor transfer in Rhizobium leguminosarum. J. Gen. Microbiol. 84:188-198. Beringer, J. E., J. L. Beynon, A. V. Buchanan - Wollaston, and A. W. B. Johnston. 1978. Transfer of the drug - resistance transposon Tn5 1x) Rhizobium. Nature. London. 276:633-634. Beynon, J. L., Beringer, J. E., and Johnston, A. W. B. 1980. Plasmids and host range in Rhizobium leguminosarum and Rhizobium phaseoli. J. Gen. Microbiol. 120:421-429. Brewin, N. J., Beringer, J. E., Buchanan - Wollaston, A. V., Johnston, A. W. B., and Hirsch, P. R. 1980. Transfer of symbiotic genes with bacteriocinogenic plasmids in Rhizobium leguminosarum. J. Gen. Microbial. 116:261-270. Buikema, W. J., S. R. Long, S. E. Brown, R. C. van den Bos, C. Earl, and F. M. Ausubel. 1983. Physical and 'genetic characterization of Rhizobium meliloti symbiotic mutants J. Mol. Appl. Genet. 2:249-260. Cadmus, M. C., K. A. Burton, and M. E. Slodki. 1982. Growth - related substituent changes in the exopolysaccharides (n5 fast-growing rhizobia. Appl. Environ. Microbiol. 44:242-245. Christensen, A. H., and K. R. Schubert. 1983. Identification of a Rhizobium trifol ii plasmid coding for nitrogen fixation and nodulation genes and its 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 6O interaction with pJBSJI, a Rhizobium leguminosarum plasmid. 156:592-599. Corbin, D., L. Barran, and G. Ditta. 1983. Organization and expression of Rhizobium meliloti nitrogen fixation genes. Proc. Natl. Acad. Sci. U.S.A. 80:3005-3007. Dazzo, F. B. 1982. Leguminous root nodules, p. 431-446. In R. G. Burns and J. H. Slater (ed.). Experimental Microbial Ecology. Blackwell Sientific Publications, Oxford. Dazzo, F. B. 1984. Bacterial adhesion to plant root surfaces, p. 85-93. lg K. C. Marshall (ed.), Microbial Adhesion enui Aggregation. Springer-Verlag, Berlin, Heidelberg, New York, Tokyo. Dazzo, F. B., and W. J. Brill. 1979. Bacterial polysaccharide which binds Rhizobium trifolii to clover root hairs. J. Bacteriol. 137:1362-1373. Dazzo, F. B., and D. H. Hubbell. 1975. Cross-reactive antigens and lectin as determinants of symbiotic specificity 1J1 the Rhizobium-Clover association. Appl. Microbiol. 30:1017-1033. Dazzo, F. B., and G. L. Truchet. 1983. Interactions of lectins and their saccharide receptors in the Rhizobium - legume symbiosis. J. Membrane Biol. 73:1-16. Dazzo, F. B., G. L. Truchet, J. E. Sherwood, E. M. Hrabak, M. Abe, and S. H. Pankratz. 1984. Specific phases of root hair attachment in the Rhizobium trifolii - clover symbiosis. Appl. Environ. Microbiol. 48:1140- 1150. Dazzo, F. B., W. E. Yanke, and W. J. Brill. 1978. Trifoliin: a Rhizobium recognition protein from white clover. Biochim. Biophys. Acta 539:276-286. Downie, J. A., G. Hombrecher, Q-S. Ma, C. D. Knight, B. Wells, and A. W. B. Johnston. 1983. Cloned nodulation genes of Rhizobium leguminosarum determine host-range specificity. Mol Gen. Genet. 190:359-365. Downie, J. A., C. D. Knight, A. W. B. Johnston, and L. Rossen. 1985. Identification of genes and gene products involved in the nodulation of peas by Rhizobium leguminosarum. Mol. Gen. Genet. 198: 255-262. Downie, J. A., Q.- S. Ma, C. D. Knight, G. Hombrecher, and A. W. B. Johnston. 1983. Cloning of the symbiotic region of Rhizobium leguminosarum: the nodulation genes 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 61 are between the nitrogenase genes and nifA-like gene. EMBO J. 2:947-952. Dubois, M., K. A. Gilles, J. K. Hamilton, P. A. Rebers, and F. Smith. 1956. Colorimetric method for determination of sugars and related substances. Anal. Chem. 28:350-356. Eckhardt, T. 1978. A rapid method for the identification of plasmid deoxyribonucleic acid in bacteria. Plasmid 1:584-588. Engel, J. D., and Dodgson, J.i3. 1981. Histone genes are clustered but not tandemly repeated in the chicken genome. Proc. Nat. Acad. Sci. U.S.A. 78:2856-2860. Fahraeus, G. 1957. The infection of clover root hairs by nodule bacteria studied by a single glass slide technique. J. Gen. Microbiol. 16:374-381. Forrai, T., E. Vincze, Z. Banfalvi, G.B . Kiss, G. S. Randhawa,and A. Kondorosi. 1983. Localization of symbiotic mutations in Rhizobium meliloti. J. Bacteriol. 153:635-643. Grundy, F. J., and M. H. Howe. 1984. Involvement of the invertible (3 Segment 1J1 bacteriophage bhi tail fiber biosynthesis. Virology. 134:296-317. Higashi, S. 1967. Transfer of clover infectivity of Rhizobium trifolii to Rhizobium phaseoli as mediated by an episomic factor. J. Gen. Appl. Microbiol. 13:391-403. Higashi, S., and M. Abe. 1978. Phage- induced depolymerase for exopolysaccharide of Rhizobiaceae. J. Gen. Appl. Microbiol. 24:143-153. Hirsch, P. R., van Montagu, M., Johnston, A. W. 13., Brewin, N. J. and Schell, J. 1980. Physical identification of bacteriocinogenic, nodulation and other plasmids in strains of Rhizobium leguminosarum. J. Gen. Microbiol. 120:403-412. Hollingsworth, R. I., M Abe, J. E. Sherwood, and F. 3 . Dazzo. 1984. Bacteriophage-induced acid heteropoly saccharide lyases that convert the acidic heteropolysaccharides of Rhizobium trifolii into oligosaccharide units. J} Bacteriol. 160:510-516. Hooykaas, P. J., A. A. N. van Brussel, H. den Hulk-Ras, G. M. S. van Slogteren, and R. A. Schilperoort. 1981. gym plasmid of Rhizobium trifolii expressed in different rhizobial species and Agrobacterium tumefaciens. Nature.(London) 291:357-353. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42. 43. 44. 62 Howe, M. M. (1973) Prophage deletion mapping of bacteriophage Mu-l. Virology 54:93-101. Howe, M. M., J. W. Shumm, and A. L. Taylor. 1979. The S and. U 50 cells per field). The percentage of cells binding the lectin was calculated by comparing the number of immunofluorescent cells to the total number of cells in the same field recognized by phase-contrast microscopy. Methods for isolaticui of’ polysaccharide: depolymerase PD-I from a phage lysate of E; trifolii 4S, and oligosaccharide isolation from CPS of plate-grown cultures have been described in Chapter I. Kinetic study of CPS depolymerization rates. Depolymerization rates of CPS samples were measured under saturating conditions of substrate and identical protein concentration. CPS solutions 111 25 mM Tiis-HCl buffer (pH 7.2), 2 mM CaCl were adjusted to a final concentration of 2 0.15 mg/ml after assaying for total carbohydrate by the phenol-sulfuric acid. :method (3). Five samples with decreasing substrate concentrations were used to establish enzyme saturation with each CPS. PD-I (4 M protein, 0.5 Units) was added to l-ml samples of each CPS solution and the depolymerization activity (measured by EH1 increase in absorbance at 235 nm) was followed for 4-5h in a Gilford Response spectrophotometer interfaced with a: computer kinetic program. Three to seven independent experiments were run with each substrate. The computer program calculated a quadratic equation of the data by the least-square fitting method. Initial rates were calculated from the value of the derivative of the quadratic expression at time zero. One 74 unit of PD-I depolymerase was defined as the amount of enzyme producing an increase of absorbance at 235 nm of 0.01 per minute. lH-NMR analysis of oligosaccharides. Oligosaccharide (OS) samples were deuterium-exchanged and the spectra were recorded on a Bruker WM-250 instrument at room temperature (8). The chemical shifts were measured relative to an external tetramethyl-silane standard. Oligosaccharide composition. Glycosyl composition of the oligosaccharide repeating' unit. was analyzed by gas- liquid chromatography-mass spectrometry as described in detail elsewhere (8). Uronic acid residues of OS were methylated and then reduced with sodium borodeuteride. Deuterium - labeled, carboxyl reduced OS were hydrolyzed with trifluoroacetic acid. Monosaccharides were then reduced with sodium borohydride, and the resulting alditols were peracetylated. The ratio of glucose to glucuronic acid was determined by selective ion monitoring of the ions m/z 217 and 219 derived from the unlabeled and deuterium - labeled glucitol hexa-acetates, respectively. 75 RESULTS Symbiotic ;phenotype. Typical root hair deformation responses on Trifolium repens var. Louisiana Nolin induced by wild type, pSym-, and mutant strains in regions I, II, and III in slide cultures are shown in Figure 2. Trifoliin A - binding activity. E; trifolii 851 (nod (I)::Tn5) and E; trifolii 845 (pSym-) had significantly lower abilities to bind the clover lectin trifoliin A than the wild type strain in both mg glgg and gg-planta assays (Table 2 and Figure 3). When the cells in the 1g gggg assay with seedlings were compared by FITC-epifluorescence and phase contrast. microscopy, three subpopulations could. be distinguished, based on the distribution of trifoliin A bound to the bacterial surface. Wild-type cells either bound trifoliin A uniformly, at one cell pole, or were unreactive. The relative proportions of these subpopulations for the wild-type strain, compared to E; trifolii 851 and to E; trifolii 845 in the root environment are presented in Table 2. Interestingly, neither mutant bound trifoliin.1\ at one cell pole, unlike the wild-type strain. The low level of lectin binding ability of E; trifolii 851 in the gm glanta assay was transient with culture age as it was for the wild type strain (Figure 3). Binding of trifoliin A to the pSym- mutant strain 845 was negative (Figure 3). FIGURE 2. Typical root hair response induced by wild type and mutant strains. Strains are E; trifolii 843 (wild type), 3; trifolii 851 (nod I::Tn5), E; trifolii 262 (nod II::Tn5), g; trifolii 297 (nod III::Tn5), and E; trifolii 845 (pSym-). 77 TABLE 2. lg situ binding of trifoliin A to E; trifolii in clover slide cultures. % of cells with bound trifoliin Aa (R i s. d.) Strain Uniform Polar Unreactive ANU 843 Hac+Nod+ 11.1:2.3 24.7:2.6 64.2:3.8 ANU 851 Hac’Nod' l.8:1.5 0.0:0.0 97.2:2.6 ANU 845 Hac-Nod- 0.0:0.0 0.3:0.7 99.7:0.8 a 2x107 bacteria were inoculated per seedling of l; repens var. Louisiana Nolin. Cells were removed from the root environment after incubation ftm"7 days and heat-fixed to slides. Trifoliin A bound to cells was detected by indirect immunofluorescence. 78 < E 5 2 25+ E g. a 20‘ E o E . n 15 d 10* o 3 5‘ a! 3 4 s s 7 a 9 10 AGE or PLATE CULTURE, DAYS FIGURE 3. Eg-planta binding of trifoliin A to wild type and symbiotically defective mutants of I; trifolii 843. Cells were grown on BIII agar plates for 3 to 10 days and heat fixed to slides. Binding to trifoliin A was examined by indirect immunofluorescence and recorded as the percentage Of cells showing fluorescence. ( O ) R_. trifolii 843 (wild tyPe), ( a ) R_. trifolii 851 (M (I)::Tn5), and ( as ) R. trifolii 845 (pSym-) . 79 Kinetic study cfif CPS depolymerization. PD-I depolymerase enzyme was saturated in) to one fourth (0.04mg/ml) of the CPS concentration used in the assays and the increase in absorbance of oligosaccharide products at 235 nm was proportional to the amount of enzyme employed. Completion of depolymerization occurred in approximately 100 minutes for all the CPS substrates. Results of a typical kinetic experiment are shown in Figure 4, illustrating that PD-I depolymerized CPS of some mutant strains faster (e.g. 851) and other mutant strains slower (e.g. 252) than CPS of the wild type strain. Initial depolymerization rates of CPS from different mutant strains are shown in Table 3. To compare rates quantitatively, data were subjected to a one- way analysis of variance using the F distribution (Table 3). The depolymerization rate of CPS from strain 252 was significantly lower (P<0.05), whereas that for CPS of strains 851 (P<0.025), 274 (P<0.10) and 845 (P<0.10) were significantly higher than those of wild type CPS (Table 3). 1H-NMR analysis of oligosaccharides. The lH-NMR spectrum of isolated OS from plate-grown E; trifolii 843 (Figure 5) indicated the presence of acetate esters(6l.88- 2.15), ketal-linked pyruvate(61.23-l.43) and ether-linked 3- hydroxybutanoic acid(8l.03-1.17 and 2.36-2.58). The levels of pyruvic, acetic, and 3-hydroxybutanoic acid substitutions per' CPS oligosaccharide repeating unit. are presented in Table 4. Strain 252 had more pyruvic and acetic, strain 246 had more pyruvic, strain 851 had less pyruvic, and strain 80 0.70 ABSORBANCE (235 nm) 0.05 . , . . 0 6 12 1O 24 30 MINUTES FIGURE 4. Kinetic study of the relative rates of d(iipolymerization of CPS from R_. trifolii 843 and mutant Strains in region I, by PD-I enzyme under identical Conditions. Strains are I: trifolii 843, _R_._ trifolii 252 (352g A::Tn5) and E; trifolii 851 (nod D::Tn5). 81 TABLE 3. Depolymerization rates of CPS from E; trifolii 843 (wild type) and. Tn5 mutant strains in three nodulation regions using enzyme PD-I. b No. of Initial rate Significant Straina ! replicates mean : s.d.C differenced 843 (wild type) 7 48 i 5 - 277 (I) 3 52 i 3 No 252 (I) 2 34 i 3 Yes(P<0.05) 246 (I) 2‘ 55 i 5 No 274 (I) 3 56 i 5 Yes(P<0.10) 851 (I) 5 59 i 6 Yes(P<0.025) 262 (II) 4 50 i 1 No 297 (III) 4 52 i 2 No 845 (pSym") 3 56 t 2 Yes(P<0.10) a Nodulation region with Tn5 insertion is shown in parentheses. b Expressed as 104x Abs 235 nm/min. C Standard deviation of the mean. d All data were subjected to a one-way analysis of variance using the F distribution. The indicated probabilities of chance were considered indicative of a significant difference with E; trifolii 843 (wild type). 82 s€_ 5‘: FIGURE 5. A 1H-NMR spectrum of the oligosaccharides produced by depolymerization of CPS from a S-day-old culture of E; trifolii 843, using PD-I enzyme. Peaks represent (A) H-4 of 4-deoxy-L-ggggg hex-4-enopyranosyluronic acid, which results from fl-elimination of glucuronic acid catalyzed by PD-I, (B) .methylene and (B methyl of 3-hydroxy butanoic acid, 1) (C) acetate, and (D) pyruvate protons. 83 TABLE 4. Non-carbohydrate composition of oligosaccharides obtained by PD-I depolymerization of CPS from E; trifolii 843 and mutant strains in three nodulation regions. Molar proportion of non-carbohydrate components per oligosaccharide repeating unita (: 0.05) E; trifolii strain Pyruvic Acetic 3-hydroxybutanoic acid acid acid ANU 843 (wild type) 1.58 0.93 0.54 ANU 277 (I) 1.55 0.81 0.57 ANU 252 (I) 2.79 1.41 0.62 ANU 246 (I) 1.99 1.09 0.57 ANU 274 (I) 1.54 0.77 0.54 ANU 851 (I) 1.12 0.89 0.49 ANU 262 (II) 1.57 0.92 0.54 ANU 297 (III) 1.57 0.96 0.54 ANU 845 (pSym-) 1.38 0.74 0.60 a Values were obtained from 1H-NMR spectroscopy by comparing the integrals for the various groups with that of the single C-4 proton of the unsaturated terminal sugar; in each oligosaccharide. 84 845 had less pyruvic and acetic substitutions, respectively than CPS of the wild type strain. The level of substitutions for OS of other strains was similar to that of the wild type strain. Glycogyl composition analysis. Gas chromatographic and combined GLC-MS analysis of the wild type oligosaccharide repeating unit indicated that it contained glucose, glucuronic acid, and galactose. The 1H-NMR spectra of the oligomers displayed a downfield multiplet with the same chemical shift at. 6 5.8-5.9 as the signal identified as 4- deoxy-L-ggggg-hex-4-enopyranosyluronic acid :Ui E; tmifolii 0403 oligosaccharides produced during lg-lyase cleavage of the CPS at glucuronic acid (8). The molar ratio of glycosyl components is (Glu:GluUA:Gal). (5:2:1). 85 DISCUSSION Our results showed that CPS of some Tn5 mutant strains of E; trifolii 843 1J1 nodulation region 12 are altered in levels of pyruvate and acetate non-carbohydrate substitutions relative to wild type CPS. However, several strains had the same levels of these substitutions as the wild type strain, indicatimg that the entire procedure of CPS isolation, depolymerization into OS and lH-NMR spectroscopy to quantitate the non-carbohydrate substitutions is very reliable and that these levels are unaffected by the presence of Tn5 genes themselves. 851 CPS had less pyruvate, 246 CPS had more pyruvate and 252 CPS had more pyruvate and more acetate, respectively, than wild type CPS. This indicated that different locations of Tn5 insertions within region I have a different effect on the levels of pyruvate and acetate substitutions. This is consistent with the finding of four nodulation genes within this region for' '3; trifolii, 3;. meliloti and R. leguminosarum (4, 6, 9, 10, 12, and T. T. Egelhoff et al., in press). In addition to having different levels of non- carbohydrate substitutions, 851 CPS, 252 CPS, and 845 CPS also displayed different depolymerization rates with PD-I from that of wild type CPS. This is consistent with previous iuork showing that the activities of CPS depolymerase enzymes 86 are sensitive to pyruvate (PD-I enzyme) (Chapter I) and acetate (PD-II enzyme) NH non-carbohydrate substitutions. Therefore, differences between CPS from wild type and mutants can 1x3 established based on alterations in depolymerization rates with enzyme PD-I and/or the levels of non-carbohydrate substitutions measured by 1H-NMR. However, the fact that these quantities are the same for CPS of some mutants does not automatically mean that the polysaccharides are the same. 1H-NMR data are only a quantitative measurement.cnf CPS non-carbohydrate substitutions. l3C-NMR would provide further information about CPS structure. CPS with the same depolymerization rates by enzyme PD-I may have differences to which the enzyme is not sensitive. Lectin binding assays showed that mutant strains 851 and 845 with altered CPS were also significantly less reactive with trifoliin A as compared with the wild type strain. The levels CHE CPS non-carbohydrate substitutions in some of the mutants may be affected because of different levels of the corresponding transferase enzymes responsible of their incorporation into the CPS, and this study will be addresSed in Chapter III. Since the levels of pyruvate and acetate in CPS from mutant strains with Tn5 insertions in nodulation region I are different from wild type levels we conclude that there must be some expression of essential ggg genes in E; trifolii 843 grown in defined BIII medium in absence of the host plant. However, our studies with root exudate (Chapter 87 III) suggest that some ggg genes expression is affected by clover root exudate. We conclude that Tn5 insertions in certain go_d genes within region I cause changes in the CPS of E; trifolii 843 as demonstrated by quantitative determination of pyruvate, acetate and 3-hydroxybutanoate substitutions by lH-NMR spectroscopy and kinetic study of CPS depolymerization rates and that functions related to polysaccharide synthesis may be encoded in the gym plasmid. 1. 88 LIST OF REFERENCES Dazzo, F. B. 1982. Leguminous root nodules, p. 431-446. _Ig R. G. Burns and J. H. Slater (ed.). Experimental Microbial Ecology. Blackwell Scientific Publications, Oxford. Dazzo, F. B., G. L. Truchet, J. E. Sherwood, E. M. Hrabak, and .A. E. Gardiol. 1982. Alteration of the trifoliin A- binding capsule of Rhizobium trifolii 0403 by enzymes released from clover roots. Appl. Environ. Microbiol. 44:478-490. Dubois, M., K. A. Gilles, J. K. Hamilton, P. A. Rebers, and F. Smith. 1956. Colorimetric method for determination of sugars and related substances. Anal. Chem. 28:350-356. Djordjevic, M.A., P.R. Schofield, R. W. Ridge, N. A. Morrison, B. J. Bassam, J. Plazinski, J. M. Watson, and B. G. Rolfe. 1985. Rhizobium nodulation genes involved in root hair curling (Hac) are functionally conserved. Plant Molec. Biol. 4:147-160. Downie, J. A., G. Hombrecher, Q. S. Ma, C. D. Knight, B. Wells, and A. W. B. Johnston. 1983. Cloned nodulation genes of Rhizobium leguminosarum determine host-range specificity. Mol. Gen. Genet. 190:359-365. Downie, J. A., C. D. Knight, A. W. B. Jehnston, and L. Rossen. 1985. Identification of genes and gene products involved in the nodulation of peas by Rhizobium leguminosarum. Mol. Gen. Genet. 198:255-262. Higashi, S., and M. Abe. 1978. Phage-induced depolymerase for exopolysaccharide of Rhizobiaceae. J. Gen. Appl. Microbiol. 24:143-153. Hollingsworth, R. I., M. Abe, J. E. Sherwood, and F. B. Dazzo. 1984. Bacteriophage-induced acid heteropolysaccharide lyases that convert the acidic heteropolysaccharides of Rhizobium trifolii into oligosaccharide units. J. Bacteriol. 160:510-516. Kondorosi, E., Z. Banfalvi, and A. Kondorosi. 1984. Physical and genetic analysis of a symbiotic region of Rhizolium meliloti: identification of nodulation genes. Mol. Gen. Genet. 193:445-452. 10. 11. 12. .89 Rossen, L., A. W. B. Johnston, J. A. Downie. 1984 DNA sequence of the Rhizobium leguminosarum nodulation genes go_d A, B and C required for root hair curling. Nucl. Acids Res. 12:9497-9508. Schofield, P. R., R. W. Ridge, B. G. Rolfe, J. Shine, and J. M. Watson. 1984. Host-specific nodulation is encoded on a l4kb fragment in Rhizobium trifolii. Plant Mol. Biol. 3:3-11. Torok, I., E. Kondorosi, T. Stepkowski, J. Posfai, and A. Kondorosi. 1984. Nucleotide sequence of Rhizobium meliloti nodulation genes. Nucl. Acids Res. 12:9509- 9524. CHAPTER III RHIZOBIUM TRIFOLII CAPSULAR POLYSACCHARIDE BIOSYNTHESIS: PYRUVYLATION OF LIPID-BOUND SACCHARIDES AS AFFECTED BY TN5 MUTATIONS IN THE SYMBIOTIC PLASMID. ABSTRACT An lg vitro assay to measure the enzymatic incorporation of pyruvate into E; trifolii CPS was developed using EDTA-treated cells, UDP-sugar donors, and [l-14 C]PEP as pyruvate donor. Pyruvylation occurred at the lipid-bound saccharide intermediate stage. CPS pyruvyl transferase activity (CPT) was measured by incorporation of radioactivity into glycoconjugates soluble in organic solvent using two wild type R_. trifolii strains- (843 and 0403 Eli): and corresponding symbiotically defective mutant strains having alterations in levels of pyruvate in their CPS. CPS pyruvyl transferase activity was affected by Tn5 mutations in the gym plasmid, and clover root exudate increased the level of CPT in E; trifolii strains having Tn5 insertions in certain nodulation (Hac) genes. 90 91 INTRODUCTION The biosynthesis of exopolysaccharide or capsular polysaccharide of different genera of bacteria including Klebsiella aerogenes (l7), Xanthomonas campestris (7), Acetobacter xylinum (3), Aerobacter aerogenes (l4), and Neisseria meningitidis (10), have been studied and shown to involve lipid-bound oligosaccharide intermediates as precursors of the polysaccharide. The key lipid carrier is a C55 polyisoprenol (undecaprenol) called bactoprenol which is linked to the growing carbohydrate chain through an acid- labile phosphodiester bond (8). Synthesis of (l->2) fz-glucan from UDP-[14C] glucose by enzyme preparations of Rhizobium jgponicum (5) and Rhizobium phaseoli (2) have ‘been reported. Lipid-bound. saccharides formed by incubation of UDP-glucose with a particulate enzyme of Rhizobium meliloti were also studied (11,12). Pyruvic acid ketal residues are present in many bacterial polysaccharides (16). However, there are very few reports (7) on the _i_g vitro pyruvylation of the polysaccharides, and none for Rhizobium polysaccharides. Rhizobium trifolii Tn5 mutant strains in different locations of nodulation region I had different levels of pyruvate substitutions in their CPS as compared to the wild type strain 1L trifolii 843 when grown on defined culture 92 medium (Chapter II). E; trifolii 251, a mutant strain of E; trifolii 0403 El: containing a single Tn5 insertion in the gym plasmid, also had higher levels of pyruvate in its CPS than did the wild type strain (Chapter I). Non-carbohydrate substitutions of E; trifolii 0403 CPS have also been related to its transient lectin-binding ability (1,15, Chapter I, and Chapter II). In view of the importance of these non-carbohydrate substitutions, we decided to study the enzymatic incorporation of pyruvate into CPS of E; trifolii. A second important objective of this study was to determine if the enzymatic lesion of the mutant strains with different levels of pyruvate in the CPS is at the level of this CPS pyruvyl transferase (CPT) activity. A third objective was to examine the effect of clover root exudate on CPT activity in the wild type and mutant strains. Results of this study showed that pyruvylation of R_. trifolii CPS occurs by a phosphoenol pyruvate donor at the saccharide-lipid bound intermediate stage. CPS pyruvyl transferase activity is affected by Tn5 mutations in the gym plasmid and clover root exudate. Portions CHE this work were .presented at the 6th International Symposium on Nitrogen Fixation, Oregon State University, Corvallis, OR, Aug. 4-9, 1985. 93 MATERIALS AND METHODS Bacterial strains. 3; trifolii ANU 843 (wild type) and corresponding mutant strains (ANU 252, 246, anui 851) were obtained from Dr. B. Rolfe, Australian National University, Canberra, Australia, (Chapter II). E; trifolii 252 has a Tn5 insertion in the coding region of ggg A gene of nodulation region I and has a Hac-Nod- phenotype. E; trifolii 246 with a Tn5 insertion before the coding region of ggg A gene has a Hac+Nod+ phenotype. E; trifolii 851 has a Tn5 insertion in the coding region of ggg D gene and has a Hac-Nod- phenotype. {R_. trifolii 0403 Ell (Wild type) was obtained from K. Nadler, Michigan State University, East Lansing. E; trifolii 251, a mutant strain of E; trifolii 0403 Eli: has a Tn5 insertion in the gym plasmid and has a Hac+Fixi phenotype. CPS pyruvylation assay (A) Enzyme preparation. Cells were grown in BIII broth for 36h to stationary phase (120 Klett units in a Klett-Summerson colorimeter with a No. 66 filter), centrifuged at 12,000 its; for 30 min, and washed with BIII medium and with 10 mM EDTA-Tris buffer, (pH 8.0). The pellet was resuspended with one volume of buffer and this cell suspension was frozen at -80°C and thawed at room temperature eight times (hereafter called EDTA-treated cells) (11). For root exudate treatments, cells grown to 94 stationary phase were then centrifuged aseptically and resuspended in BIII broth without C source (neither mannitol nor glutamate) containing 70 rml of 14d white clover root exudate (0.1 mg/ml protein) and shaken at 30°C for 24 hours. Cell pellets were then washed and EDTA-treated as described. Proteins were measured by the method of Lowry (9) with a standard of bovine serum albumin. (B) Enzyme assay. Enzymatic incorporation of pyruvate into CPS was measured with an lg vitro assay modified from the method of Tolmasky et al. (11). The standard mixture (total volume of 0.05 ml) contained: 70 mM Tris-HCl buffer, (pH 8.2), 8 mM MgCl 40 mM 2-mercaptoethanol, EDTA-treated 2, cells (0.4-l mg protein), 0.3 mM UDP-Glucose, 0.15 mM UDP- Galactose, 0.15 mM UDP-Glucuronic acid, and 0.57 mM [1-14C] phosphoenol pyruvic acid (15 mCi/mmol) as cyclohexylammonium salt (PEP) (Amersham). The reaction was performed in screw- capped plastic tubes kept at 15°C in a water bath for 20 min and stopped by adding 0.5 nu. 30 minutes. The optimal temperature was 15°C and the Optimal pH was 8.2. The concentration of UDP-sugar donors used was saturating and the addition of cold PEP to the incubation. mixture decreased the incorporation. of radioactivity in: background levels. Controls consisting of enzyme preparations inactivated by boiling gave negative results. Incorporation of radioactivity into LBO from labeled UDP-sugars. There was incorporation of radioactivity into lipid-bound saccharides from each of the three labeled UDP- sugar donors Used separately in the incubation mixture, UDP— l4 [ acid (Table 1). C]Glucose, UDP-[14C]Galactose, or UDP-[14C]Glucuronic TAE 58( SW UD‘. U0 U0 t1 ir WI 99 TABLE 1 . Incorporation of radioactivity into lipid-bound saccharides in cells of E; trifolii 843 from labeled UDP- sugars.a UDP-[14C]sugar Activityb 14 UDP-[ C]Glucose 1499 UDP-[14C1Galactose 2953 UDP-[14C]Glucuronic acid 12746 a Assay was performed as described in the text with EDTA- treated cells grown in defined culture medium and the indicated UDP-[l4 b Clsugar donor. Expressed as the radioactivity (cpm) incorporated into glycoconjugate product soluble in chloroform: methanol: water (1: 2: 0.3) in 20 min per 0.6 mg of cellular protein. Values are average of three replicates. 100 Mild acid hydrolysis. Labeled product obtained in an 14 incubation mixture containing [1 C]PEP scaled-up 30-fold was treated by mild acid to cleave the acid-labile perphosphate bond linking the lipid carrier to the carbohydrate. 'This resulted 1J1 transfer cu? all the radioactivity to the aqueous phase when the reaction system was partitioned with chloroform: methanol: water (3: 2: 1) (Table 2). Analysis of the radioactive moiety soluble in the aqueousghase after mild acid hg/drolysis. Gel permeation chromatography of the cleaved radioactive products soluble in aqueous phase (24,000 cpm) through Bio-Gel P6 gave an elution profile shown in Figure 1. Four radioactive peaks were detected. Two peaks (I and II) eluted between the octasaccharide and lactose standards; two other peaks with significantly more radioactivity (III and IV) eluted after lactose. Acid hydrolysates of fractions I, III, and IV were examined by paper chromatography and shown to contain labeled pyruvate. ' Pyruvylation of lipid-bound saccharides. CPT activities in EDTA-treated cells of E; trifolii 843 and mutant strains are shown in Table 3. All the data were subjected to a one- way analysis of variance using the F distribution. For cells grown in defined culture medium only, strain 246 showed higher and strain 851 lower CPT activity than wild type strain. The difference between 246 and 851 was statistically significant at a probability of chance P<0.10. For cells 101 TABLE 2. Mild acid hydrolysis of labeled product. Radioactivity % (cpm) recovery Labeled producta 36,000 100 Aqueous phaseb 37,000 102 a LBO extract. b Radioactivity in aqueous phase after mild acid hydrolysis and partition in chloroform: methanol: water (3: 2: l). 102 /. .x/’ Q (-~- ) p” '1. / l l ' ‘\ 1500. ; ~ . i /. N / ABSORBANCE .: 490nm (a.--) RADIOACTIVITY, cm 3:?- ,. C C d 509 . /. I' 1 ‘1 f l u /./° ;' 1‘ \. I... .lkw.\/. I", ‘1‘ \o 1) .#_-‘£:;;::__;hnnumuhns‘V - - -- 1 O 40 50 60 70 FRACNWDN IflMHBEH Figure 1. Gel filtration of the radioactive moiety soluble in aqueous phase using Bio-Gel P6. Fractions were collected and aliquots monitored for radioactivity. Unlabeled saccharides were determined by the phenol-sulfuric acid method. 103 TABLE 3. Pyruvylation of lipid-bound saccharides in cells of E; trifolii 843 (wild type) and Tn5 mutant strainsa. Pyruvic acid Activityd Strainb per oligosaccharide (x i s.d.) repeating unitC -REe +REf 843 (wild type) 1.58 303130 295:15 246 (I) 1.99 326155 447:50 252 (I) 2.79 305116 374:39 851 (I) 1.12 250:10 286110 Enzyme assay was performed with EDTA-treated cells, UDP- sugar donors and [l - l4C]PEP. b Nodulation region with Tn5 insertion is shown in parentheses. c Values were obtained from lH-NMR spectrosc0py by comparing the integrals for the various groups with that of the single C-4 proton of the unsaturated terminal sugar in each oligosaccharide (Chapter II). Pyruvylation enzyme activity is expressed as the radioactivity (cpm) incorporated into glycoconjugates soluble in chloroform: methanol: water (1: 2: 0.3) in 20 min per 0.4 mg of protein. Values are average of three replicates. e Cells grown to stationary phase in defined culture medium. f Cells grown to stationary phase were then treated for an additional 24h with clover root exudate (RE). 104 treated with root exudate, strains 246 and 252 had higher CPT activities than did the wild type strain. Treatment of cells of strains 246 and 252 with root exudate significantly increased their CPT activities at P<0.05 whereas CPT activities of strains 843 and 851 were unaffected by root exudate. Strain 251 had higher levels of CPT activity than did the corresponding wild type strain 11; trifolii 0403 rif (Table 4). 105 TABLE 4. Pyruvylation of lipid-bound saccharides in cells of 3; trifolii 0403 rif (wild type) and 3; trifolii 251.a Pyruvic acid Strainb per oligosaccharide Activityd repeating unitC 0403 rif (wild type) 1.74 518 251 2.04 685 Enzyme assay was performed with EDTA-treated cells grown to early stationary phase in defined culture medium, UDP- sugar donors and [1-14C]PEP. Strain 251 has a single Tn5 insertion in the gym plasmid of strain 0403 51;. c Values were obtained from lH-NMR spectroscopy by comparing the integrals for the various groups with that of the single C-4 proton of the unsaturated terminal sugar in each oligosaccharide (Chapter I)“ Pyruvylation enzyme activity is expressed as the radioactivity (cpm) incorporated into glycoconjugate product soluble in chloroform: methanol: water (1: 2: 0.3) in 20 min per mg of protein. Values are average of three replicates. 106 DISCUSSION We have evidence of consistent differences in the levels of pyruvate substitutions in CPS of certain Tn5 mutant strains of 31 trifolii 843 with Tn5 insertions in the nodulation region I (Hac) (Chapter II). 31 trifolii 251, a mutant strain of B1 trifolii 0403 11; containing a single Tn5 insertion in the gym plasmid, also had higher levels of pyruvate in its CPS than the CPS of the corresponding wild type strain. In 31 trifolii 0403 the level of non- carbohydrate substitutions in the CPS have an important effect on lectin binding ability (1,15). In this work we have deve10ped an _i_n 17.1.22 assay to measure the enzymatic incorporation of pyruvate into CPS. EDTA-treated cells, UDP-sugar donors and [1-14C] phosphoenol pyruvate as the source of pyruvate, were used in the _i_n giggg assay. There was no incorporation of label into glycoconjugate product soluble in chloroform: methanol: water (1: 2: 0.3) when 4-fold concentration of cold PEP was added to the incubation mixture containing labeled PEP. This indicates that PEP radioactivity is incorporated into this product. By using PEP labeled at C1, we eliminated possible incorporation of label from acetate :metabolism. Radioactivity was incorporated into the glycoconjugate [14 product when either UDP- C]Glucose, UDP-[14C1Galactose, or 107 UDP-[14C]Glucuronic acid were used in the incubation mixtures without PEP (Table 1) indicating that the saccharide moiety of the CPS is synthesized from these three UDP-sugar donors. The difference in level of incorporation using each different labeled UDP-sugar precursor probably reflects a different specific activity of the internal pool sizes with UDP-glucuronic acid being the highest since this precursor is not present in the LPS, cyclic (bl-2 glucan or cellulose polymers made by this bacterium. In addition, the label incorporated from [1-14C]PEP into product was 3—fold lower when cold UDP-sugar donors were not included in the incubation mixture (data not shown). Incorporation of 14C from PEP into an organic solvent soluble phase indicates that pyruvylation occurs at the lipid carrier intermediate stage of CPS biosynthesis. The labeled glycoconjugate product soluble in organic solvent was hydrolyzed by mild acid to liberate the carbohydrate moiety from the lipid bound carrier. This resulted in the transfer of all the radioactivity' to the aqueous phase ‘when partitioned le chloroform: methanol: water (3: 2: 1) (Table 2), indicating that the label was incorporated into the aqueous-soluble carbohydrate moiety of the glycoconjugate product and not the lipid carrier itself. Gel filtration chromatography (Figure 1) of the radioactive products soluble in aqueous phase indicated that pyruvate was incorporated into several products eluting after the oligosaccharide repeating unit of 31 trifolii 843 CPS. This indicated that pyruvylation occurs 108 at several different steps in the polysaccharide biosynthetic pathway before the lipid-bound oligosaccharide is completed. Radioactive pyruvate was detected by paper chromatography after depyruvylation of fractions I, III, and IV proving that pyruvate is the labeled moiety in these various sized fractions. We examined two mutant strains (246 and 252) having more pyruvate substitutions and one strain (851) having less pyruvate substitutions in their CPS than the wild type strain, 31 trifolii 843. As anticipated, all of the strains had CPT activity (Table 3). In the absence of root exudate, there were small differences in CPT activity between the wild type and the three mutant strains, but these were not statistically significant. However, the difference in CPT between strains 246 auui 851 was statistically significant. Clover root exudate treatment did not change the levels of pyruvyl transferase activity of either wild type 843 and mutant strain 851. However, treatment of mutants 246 with Tn5 insertion before the coding region of god A gene and 252 with Tn5 insertion within nod A gene did result in statistically' significant increase cnf pyruvyl ‘transferase activity as compared to cells grown in defined. medium. Interestingly, 31 trifolii 251, a mutant strain from a different wild type strain, 51 trifolii 0403 1;; which has more pyruvate in its CPS than wild type CPS had a correspondingly higher level of CPS pyruvyl transferase activity than the wild type strain (Table 4). This suggests 109 that other loci in the megaplasmid are also affecting CPS pyruvylation. We conclude that pyruvylation of CPS in B1 trifolii occurs from phosphoenol pyruvate at lipid-bound saccharide intermediate stages. CPS pyruvyl transferase level is affected by Tn5 insertions in the gym plasmid of 31 trifolii 843 and 0403 £11 and mutations in the essential nod A gene on this plasmid affect the response of CPS pyruvyl transferase activity to root exudate of its clover host. 10. 110 LIST OF REFERENCES Abe, M., J. E. Sherwood, R. I. Hollingsworth, and F. 3. Dazzo. 1984. Stimulation of clover root hair infection by lectin-binding oligosaccharides from the capsular and extracellular polysaccharides of Rhizobium trifolii. J. Bacteriol. 160: 517-520. Amemura, A. 1984. Synthesis of (1->2)-[5-D-Glucan by cell—free extracts of Agrobacterium radiobacter IFO 12665b1 and Rhizobium phaseoli AHU 1133 Agric. Biol. Chem. 48(7): 1809-1817. Couso, R. 0., L. Ielpi, R. C. Garcia, and M. A. Dankert. 1982. Synthesis (If polysaccharides 1J1 Acetobacter xylinum. 123:617-627. Dazzo, F.a». 1982. Leguminous root nodules, p. 431-446. _Ig R. G. Burns and J. H. Slater (ed.). Experimental Microbial Ecology. Blackwell Scientific Publications, Oxford. Dedonder,R. A. and W. Z. Hassid, 1964. The enzymatic synthesis of a (0-1,2-)-1inked glucan by an extract of Rhizobium japonicum, Biochim, Biophys. Acta 90:239-248. Dubois, M., K. A. Gilles, J. K. Hamilton, P. A. Rebers, and F. Smith. 1956. Colorimetric method for determination of sugars and related substances. Anal. Chem. 28:350-356. Ielpi, L., R. O. Couso, and M. A. Dankert. 1981. Xanthan gum biosynthesis . Biochem. B iophys . Res . Commun . 102:1400-1408. Lennarz, W. J. and H. G. Scher. 1972. Metabolism and function of polyiSOprenol sugar intermediates in membrane-associated reactions. Biochim. iBiophys. Acta 265:417-441. Lowry, O. H., N. J. Rosebrough, A. L. Farr, and R. J. Randall. 1951. Protein measurement with the Folin phenol reagent. J. Biol. Chem. 193: 265-275. Masson, L., and B. E. Holbein. 1985. Role of lipid intermediate(s) in the synthesis of serogroup B Neisseria meningitidis capsular polysaccharide. J. Bacteriol. 161:861-867. ll. 12. 13. 14. 15. l6. l7. ‘ 111 Tolmasky, M. E., R. J. Staneloni, R. A. Ugalde, and L. F. Leloir. 1980. Lipid. - bound sugars in Rhizobium meliloti. Arch.‘Biochem~ Biophys. 203:358-364. Tolmasky, M. E., Staneloni, R. J. and Leloir, L. F. 1982. Lipid - bound saccharides in Rhizobium meliloti. J3 Biol. Chem. 257:6751-6757. Trevelyan, W. E., D. P. Procter, and J. S. Harrison, (1950). Detection of Sugars on Paper Chromatograms. Nature (Lond.) 166:444-445. Troy, F. A., F. E. Frerman, and E. C. Heath. 1971. The biosynthesis of capsular polysaccharide in Aerobacter aerogenes. J; Biol. Chem. 246:118-133. Sherwood, J. E., J. M. Vasse, F. B. Dazzo, and G. L. Truchet. 1984. Development and trifoliin A-binding ability of the capsule of Rhizobium trifolii. -J. Bacteriol. 159: 145-152. Sutherland, I. W. 1979. Microbial exopolysaccharides. Trends Biochem. Sci. 4:55-59. Sutherland, I. W., and Norval, M. (1970). Synthesis of exopolysaccharide by Klebsiella aerogenes membrane preparations and the involvement 0f lipid intermediates. Biochemical Journal 120:567-576. CHAPTER IV SUCCINATE METABOLISM AS RELATED TO GROWTH, BACTEROID DIFFERENTIATION AND FUNCTION IN RHIZOBIUM MELILOTI ABSTRACT The effect of succinate metabolism on growth and the in vitro" and in vivo bacteroid differentiation and function of Rhizobium meliloti was investigated with a wild type strain (LS-30), a succinate-dehydrogenase mutant strain (UR6), and a spontaneous revertant strain (UR7). Meristematic nodules induced by strains L5-30 and UR7 on alfalfa plants were effective (Fix+) in acetylene reduction, and had typical histologies and ultrastructures. Nodules induced by UR6 were ineffective (Fix-). In these Fix- root nodules, three weeks after inoculation, very few transformed bacteroids were observed and premature degeneration and lysis occurred in the central 4zone. Six weeks after inoculation, the Fix- nodules were fully senescent and the bacteria were lysed in all zones. In contrast, L5-30 and UR7 bacteroids were typically degenerated only i 1 the normal senescent zone (H? the Fix+ 112 113 nodules. This difference in nodule ultrastructures indicated that UR6 had a Bad- symbiotic phenotype (defective in bacteroid differentiation). 13 yitgg effects of succinate on growth and cell morphology in defined culture medium were concentration dependent. At a low concentration (5-10 mM), succinate was utilized preferentially before (5-10 mM) glucose or mannitol and diauxic growth was observed for the wild type strain L5- 30 but not for UR6. No changes in cell morphology were detected for any of the strains under these growth conditions. However, at a higher concentration (20 mM), both succinate or malate affected growth in mannitol (27. mM) minimal medium. Under these conditions, growth of L5-30 was inhibited, accompanied by cell elongation and pleomorphism in 15% of the population, and [14 C] mannitol uptake and mannitol dehydrogenase activities were lowered. At 20 mM, succinate did not inhibit growth, induce cell pleomorphism, or repress mannitol uptake and mannitol dehydrogenase activities in strain UR6. These results suggest that metabolism of succinate through a complete TCA cycle is necessary for normal differentiation and maintenance of alfalfa bacteroids, as well as for the in vitro effects of succinate on growth and the induction of bacteroid - like cell morphologies in Rhizobium meliloti. 114 INTRODUCTION Succinate and malate are abundant organic acids within legume root nodules (9, 18, 33) and succinate is transported and metabolized by free-living bacteria (12, 16, 22, 27) and bacteroids (6, 16, 28, 32) of different Rhizobium species. Organic acids support the highest rate of oxygen respiration by bacteroid suspensions (37) and succinate is a very effective substrate for supporting nitrogen fixation by both free—living Rhizobium species (7, 42) and bacterOids (26, 34). Utilization. of tricarboxylic acid intermediates are related to symbiotic effectiveness (1), and Rhizobium mutant strains defective in their uptake and metabolism are ineffective in nitrogen-fixing symbiosis with the host plant (13, 15, 27). :12 vitro" induction of bacterobd - like .cell morphology by succinate in _R_. trifolii (39, 40, 41) and induction of sphere to rod morphogenesis in Arthrobacter crystallopoietes (19) have been described. The objective of this work. was to' investigate the effect of succinate metabolism on growth and the in vivo and "in vitro" bacteroid differentiation in Rhizobium meliloti L5-30 (wild type effective strain). We used for this study a succinate dehydrogenase mutant strain (UR6) unabLe to grow on succinate and ineffective in symbiosis and a spontaneous 115 revertant strain (UR7) which regained succinate dehydrogenase activity, normal growth phenotype, and was effective in symbiotic nitrogen fixation (15). The present study of the histology and ultrastructure of alfalfa nodules by combined light and transmission electron microscopy showed that UR6 was released from infection threads but was defective in bacteroid differentiation in vivo". The effect of succinate metabolism (Hi the 'WJI vitro" induction (fl? bacteroid-like cell morphology was found to be concentration dependent. At low concentration (5-10 mM), succinate was used preferentially to other carbon sources and did not induce bacteroid-like cell morphology. At high concentrations (20 mM) succinate inhibited growth and induced in vitro" bacteroid formation in the wild type strain but not in the mutant strain unable to grow on succinate. This work was presented in part at the 5th International Symposium on Nitrogen Fixation, Noordwijkerhout, The Netherlands, August 28 - September 3, 1983 (p. 253). 116 MATERIALS AND METHODS Bacterial strains. 51 meliloti L5-30 (wild type strain) was kindly provided by J. Denarie. UR6 is a succinate- dehydrogenase mutant strain of R1 meliloti L5-30 and UR7 a spontaneous revertant strain. UR6 was unable to grow on succinate, lacked succinate dehydrogenase enzyme activity and induced ineffective nodules on alfalfa plants (15). UR7 regained succinate dehydrogenase enzyme activity, had normal growth phenotype and induced effective nodules in symbiosis with alfalfa plants. Symbiotic phenotype. Surface-sterilized seeds (8) of Medicago sativa var. Vernal were germinated, transferred to Jensen agar slopes (18) and were inoculated with 5x106 cells per seedling with an inoculum of cells grown for 5 days on minimal medium (MM) (15) mannitol agar slants. The tubes were incubated in a plant growth chamber with a 14-h photoperiod at 22°C (26,900 lux) and 10-h darkness at 20°C. Roots were scored for nodulation every week and plants were assayed for nitrogen fixation by the acetylene reduction technique (8) after 6 weeks. The mutant strain recovered from surface-sterilized nodules had the original Succ- Strr phenotype. Electron and light microscopy of nodules. Nodules were fixed in 4% glutaraldehyde in sodium cacodylate buffer (pH 117 7.2), post—fixed in 1% 050 dehydrated with ethanol, 4, embedded in Epon 812, and thin sections were stained by the uranyl acetate—lead citrate method (35). Transmission electron microscopy (TEM) was performed with a Philips TEM 300 at 80 kv. Semi-thin sections of Epon embedded material were deposited on glass slides and treated by the basic fuchsin - methylene blue method (17) for direct light microscopy. Growth conditions. Bacteria were grown in minimal medium containing streptomycin (100 g/ml)(MM) (15) and kept on mannitol MM slants. For growth studies, a standardized inoculum size of 0.596 glycerol-grown cells was inoculated into MM containing the indicated carbon sources. Growth was followed turbidimetrically at 660 nm in side-arm flasks with a Klett-Summerson colorimeter using the No. 66 red filter. Data shown are mean values of three replicate parallel cultures. Cell-free extract preparation. Cell-free extracts were obtained as described (2) by three l-min treatments of cell suspensions with a Branson Sonifier instrument. Protein determination. Protein was measured by the method of Lowry (20) using bovine serum albumin as standard. Glucose utilization. Glucose was measured (25) in the supernatants of culture samples centrifuged at 12,000 x g. SuCcinate utilization. The culture supernatants obtained by 12,000 x g centrifugation were filtered through 0.45-/4.m-porosity membrane filters (Type HAWP; Millipore 118 Corp., Bedford, Mass.) and succinate in the filtrates was quantitated on a Water's Associates HPLC with a Bio-Rad HPX - 87 H cuganic Acid Analysis column at 50°C using 0.005 N H2804 as the solvent, and was detected at 210 nm with a Gilson Holochrome UV monitor. Enzyme activities. Mannitol uptake was measured as previously described (14). The incubation mixture consisted of 0.05 ml of cell suspension (7x109cells/ml); 0.45 ml of M; 10 nmol of unlabeled mannitol and 2 nmol of [14C] mannitol (30,000 dpm/nmol). 0.1 ml samples were removed each 45 sec and filtered through 0.45-km filters (Type HAWP; Millipore Corp., Bedford, Mass.). The filters were washed with 2ml of MM, and the radioactivity retained was counted in an 153 7000 Liquid Scintillation-Counter (Beckman Instruments, Irvine, Cal.) using Scinti Verse II (Fisher Scientific Co., Pittsburgh, Pa.) as the scintillation mixture. Mannitol dehydrogenase activities were measured by following the reduction of NAD+ at 340 nm in a' Gilford updated Beckman model DU spectrophotometer as described (21). Electron microscopy of cells. 2-ml culture samples were removed at various times, centrifuged at 12,000 x g and the cells washed twice in phosphate-buffered saline (PBS, 10 mM potassium phosphate, 140 mM NaCl, pH 7.2). Cell suspensions deposited on Formvar-coated grids were stained by the glutaraldehyde-ruthenium red-uranyl acetate technique (24) and observed by TEM. The percentage of cells displaying 119 elongation and/or pleomorphism was recorded from a total of 100 cells examined for each sample. Oxygen consumption. L5-30 and UR6 strains were grown as shaken cultures in MM broth containing 27 mM mannitol plus 20 mM succinate. At indicated times shaking was stopped, an oxygen electrode (New Brunswick Scientific, N. J.) was introduced immediately into the cultures, and the decrease of dissolved oxygen was recorded over 21 30 minutes period. SO The oxygen electrode was calibrated with 5% Na (0%) and 2 3 distilled water saturated with air (100%) at 25°C according to manufacturer's instructions. '1“ 120 RESULTS Ultrastructure of nodules. Meristematic nodules induced by strains L5-30 and UR7 were effective in nitrogen fixation (acetylene reduction), and had typical histologies and ultrastructures throughout a growth period of 6 weeks (Figure 1). Nodules induced by UR6 were ineffective in acetylene reduction. Examination of these Fix- nodules 3 weeks after inoculation showed that the UR6 cells were released from infection threads into the host cells, but very few of the bacteria had transformed into bacteroids. Also premature degeneration and lysis of the bacteria and host cells occurred in the central zone (Figure 2). 6 weeks after inoculation, the nodules induced by UR6 were fully senescent and the bacteria were lysed in all zones. At this time, L5-30 and UR7 bacteroids’were typically degenerated only in the normal senescent zone of the Fix+ nodules. Growth studies, cell morphology and enzyme activities. 31 meliloti L5-30 had a higher growth rate in succinate (10 mM) than in glucose (10 mM) (Figure 3A). L5-30 grown in 10 mM glucose plus 10 mM succinate (Figure 3A) underwent diauxic growth and the glucose concentration in the medium decreased only during the second phase of growth. UR6 grown under the same conditions did not undergo diauxic growth (Figure 3B). L5-30 grown in 10 mM mannitol plus 10 mM FIGURE 1. Ultrastructure of root nodules on alfalfa 6 weeks after inoculation. with UR7 strain. (A)(B) Enlargement of plant cells in the central zone filled with bacteroids. Results with wild type L5-30 were identical.x 12,000. 122 FIGURE 2. Ultrastructure of root nodules on alfalfa induced 3 weeks after inoculation with UR6 strain. (A) Some degenerated bacteria in the proximal infection zone, X 4,100. (B) Enlarged profiles of rough endoplasmic reticulum in the distal infection zone, X5,500. (C) Lysosomes (arrows) in distal infection zone, X9,000. (D) Electron-dense granules (arrow) associated with bacteria in the distal infection zone, X3,700. (E) Bacteria undergoing degeneration in host cells in distal central zone. Note the ultrastructurally well preserved host cytoplasm, X6,800. 3001 1110 200 8 l 1 12 L 6.. S 100. 1* 1 5 » 4. _, E K 2 50 1 , 4o 0 60 1504 o— 4.3 “f 1001 /o /:t 2 /: a g /:/ O .- /"/ E g40 1 3o 4 25 1"°—‘—‘°_° ° ° :35 45 55 65 ”nouns 85 FIGURE 3. Growth of R_. meliloti L5-30 (A) and UR6 (B) in minimal medium containing 10 mM glucose plus 10 mM succinate (x); 10 mM glucose (0); or 10 mM succinate (o). ( -— ) Klett units; ( -- ) glucose concentration in the culture medium (mM) . 124 succinate also grew biphasically, although not as markedly as for glucose (Figure 4A). Succinate was consumed in the first phase of growth although at a slower rate than with cells grown in succinate alone (Figure 4A). In contrast, UR6 growth in 10 mM mannitol plus 10 mM succinate was the same as in 10 mM Mannitol (Figure 4B). Succinate or malate affected cell growth of L5-30 in 27 mM mannitol minimal medium. At 20 mM succinate or malate, L5-30 had lower yields at stationary phase (Figure 5). At times indicated by arrows in Figure 5, 15% of the cells were elongated and pleomorphic (Figure 6). UR6 grew on fumarate or malate and was able to transport but not to grow on succinate. 20 mM succinate or malate neither inhibited growth of UR6 in mannitol (Figure 7), nor induced cell pleomorphism (Figure 8). Under these conditions, mannitol uptake and mannitol dehydrogenase activities were lowered for the wild type strain but not for the UR6 strain (Table 1). Oxygen consumption. L5-30 consumed oxygen at a higher rate in 27 mM mannitol plus 20 mM succinate than did UR6 (Figure 9) . 300 . x 1 /X/ ‘/——-‘_'.——I , A 1. o .. 200. Bi. I a _’ E t g 0 1004 ’4 .- . 53 a ‘ 2 50 11 0 2'5 BOURS 200* O o xe 07 1 I 100‘ in t z B a " so « E _l x 25 25 35 43 1 {5 HOURS FIGURE 4. Growth of R_. meliloti L5-30 (A) and UR6 (B) in minimal medium containing 10 mM mannitol plus 10 mM succinate (l); 10 mM mannitol (O); or 10 mM succinate (0). (-—— ) Klett units: (----) succinate concentration in the culture medium (mM). 400 k // A 10 15 20 25 30 1 KLETT UNITS 400 200‘ ./° //.3:: 1.5 KLETT UNITS \' ° n \\\° 0 5.. '/ 1% 1 30 - - - - 10 15 20 25 30 HOURS FIGURE 5. Growth of _R_._ meliloti L5-30 in mannitol minimal medium containing (A) succinate or' (B) malate. In (A), labels are 27 mM mannitol only (I); 27 mM mannitol plus 20 mM succinate (x): 20 mM succinate only (0): in (B), they are 27 mM mannitol only (0); 27 mM mannitol plus 20 mM malate (I); 20 mM malate only (0). fl .‘II. x a b ..e . \ \ .' ~ c 6 FIGURE 6. Cell morphology of L5-30 strain grown in different media. Culture samples removed at indicated times (arrows) in Figure 5, were observed by TEM. L5-30 grown in mannitol (a); mannitol plus succinate (b) (c); (d). X11,000. mannitol plus malate 400: 1:3: 2001 .//' I .2 a? s ./ . _l x /o' 1 -/ so 1 ‘// m——¢-—w-r—0 3 3 a 40 fl - - , 20 40 60 HOURS 4oo« ./'/—_::: / 200‘ f4“? /.¢= KLETT UN ITS p.- O A? X i C D 501 40 - a - v 20 40 60 HOURS FIGURE 7. Growth of _R_._ meliloti UR6 in mannitol minimal medium containing (A) succinate or (B) malate. (A) 27 mM mannitol only (0); 27 mM mannitol plus 20 mM succinate (X); 20 mM succinate only (a). (B) 27 mM mannitol (o): 27 mM mannitol plus 20 mM malate (x); 20 mM malate (O). 129 \ x {-1 FIGURE 8. Cell morphology of UR6 strain grown in different media. Culture samples removed at indicated times (arrows) in Figure 7, were observed by TEM. UR6 grown in mannitol (a); mannitol plus succinate (b): mannitol plus malate (c). x 3,200. 130 TABLE IL. Mannitol uptake and mannitol dehydrogenase activities in L5-30 and UR6 strains.a Mannitol Mannitol Strain Carbon uptake %b dehydrogenase %c source nmol/min/mg nmol/min/mg protein protein L5-30 Mtl 150 100 76.0 100 " Mtl+Succ 45 29 50.5 65 " Mtl+Mal 63 42 52.2 68 UR6 Mtl 92 100 102.9 100 " Mtl+Succ 86 94 107.9 105 " Mt1+Mal 72 78 87.3 85 a Cells were grown in parallel cultures in MM containing the indicated carbon sources, Mannitol (Mtl) was used at 27 mM and succinate (Succ) or malate (Mal) at 20 mM. Cells were harvested when the cell density of mannitol-grown cells reached a Klett value of 150. [l4 C] mannitol uptake and mannitol dehydrogenase activities were measured as described in the text. b,c Expressed as percentage of the respective activity of each strain grown in mannitol. 131 96 Dissolved Oxlgon , T T T 1 V 0 4 I 12 16 20 24 MINUTES FIGURE 9. Decrease of dissolved oxygen in cell suspensions of L5-30 ( ——-) and UR6 (-——-) strains. Cells were grown as for Figures 5 and 7. At indicated times (arrows) shaking was stopped, a NBS oxygen electrode was introduced into the cultures and the decrease of dissolved oxygen was followed as a function of time. The oxygen electrode was calibrated with 5% Na 80 (0%) and distilled water saturated with air 2 3 (100%) at 25°C. 132 DISCUSSION Ultrastructural study (M3 the Fix- nodules induced by the succinate dehydrogenase mutant strain UR6 indicated that this _R_._ meliloti strain was defective in bacteroid differentiation (Bad-) in symbiosis with alfalfa plants, while a spontaneous revertant strain UR7 possessed the same Bad+ Fix+ characteristics as the wild type strain, 31 meliloti L5-30 (Figures 1 enui 2). Specific ultrastructural changes in nodules induced by the mutant strain were different from the general changes that result of nitrogen deficiency in the case of an ineffective symbiosis (36). These results indicated that symbiosis by UR6 was blocked at a step following release of bacteria from the infection threads and that a complete TCA cycle is required for normal bacteroid differentiation of R1 meliloti in alfalfa nodules. The use of this combination of strains was ideal for studying the "in vitro" effects caused in & meliloti by succinate. The in giggg succinate effect was concentration dependent. At low concentration, succinate was utilized preferentially to glucose or mannitol (Figures 3 and 4) and did not induce a bacteroid-like cell morphology. The effect of succinate on glucose growth was more accentuated than on Imannitol growth (Figures 3 and 4). In. media containing gllicose plus succinate, there was no co-utilization of 133 substrates by the wild type strain and glucose was only consumed in the second phase of diauxic growth (Figure 3). Biphasic growth has been reported for R1 meliloti grown in succinate plus lactose, and cmflls iJI the second phase of growth had higher levels of {O-galactosidase activity than did the cells in the first phase and this effect was not reversed by addition of CAMP (38). Succinate was totally consumed in the first phase of growth in media containing both mannitol plus succinate, but this consumption occured at slower rate than with cultures grown with succinate alone (Figure 4). This indicated that succinate and mannitol were co-utilized. The differences in the effect of succinate on glucose and mannitol growth could be because R1 meliloti L5- 30 grows slower on glucose than on mannitol. Similar effects have been reported in R1 leguminosarum 3841 (11) grown on a mixture. of carbon sources, e. g. succinate plus p- hydroxybenzoate. The cells co-utilized both substrates, however consumption of p-hydroxybenzoate was substantially lower than that of succinate. Consistent with results using 31 trifolii 0403 (41), succinate at high concentrations (20 mM) decreased growth yields and induced bacteroid-like morphology in R1 meliloti L5-30 (Figures 5 and 6). Finan at al. (12) reported that inhibition of R1 leguminosarum growth caused by succinate was eliminated in a medium 2 and 0.2 mM Ca+2, and they suggested containing 4 mM Mg+ that the succinate effect on cell morphology was due to chelation of divalent cations and extraction from the cells. 134 The pleomorphism induced by glycine in rhizobia (29, 29, 30, 31) can be avoided by increasing the calcium concentration in glycine-containing media (29) . The medium used in our 2 and 0.4 mud Ca+2 and. does not study contains 1 mM Mg+ contain aminoacids. Although, chelation of divalent cations may contribute to the succinate effect, this is not the only reason since the mutant strain grown under identical conditions of high succinate concentration as the wild type strain did not show growth inhibition or cell pleomorphisms at all (Figures 7 and 8). Furthermore, studies in R. trifolii (41) showed that succinate induces bacteroid-like swellings in the presence of Ca+2 in excess of the amount that could be chelated by succinate. Neither of these effects occuring with wild type strain were obtained for the succinate dehydrogenase mutant strain UR6 under identical conditions, indicating that normal metabolism of succinate is necessary for the effects of succinate on growth (at low concentration) and for the "in vitro" induction (at high concentration) of bacteroid-like cell morphology in R1 meliloti. Succinate also lowered the levels of mannitol uptake and lnannitol dehydrogenase activities in cells grown on mannitol plus succinate as compared to cells grown on mannitol alone (Table 1). In R1 leguminosarum 3841 (11) glucose or succinate produced about 50% repression of the inducible p-hydroxybenzoate catabolic system when the cells were grown on a nuxture of carbon sources, e. g. succinate 135 plus p-hydroxybenzoate. Repression of mannose uptake by succinate in R1 meliloti (3) and of glucose uptake by malate in R1 leguminosarum (10) have also been reported. Bacteroid characteristics were reported to develop in Rhizobium 1J1 microaerobic conditions (4, EH. Also, the concentration of dissolved oxygen regulated the pathway of glucose: metabolisn1 in. Pseudomonas aeruginosa (23). L5-30 decreased the concentration of dissolved oxygen in media containing succinate at a much higher rate than did UR6 (Figure 9). Presumably there are regulatory molecules in R1 meliloti that respond to the catabolic situation in the cell. We conclude that nmmabolimn of succinate through the TCA cycle is necessary in R1 meliloti for normal differentiation and function of alfalfa bacteroids and for the "in vitro" effects of succinate on growth and induction of bacteroid-like cell morphology. f 1. 10. 136 LIST OF REFERENCES Antoun, H., L. M. Bordeleau, and R. Sauvageau. 1984. Utilization. of the tricarboxylic acid cycle intermediates and symbiotic effectiveness in Rhizobium meliloti. Plant and Soil 77:29-38. Arias, A., C. Cervenansky, A. Gardiol, and G. Martinez- Drets. 1979. Phospho-glucose isomerase .mutant. of Rhizobium meliloti. J} Bacteriol. 137:409—414. Arias,'A., A. Gardiol, and G. Martinez-Drets. 1982. Transport and catabolism of D-mannose in Rhizobium meliloti. J) Bacteriol. 151:1069-1072. Avissar, Y. J., and R. Gollop. 1982. Bacteroid characteristics in microaerobic Rhizobium. Israel J. Botany 31:112-118. Avissar, Y. J., and K. D. Nadler. 1978. Stimulation of tetrapyrrole formation. in thizobium. Ajaponicum. Iby restricted aeration. J. Bacteriol. 135:782-789. Bergersen, F. J. 1977. Physiological chemistry of dinitrogen fixation by legumes p. 519-555. lg R. W. F. Hardy and W. S. Silver (ed.), A treatise on dinitrogen fixation, section III. John Wiley and Sons, New York. Bergersen, F. J., G. L. Turner, A. H. Gibson, and W. F. Dudman. 1976. Nitrogenase activity and respiration of cultures of Rhizobium spp. with special reference to concentration of dissolved oxygen. Biochim. Biophys. Acta 444:164-174. Dazzo, F. B. 1982. Leguminous root nodules, p. 431-446. 13 R. G. Burns and J. H. Slater (ed.). Experimental Microbial Ecology. IBlackwell Scientific Publications, Oxford. de Vries, G. E., P. In't Veld, and J. W; Kijne. 1980. Production of organic acids in Pisum sativum root nodules as a result of oxygen stress. Plant Sci. Lett. 20:115-123. de Vries G. E., A. A. N. van Brussel, and A. Quispel. 1982. Mechanism and regulation of glucose transport in Rhizobium leguminosarum. J; Bacteriol. 149:872-879. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 137 Dilworth, M. J., I. McKay, M. Franklin, and A. 1. Glenn. 1983. Catabolite effects on enzyme induction and substrate utilization in Rhizobium leguminosarum. J. Gen. Microbiol. 129:359-366. Finan, T. M., J. M. Wood, and D. C. Jordan. 1981. Succinate transport in Rhizobium leguminosarum. J. Bacteriol. 148:193-202. Finan, T. M., J. M. Wood, and D. C. Jordan. 1983. Symbiotic properties of C -dicarboxylic acid transport mutants of Rhizobium leguminosarum. J. IBacteriol. 154:1403-1413. Gardiol, A., A. Arias, C. Cervenansky, C. Gaggero, and G. Martinez-Drets. 1980. Biochemical characterization of a fructokinase mutant of Rhizobium meliloti. J. Bacteriol. 144:12-16. Gardiol, A., A. Arias, C. Cervenansky, and G. Martinez- Drets. 1982. Succinate dehydrogenase mutant of Rhizobium meliloti. J3 Bacteriol. 151:1621-1623. Glenn, A. R., P. S. Poole, and J. F. Hudman. 1980. Succinate uptake by free-living and bacteroid forms of Rhizobium leguminosarum. J. Gen. Microbiol. 119:267-272. Huber, J. D., F. Parker, and G. F. Odland. 1968. A basic fuchsin and alkalinized methylene blue rapid stain for epoxy-embedded tissue. Stain Techn. 43:83-87. Kawai, S., and S. Mori. 1984. Detection of organic acids in soybean nodules by high performance liquid chromatography. Soil Sci. Plant Nutr. 30:261-266. Krulwich, T. A., and J. C. Ensign. 1969. Alteration of glucose metabolimn of Axthrobacter crystallopoietes by compounds which induce sphene to rod morphogenesis. J. Bacteriol. 97:526-534. Lowry, O. H., N. J. Rosebrough, A. L. Farr, and R. J. Randall. 1951. Protein measurement with the Folin phenol reagent. J; Biol. Chem. 193:265-275. Martinez-Drets, G., and A. Arias. 1970. Metabolism of some polyols by Rhizobium meliloti. J. Bacteriol. 103:97-103. McAllister, C. F., and J. E. Lepo. 1983. Succinate transport by free-living forms of Rhizobium japonicum. J. Bacteriol. 153:1155-1162. Mitchell, C. G., and E. A. Dawes. 1982. The role of oxygen in the regulation of glucose metabolism, 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. I38 transport and the tricarboxylic acid cycle in Pseudomonas aeruginosa. J. Gen. Microbiol. 128:49-59. Mutaftschiev, S., J. Vasse, and G. Truchet. 1982. Exostructures of Rhizobiunl meliloti. FEBS Microbiol. Lett. 13:171-175. Park, J. T., and M. J. Johnson. 1949. Submicro determination of reducing carbohydrates. J. Biol. Chem. 181:149-152. Patterson, T. G., J} B. Peterson, and T. A. LaRue. 1983. Effect of supra-ambient oxygen on nitrogenase activity (C H2) and root respiration of soybeans and isolated soybean bacteroids. Plant Physiol. 70:695-700. Ronson, C. W., P. Lyttleton, and J. G. Robertson. 1981. C -dicarboxylate transport mutants of Rhizobium trifolii form ineffective nodules on Trifolium repens. Proc. Natl. Acad. Sci. USA. 78:4284-4288. ‘ Saroso, S., A. R. Glenn, and M. J. Dilworth. 1984. Carbon utilization by free living and bacteroid forms of cowpea Rhizobium strain NGR-234. J. Gen Microbiol. 130:1809-1814. Sherwood, M. T. 1972. Inhibition of Rhizobium trifolii by yeast extracts or glycine is prevented by calcium. J. Gen. Microbiol. 71:351-358. Staphorst, J. L., and B. W. Strijdom. 1971. Infectivity and effectiveness of colony isolates of ineffective glycine resistant Rhizobium meliloti strains. Phytophylactica 3:131-136. Staphorst, J. L., and B. W. Strijdom. 1972. The effect of yeast extract concentration in media on strains of Rhizobium meliloti. Phytophy lactica 4:29-32. Stovall, I., and M. Cole. 1978. Organic acid metabolism by isolated Rhizobium japonicum bacteroids. Plant Physiol. 61:787-790. Stumpf, D. K., and R. H. Burris. 1979. A micromethod for the purification and quantification of organic acids of the tricarboxylic acid cycle in plant tissues. Anal. Biochem. 95:311-315. Trinchant, J. C., and J. Rigaud. 1979. Sur les substrats energetiques utilises, lors de la reduction de C H2, par les bacteroides extraits des nodosites de Pgaseolus vulgaris. L. Physiol. Veg. 17:547-556. 35. 36. 37. 38. 39. 40. 41. 42. 139 Truchet, G. L. and F. B. Dazzo. 1982. Morphogenesis of lucerne root nodules incited by Rhizobium meliloti in the presence of combined nitrogen. Planta 154:352-360. Truchet, G., M. Michel, and J. Denarie. 1980. Sequential analysis of the organogenesis of lucerne (Medicago sativa) root nodules using symbiotically-defective mutants of Rhizobium meliloti. Differentiation 16:163- 172. Tuzimura, K. and H. Meguro. 1960. Respiration of substrates by Rhizobium in the nodules. J. Biochem. 47: 391-397. Ucker, D. S., and E. R. Signer. 1978. Catabolite- repression- like phenomenon in Rhizobium meliloti. J. Bacteriol. 136: 1197-1200. Urban, J. E. 1979. Nondividing, bacteroid-like Rhizobium trifol ii: g3 vitro induction via nutrient enrichment. Appl. Environ. Microbiol. 38:1173-1178. Urban, J. E., and D. B.E3ecttel. 1984. Fine structure of succinate-swollen Rhizobium trifolii 0403. Appl. Environ. Microbiol. 47:1178-1181. Urban, J. E., and.Fh E3. Dazzo. 1982. Succinate-induced morphology of Rhizobium trifolii 0403 resembles that of bacteroids in clover nodules. Appl. Environ. Microbiol. 44:219-226. Wilcockson, J., and D. Werner. 1979. Organic acids and prolonged nitrogenase activity by non-growing free living iRhizobium (japonicuma .Arch. .Microbiol. 122:153- 160. 140 SUMMARY The Rhizobium - legume symbiosis involves a complex sequence of interactions resulting in the formation of a root nodule that fixes nitrogen. In this dissertation, symbiotically defective mutant strains of Rhizobium were isolated and characterized with the objective of identifying biochemical events leading to a successful symbiosis. In Chapter I, _R_. trifolii 0403 mi_f_ was subjected to transposon (Tn5) mutagenesis. The most useful mutant strain obtained was 31 trifolii 251 with a single Tn5 insertion in the gym plasmid. R1 trifolii 251 showed a significant increase in attachment to clover root hairs and in lectin- binding ability when compared to the wild type strain. This strain could be useful for studying the role of improved attachment in interstrain competition in the rhizosphere. To verify that the mutant phenotype is caused by the transposon insertion, the mutation could be transferred (6)(17) to a wild type background. The linkage of the Tn5-conferred kanamycin resistance phenotype ‘with the lnutant. symbiotic phenotype would be indicative that the Tn5 insertion is the cause of the .mutation. .Alternatively, the fragment into which Tn5 has inserted could be cloned, the gene inserted into 31 trifolii wild type background (24) and the resulting mutant strains tested for the mutant symbiotic phenotype. The 251 capsular polysaccharide (CPS) was different from 141 wild type CPS in terms of non-carbohydrate substitutions. To analyze the CPS, we .used a glucuronic acid-specific lyase (PD-I) (isolated from a phage lysate of &_ trifolii 48) to hydrolyze the CPS into its oligosaccharide repeating unit (08). This approach was very useful since it allowed the detection of differences between CPS from wild type and mutant strains based. on both the kinetic study of CPS depolymerization rates with enzyme PD-I and the quantitative lH-NMR measurement of the levels of non-carbohydrate substitutions (pyruvate, acetate, and 3-hydroxybutanoate) of OS. CPS from strain 251 depolymerized at a lower rate and contained more pyruvate but less acetate substitutions than wild type CPS. These results are consistent with the proposed role of CPS non-carbohydrate substitutions in _R_._ trifolii attachment to clover root hairs and the finding that trifoliin binding to CPS is sensitive to small changes in non-carbohydrate substitutions (1). In Chapter II, we examined the CPS isolated from 131 trifolii 843 and. Tn5-induced. mutant strains obtained by extensive mutagenesis of a 14 kb DNA fragment of the gm plasmid carrying nodulation genes. CPS from mutant strains in region I (Hac) had different rates of depolymerization and different levels of pyruvic and acetic acid substitutions when compared with wild type CPS. We will discuss this more below. An 1m ylgmg assay was developed to measure the enzymatic incorporation of pyruvate into 31 trifolii CPS in 142 Chapter III. Pyruvylation was shown to occur at a lipid- bound- oligosaccharide intermediate stage. This method (was used to measure CPS pyruvyl transferase activity (CPT) in wild type and mutant strains of 31 trifolii 843 and 0403 11: having altered levels of pyruvate in the CPS. This procedure, with the necessary modifications, could be useful to study CPS acetylation. The results obtained in these two Chapters (summarized in Table 3, Chapter III) have to be interpreted in the context of the limited available knowledge about conditions and regulation of Rhizobium nod genes expression. Recent regulation data (19) (B. Rolfe and M. Djordjevic, personal communication) from research groups working with the four common mgg (D and ABC) genes (affecting root hair curling (Hac) and nodulation (Nod) phenotypes) of different Rhizobium species suggests the following: a) nod D gene is constitutively expressed at a low level in defined culture medium; and b) mo_d D gene is required in concert with a plant factor present in root exudate (plant stimulator) for the induction of 39g ABC genes which are expressed coordinately in.e: single transcriptional unit (4)(10)(l9). The results of studies of R1 trifolii 843 (wild type) and mutant strains 246, 252, and 851 desoribed in Chapters II and III are related to the control of expression of n_od D and mgg ABC genes. These results will be explained on the basis of data obtained from B. Rolfe and M. Djordjevic (personal communication). 143 R. trifolii 843. In defined culture medium, PD (the promoter of nod D) is available for binding of RNA polymerase. The strength of P is greater than that of P D ABC (promoter of 29g ABC), and 39g D gene is therefore expressed constitutively. mgg ABC genes are expressed transiently at late exponential/early stationary phase 111 defined liquid culture medium and in cells grown for five days on BIII defined agar medium. This suggests that the nutritional status of the cells may affect expression of nod genes. This figure shows the relative location of mgg D and mgg ABC genes and their respective promoters. These genes are transcribed in opposite directions. In presence of root exudate, the nascent BBQ D gene product together with a plant signal (plant stimulator) improves the quality of PABC as a promoter. As a result mRNA to 39g, ABC genes is synthesized and because two opposing promoters can not be read at the same time ggg D gene is presumably turned off. It is not known if the transient expression of mgg ABC genes in defined culture medium is due to accumulation in the cells at a certain phase of growth of the same factor (plant stimulator) that is present in the root exudate. Mutant strain 246. In this strain Tn5 is inserted between PD and the coding region of nod A gene. The Tn5 is 144 positioned so that it can not affect the expression of mgg D gene. However, it does separate effectively the mgd ABC genes from their promoter. One would predict then that this mutant would be Nod—, however it is Nod+. Tn5 is known to have weak promoters at each end of the transposon. The most likely explanation is that expression of E_d ABC genes is taking place from a Tn5 promoter. Mutant strain 252. Tn5 is inserted in the coding region of mgg A gene. This destroys mgg A or makes short nonsense products. Either (a) 29g BC genes are expressed from a Th5 promoter (A-B+C+) or (b) not expressed and the result will be (A-B—C-). A recent report suggested that the expression of mgg C was obtained from a Tn5 promoter located in mgg A in a mutant of R1 meliloti (19). Mutant strain 851. In this strain, Tn5 has inserted into the mgg D gene. Tn5 should destroy 99g D or make short nonsense products and n_og ABC genes will not be expressed because the mgg D gene product is required along with a presumptive plant signal. Cells for CPS isolation were grown on BIII plates for 5 days because the level of non-carbohydrate substitutions and lectin-binding ability of CPS changes with culture age and, at this time, CPS lectin-binding ability is maximal (1). 246 and 252 CPS had more pyruvate and 851 CPS less pyruvate than wild type CPS. CPS from strain 277 with a Tn5 insertion in the Egg C gene had levels equal to wild type CPS. Therefore the nod C gene product does not seem to be involved in 145 controlling CPS pyruvate levels. Although strain 845 lacks the ng plasmid, its CPS still had pyruvate and acetate substitutions suggesting that the role of 99g genes on the levels of pyruvate of CPS may be regulatory. This effect (direct or indirect) may be at the level of the CPS pyruvyl transferase enzyme (affecting synthesis or activity) or at the level of the assembly of the oligosaccharide substrate. A possible working hypothesis to explain our results (Table 3, Chapter III) would be that the mgg A gene product is a negative regulator of the levels of pyruvate in the CPS. In the wild type cells used for CPS isolation, mgg ABC genes are expressed transiently from PABC and therefore this strain has the "right level" of non-carbohydrate substitutions. Strain 246 is Nod+. Therefore there must be some expression of the mo_d A gene in this strain. We must assume that this expression takes place from P The nod Tn5“ A gene would then be expressed from PTnS in 246 and not expressed in 252. If the mgg A.gene product is a negative regulator, the lack of expression of the mgg A gene would result in higher levels of pyruvate in these two strains. In the 851 strain, the mgg ABC genes are not expressed because a functionalegg D gene is not available. This strain should also have higher levels of pyruvate. The fact that it has less pyruvate may be explained if 39g D gene product is also affecting the levels of pyruvate. The possibility that mgg A gene is responsible for the phenotype could be verified by complementing strain 252 with a cloned wild type nod A gene 146 and recovering the wild type levels of pyruvate in the CPS. Cells for measurement of CPT activity were grown in liquid BIII medium (-RE) to stationary phase. This assay measures incorporation of labeled pyruvate into lipid-bound oligosaccharides. The differences in CPT levels between the strains was not as marked as the differences in levels of pyruvate of the CPS (Table 3, Chapter III). This may be explained because the expression of 39g ABC genes in liquid culture occurs at early, not late stationary phase, when the cells were harvested. However, the mgg ABC genes should have been expressed transiently in the wild type strain. CPT activity was also measured in cells grown to stationary phase and then treated for 24h with root exudate (+RE) isolated after 14 days of incubation of clover seedlings with plant medium. A single batch was used for simultaneous treatment of cells from all the strains. In the presence of an active plant stimulator (+RE) and the Egg D gene product, mgg ABC genes should be expressed in the wild type strain from P and subsequently nod D should be turned off. It ABC' has been reported that low level expression of mgg D from a megaplasmid copy permitted only a 2- to 3- fold increase of mgg C expression by plant exudates as opposed to a 30-fold increase when M D was expressed at a high level from a inc-P vector (19). Since appreciable activation of the Egg ABC genes requires high levels of nod D gene .product, activation of nod ABC in presence of plant exudate may in 147 fact be low for the wild type strain. In the 246 and 252 strains the mgg D gene should be expressed all the time. The higher values obtained for CPT levels (+RE vs.-RE) may be explained if other n_o_d genes expressed in the presence of the plant stimulator affect the levels of pyruvylation. Consistent. with this possibility, other“ megaplasmid loci have been reported to be involved in the regulation of both mgg D and mo_d C genes expression in R_._ meliloti (19). Interestingly, R1 trifolii 251 a mutant strain from a different wild type strain 31 trifolii 0403 51: which has more pyruvate in its CPS than wild type CPS had a correspondingly higher level of CPT activity than the wild type strain. Strain 251 is Hac+. These results suggest that other loci in the megaplasmid different from the Hac region are affecting CPS pyruvylation. Alternatively, the mgd B gene product may be responsible for the phenotype. The 292 B gene should be or not expressed from P in 246 and expressed from P Tn5 Tn5 expressed in 252 strain. If the mom B gene product is a negative regulator of the level of CPS pyruvylation; the lack of 11951 B gene expression would also result in higher pyruvate values in CPS and a similar reasoning to the one described for mgg A would be valid. If this is the case, a Tn5 insertion in the coding region of the mgg B gene should produce higher levels of pyruvate in CPS of the mutant than in wild type CPS and should be complemented with a clone of the wild type nod B gene. If this mutant strain has normal 148 levels of pyruvate, the participation of mgd B gene can be ruled out. A third alternative would be that the mom B gene product has a positive effect on the levels of pyruvylation and the possible constitutive expression of mgg B gene from P in strains 246 and 252 versus transient expression in Tn5 the wild type strain is enough to have a positive effect on the levels of pyruvylation. A mutant of strain 252 with a Tn5 mutation in the coding region of 39g B should have lower levels of pyruvate than 252 if this alternative is correct. This mutant should be restored to wild type levels with a clone of the Egg B gene. The knowledge about conditions and factors affecting mgg genes expression is still developing. In addition, the presence and activity of a plant stimulator in root exudate is highly variable and depends on the experimental conditions for its isolation (e. 9. time of incubation of seedlings with plant medium and conditions of storage). More information is needed about the regulation of mgg genes expression, different possible effects of root exudate, and CPS structure of wild type and mutant strains under different conditions to be able to elaborate a definitive model explaining the reproducible and significant differences detected in the levels of pyruvylation in CPS from wild type and mutant strains. In Chapter IV, the effect of succinate metabolism on growth and bacteroid differentiation of Rhizobium meliloti 149 was investigated with wild type (LS-30), a succinate dehydrogenase mutant (UR6), and a spontaneous revertant (UR7) strains. UR6 had a Bad_ phenotype (defective in bacteroid differentiation) while L5-30 and UR7 were Bad+. Ex planta "succinate effects" were concentration dependent. Succinate (at low concentration) was utilized preferentially before glucose. At higher concentration, succinate decreased growth yield and induced bacteroid-like morphology in 15% of the cell population. These effects were observed for the wild type strain but not for the mutant strain, suggesting that metabolism of succinate is necessary for lg yiyg normal differentiation of alfalfa bacteroids and to produce the ex planta "succinate effects" on growth and morphogenesis in R1 meliloti. The mechanism by which succinate is utilized preferentially before glucose in R_. meliloti has not yet been established and deserves special consideration. In E1 9911 and other enteric bacteria the induction of several catabolic Operons responds to the intracellular concentration of cAMP, which is determined by the carbon source available to the cell (5). The mechanism of catabolite repression of enzyme synthesis (15) is implicit in the model for the role of the CRP(cAMP receptor protein)- cAMP regulatory complex in initiation of transcription (20). There is mounting evidence that additional factors are involved 111 catabolite repression (30). 11 factor' called catabolite modulator factor (CMF) of low molecular weight, 150 stable to acid, base and heat, has been proposed as a negative factor by contrast with cAMP, a positive factor in regulation of gene expression (2). Indole acetic acid and imidazole acetic acid can replace cAMP for expression of the arabinose operon (11). The "catabolic potential" of the cell, where cAMP is one factor determining this potential, has been related to catabolite repression (31). Sugars transported by a variety of mechanisms including the PTS mechanism for glucose and. mannitol, proton symport mechanisms such as for lactose, and the facilitated diffusion of glycerol, can all inhibit CAMP production (3,25,26). The sugar does not need to be metabolized for this inhibition to occur. In contrast to the facultative anaerobes, in Pseudomonas species the compounds producing the most severe catabolite repression are succinate and other intermediates of the tricarboxylic acid cycle (21,27). Pseudomonas is a genus of soil bacteria closely related to fast-growing Rhizobium species in its oxidative metabolism. In addition, the presence of a PEP-PTS system has not been demonstrated in these bacteria, nor is this system functional in R1 meliloti (8). In Pseudomonas species, the intracellular concentration of cAMP is not markedly affected by the carbon source used by the cells (21,27) and certainly not to the extent seen in E1 g_o_l_i (5). Exogenously applied cAMP does not reverse catabolite repression due tx> succinate (21,27) and the intracellular cAMP levels are unchanged by the onset 151 of succinate-related catabolite repression (21). The pathway of glucose metabolism is regulated by the concentration of dissolved oxygen in Pseudomonas aeruginosa (18). Regulatory processes in carbon metabolism in Rhizobium and. Bradyrhizobiunl are run: well understood. Glucose catabolite repression of polyol metabolism has been reported for R1 trifolii (23) and R1 meliloti (16) but not for B1 japonicum (12). Differences have been observed between slow (B1 japonicum) and fast (31 meliloti) growing species. Malate represses hydrogenase activity' ha B1 japonicum (14) and exogenously applied cAMP can overcome this repression. Intracellular cAMP pool sizes have also been reported to change with the carbon source used in B1 japonicum (14, R. McClung and B. K. Chelm, Abstr. Second International Symposium on the Molecular Genetics of the Bacteria-Plant Interaction. 1984, #59). Intracellular cGMP levels were also found to change in microaerophilic-aerobic shift experiments :hi‘BL japonicum (13). 2h) contrast, catabolite repression- like phenomena caused by succinate on galactosidase activity in 31 meliloti (29) was not reversed by addition of exogenous cAMP. It has been reported that during growth in two carbon sources, succinate dehydrogenase activities paralleled intracellular cAMP which varied independently of the carbon source whereas peak succinate-transport rates of succinate- grown cells were 4-fold those of arabinose-grown cells in B1 japonicum (S. Ferrenbach, and J. Lepo Abstr. 6th 152 International Symposium (n1 Nitrogen Fixation 1985,#4-23). C4-dicarboxylates, succinate, fumarate and lnalate have a symport transport with protons in E1 9911 (9) and are actively transported by a common inducible system in 31 leguminosarum (7), 31 trifolii (22) and g; japonicum (28). In wild type R. meliloti L5-30 we obtained similar ex planta effects with succinate and malate suggesting a common cause. However, the dicarboxylate transport is not likely to be responsible for these effects since the UR6 succinate dehydrogenase mutant strain is able to tmansport succinate and malate and did not show these effects. In contrast, our results suggest that metabolism of succinate rather than uptake is necessary for the 1m y1y9 and ex planta "succinate effects". In addition, 'we found. that. the *wild. type and mutant strains consumed oxygen at different rates in the presence of succinate. These results suggest that there may be regulatory molecules in R1 meliloti that respond to the metabolism of succinate. LI ST OF REFERENCES l. Abe, M., J.E. Sherwood, R.I. Hollingsworth, and F.B. Dazzo. 1984. Stimulation of cdover root hair infection by lectin-binding oligosaccharides from the capsular and extracellular polysaccharides of Rhizobium trifolii. J. Bacteriol. 160:517-520. 2. Dessein, A., F. Tillier, and A. Ullmann. 1978. Catabolite modulation factor: physiological properties and in vivo effects. Mol. Gen. Genet. 162:89-94. 3. Dills, S. S., A. Apperson, M. R. Schmidt, and M. H. Saier, Jr. 1980. Carbohydrate transport in bacteria. Microbiol. Rev. 44:385-418. 153 4. Egelhoff, T. T., Fischer, R. F., Jacobs, T. W., Mulligan, J. T. and S. R. Long. 1985. DNA 4:241-248. 5. Epstein, W., L. B. Rothman-Denes, and J. Hesse. 1975. Adenosine 3',5'-cyclic monophosphaate as mediator of catabolite repression in Escherichia coli. Proc. Natl. Acad. Sci. USA 72:2300-2304. 6. Finan, T. M., A.M. Hirsch, J.A. Leigh, E. Johansen, G. A. Kuldan, S. Deegan, G. C. Walker, and E. R. Signer. 1985. Symbiotic mutants of R; nmfliloti that uncouple plant from bacterial differentiation. Cell 40:869-877. 7. Finan, T. M., J. M. Wood, and D. C. Jordan. 1981. Succinate transport in Rhizobium leguminosarum. J. Bacteriol. 148:193-202. 8. Gardiol, A., A. Arias, C. Cervenansky, C. Gaggero, and G. Martinez-Drets. 1980. Biochemical characterization of a fructokinase mutant of Rhizobium meliloti. J. Bacteriol. 144:12-16. 9. Gutowski, S. J. and H. Rosenberg. 1975. Biochem. J. 152:647-654. 10. Jacobs, T. W., Egelhoff, T. T. and S. R. Long. 1985. J. Bacteriol. 162:469-476. v 11. Kline, E. L., C. S. Brown, V. Bankaitis, D. C. Montefiori, and K. Craig. 1980. Metabolite gene regulation: imidazole and imidazole derivatives which circumvent cyclic adenosine 3',5'-monophosphate induction of the E_. coli L- arabinose operon. J. Bacteriol. 141:770-778. 12. Kuykendall, L. D. and G. H. Elkan. 1977. Some features of mannitol metabolism in Rhizobium japonicum. J. Gen. Microbiol. 98:291-295. 13. Lim, S. T., H. Hennecke, and B. Scott. 1979. Effect of cyclic guanosine 3',5'-monophosphate on nitrogen fixation in Rhizobium japonicum. J. Bacteriol. 139:256-263. 14. Lim, S. T., and K. T. Shanmugan. 1979. Regulation of hydrogen utilization in Rhizobium japonicum by cyclic AMP. Biochim. Biophys. Acta 584:479-492. 15. Magasanik, B. 1961. Catabolite repression. Cold Spring Harbor Symp. Quant. Biol. 26:249-262. 16. Martinez-Drets, G. and A. Arias. 1970. Metabolism of some polyols by Rhizobium meliloti. J. Bacteriol. 103:97- 103. 154 17. Meade, H. M., S. R. Long, G. B. Ruvkun, S. E. Brown, and F. M. Ausubel. 1982. Physical and genetic characterization of symbiotic and auxotrophic mutants in Rhizobium meliloti induced by transposon Tn5 mutagenesis. J. Bacteriol. 149:114-122. 18. Mitchell, C. G. and E. A. Dawes. 1982. The role of oxygen in the regulation of glucose metabolism, transport and the tricarboxylic acid cycle in Pseudomonas aeruginosa J. Gen. Microbiol. 128:49-59. 19. Mulligan, J. T. and S. R. Long. 1985. Induction of Rhizobium meliloti nod C expression by plant exudate requires nod D. Proc. Natl. Acad. Sci. USA 82:6609-6613. 20. Pastan, I., enui S. Adhya. 1976. Cyclic adenosine 5'- monophosphate in Escherichia coli. Bacteriol. Rev. 40:527- 551. 21. Phillips, A. T., and L. M. Mulfinger. 1981. Cyclic Adenosine 3',5'-Monophosphate Levels in Pseudomonas aeruginosa during induction and carbon catabolite repression of histidase synthesis. J. Bacteriol. 145:1286-1292. 22. Ronson, C. W., P. Lyttleton, and J. G. Robertson. 1981. C4-dicarboxylate transport mutants of g; trifolii from ineffective nodules on Trifolium repens. Proc. Natl. Acad. Sci. USA. 78:4284-4288. 23. Ronson, C. W., and S. B. Primrose. 1979. Effect of glucose on polyol metabolism by Rhizobium trifolii. J. Bacteriol. 139:1075-1078. 24. Ruvkun, G. B., and F. M. Ausubel. 1981. A general method for' site-directed. :mutagenesis 1J1 prokaryotes. Nature (London) 289:85-88. 25. Saier, M. H., Jr. 1977. Bacterial phosphoenolpyruvate:sugar phosphotransferase systems: structural, functional, and evolutionary relationships. Bacteriol. Rev. 41:856-871. 26. Saier, M. H., Jr. 1979. The role of the cell surface in regulating the internal environment, p. 169-225. 111 J. R. Sobatch and L. N. Ornston (ed.), The bacteria, vol.3. Academic Press, Inc., New York. 27. Siegel, L. S., P. B. Hylenon, and P. V. Phibbs, Jr. 1977. Cyclic adenosine 3',5'-monophosphate levels and activities of adenylate cyclase and cyclic adenosine 3',5'- monophosphate jphosphodiesterase iJl Pseudomonas and Bacteroids. J. Bacteriol. 129:87-96. 155 28. Stovall, I., and M. Cole. 1978. Organic acid metabolism by isolated Rhizobium japonicum bacteroids. Plant Physiol. 61:787-790. 29. Ucker, D. S., and E. R. Signer. 1978. Catabolite- repression-like phenomenon in Rhizobium meliloti. J. Bacteriol. 136:1197-1200. 30. Ullmann, A. and A. Danchin. 1979. 1980. Role of cyclic AMP 1J1 regulatory mechanisms 1J1 bacteria. Trends Biochem. Sci. 5:95-96. - 31. Wanner, B. I“, 12. Kodaira, and F2 (3. Neidhardt. 1978. Regulation. of lac (operon expression. Reappraisal of the theory of catabolite repression. J. Bacteriol. 136:947-954. IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII 1111111111111 111 11111 11111111 1111111 11111111111111111111111