TREHALOSE IS NOT a DETERMINANT of NODULATION DOMINANCE in Rhizohium Leguminosarum BIOVAR Phaseoli

TREHALOSE IS NOT A DETERMINANT OF NODULATION DOMINANCE IN Rhizohium leguminosarum BIOVAR phaseoli

MARY CHRISTINA RWAKAIKARA SILVER

A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY

UNIVERSITY OF FLORIDA

1994 ACKNOWLEDGMENTS

I am very grateful to Dr. David Hubbell, the chairman of my advisory committee, for the effort he made to get this work accomplished. I appreciate the constructive comments from the rest of the committee members, received in the course of the

study. I am thankful to Dr. E. Triplett for his advice and

Jocylene Milner for supplying the Rhizobium strains.

I am extremely indebted to Dr. Coleman for all she has

done for me and my son while we lived in the USA. She helped me both academically and socially, and this helped me get over many hurdles.

I appreciate the assistance in chromatographic analytical

procedures rendered by Drs. Bill Davis and Linda Lee and their

laboratory staff, Dong Ping Dai and Shihsien Chen. I would

like to thank Dr. S. Rao for the use of his equipment and

supplies. I am thankful to the following: Mary McLeod, Dr.

Amiel Jarstfer, Chris Pedersen, Larry Schwandes, Loan Ngo,

Jean Thomas, Odemari Mbuya, Mark Ou, Amiel Varshovi, Marla

Buchanan, Kafui Awuma, Daniel Mugeni , Pauline Rothwell, Mona

Mehta, Jimmy Rojas, and P. Srikanth for their help in

facilitating my work.

I am very much indebted to the following: Mr. and Mrs.

Mukasa, Mrs. Ouko and family, Mr. Awuma Kafui, Drs. Peter and

Rose Kizza, all of whom, at different times, took care of my

11 .

son. I appreciate help in various aspects rendered by Mr. and

Dr. Olivia Mijumbi, Dr. Robinah Ssonkko, Ruth Muwonge, Paul

Sendi, George Lukwago and Jerry Lwanga. I acknowledge moral support from friends outside Gainesville, in particular, Drs

Keren Kabadaaki, Stephen Tenywa, Moses Tenywa, Matthias

Magunda, Wellington Masamba, Mr. Jovan Tibezinda, Mr. Alfred

Yamoah, Mrs. Justine Bukenya and Mrs. Drusilla Mukasa.

I would like to thank all my friends in Uganda for helping my family members, taking care of other administrative concerns and their constant letters and messages of encouragement. I am particularly grateful to Mr. and Mrs.

Ssali, Dr. E. Sabiiti, Dr. S. Kyamanywa and Dr. Mary Okwakol.

I am thankful to my family for all their love and support.

I appreciate the great responsibility of looking after my son

for three years.

I would like to register my thanks to the financial

institutions for supporting my study program. I am thankful

to USAID/UGANDA/MFAD Program and the University of Florida.

I also appreciate the contribution from Drs. J. Sartain and

Peter Kizza. I appreciate the prompt interventions by Hon.

Victoria Ssekitoleko and Hon. Karen Thurman.

I am grateful to Cindy Roark for helping in the

preparation of the dissertation.

Lastly, but not least, I appreciate the role played by

Maria Cruz and Betty Fisher for facilitating my International

dealings and other administrative matters.

Ill , 5

TABLE OF CONTENTS PAGE ACKNOWLEDGMENTS

ABSTRACT ^

CHAPTERS

1 I INTRODUCTION

II LITERATURE REVIEW 6

Biological Nitrogen Fixation 6 The 'Lequme-Rhizobium Symbiosis 6 Factors Affecting the l,equme-Rhizobium 7 Symbiosis ..... ^ Interstrain Competition or Nodulation Dominance 7 Trehalose IB Physical Properties IB Occurrence ..... 14 Trehalose Biosynthesis 1 Trehalose Catabolism 17 Trehalose 20 Functions of _ Trehalose in Plant-microbe Interactions . . . . 26 L. . . . . 29 The Common Bean {PhssBolus vulgsris ) B2 III MATERIALS AND METHODS

Experiment la: Accumulation of Trehalose by Rhizobium leguminosarum Biovar phaseoli Strain CEB and its Tn5 Derivative Mutants Cultured Under Varying Conditions B2 Screening for Trehalose Negative Mutants . . B2 Quantification of Trehalose and Other Saccharides in Bacterial Cells Grown

. . BB Under Different Salt Concentrations . Analysis of Nodules for Trehalose Content . B4 Fixation B7 Sectioning 31 Light Microscopy 31 Electron Microscopy 31 Dry Matter Yield and Nitrogen Content of Bean Plants B8 Experiment lb: Growth of CEB and Tn5 Mutants in YEMB of Varying Salt Concentrations . . . B8

IV f

CHAPTERS PAGE

2oh -i ti/n iG^uitiino 2 3. • Growth of Rh sarum Biovar phaseoli Strain CE3 and Tn5 Derivative Mutants on Trehalose Medium. . . . Experiment 2b: Growth of CE3 and Tn5 Derivative Mutants in Yeast Extract Mannitol Broth of Varying pH Experiment 3: Modulation Dominance Trials . . . Statistical Analysis

IV RESULTS 42 Characterization of Strains Used in the Study 42 Accumulation of Trehalose by Rhizohiuui leguminosaruiu Biovar phasBoli Grown Under Different Salt Concentrations 42 Effect of Sodium Chloride on the Growth of Rhizobium leguminosarum Biovar phaseoli Strain CE3 and its Tn5 Mutants 43 Carbohydrate Content in Nodules 43 Dry Matter Yield and Nitrogen Content of Bean Plants (Phaseolus vulgaris Cultivar with Rhizobium jCarbon Utilization by Rhizobium leguminosarum Biovar phaseoli Strain CE3 and its Tn5 Mutants grown on Trehalose and Mannitol 83 Accumulation of Trehalose Under Varying pH Conditions 84 The Response of Rhizobium leguminosarum Biovar phaseoli to Chloramphenicol 89 The Effect of Temperature on the Accumulation of Trehalose and Arabinose in Strains CE3 and CE3::Tn514b 91

V CHAPTERS PAGE

Nodulation Studies 9'^

VI SUMMARY AND CONCLUSIONS 101

APPENDICES 107

LITERATURE CITED HO

BIOGRAPHICAL SKETCH 145

VI LIST OF TABLES

TABLE page

4-1 Trehalose content of Rhizobium leguminosarum biovar phaseoli strain CE3 and its Tn5 derivative mutants grown under increasing salt concentrations 45

4-2 Accumulation of trehalose and arabinose by Rhizobium leguminosarum biovar phaseoli strain CE3 and its derivative Tn5 mutants grown in basal medium ^5

4-3 Arabinose content in Rhizobium leguminosarum biovar phaseoli strain CE3 and its Tn5 mutants grown under increasing salt concentrations 46

4-4 Carbohydrate content of nodules formed by Rhizobium leguminosarum biovar phaseoli strain CE3 and its derivative mutants on bean plants 18 DAP. 47

4-5 Carbohydrate content of nodules formed by Rhizobium leguminosarum biovar phaseoli strain CE3 and its derivative Tn5 mutants on Phaseolus vulgaris cultivar Midnight, 18 DAP 48

4-6 Carbohydrate content in nodules of salt stressed plants formed by Rhizobium leguminosarum biovar phaseoli strain CE3 and its derivative Tn5 mutants 50

4-7 Dry matter yield of plants inoculated with Rhizobium leguminosarum biovar phaseoli strain CE3 and its Tn5 mutants, 18 DAP 60

4-8 Nitrogen content of plants inoculated with Rhizobium leguminosarum biovar phaseoli strain CE3 and its derivative Tn5 mutants, 18 DAP 60

4-9 Nitrogen content of plants inoculated with Rhizobium leguminosarum biovar phaseoli strain CE3 and its derivative Tn5 mutants grown under salt stress conditions 61

4-10 Accumulation of Arabinose by Rhizobium leguminosarum biovar phaseoli strain CE3 and its derivative Tn5 mutants under varying pH conditions 62

vii TABLE PAGE

4-11 Effect of chloramphenicol on trehalose and arabinose content in Rhizohium leguminosarum

biovar phaseoli strain CE3 and its mutants . . . . 64

4-12 Accumulation of arabinose by strain CE3 and CE3:;Tn5 under different temperature regimes . . . 68

4-13 Effect of inoculum ratios on percent nodule occupancy by strain CE3 and CE3::Tn514b 69

4-14 The effect of sodium chloride on nodule occupancy by strains CE3 and CE3::Tn514b 70

4-15 Dry matter yield as affected by inoculum ratio . . 71

4-16 Competitive ability of strains CE3 and CE3::Tn514b in mixed inocula 71

Vlll LIST OF FIGURES

FIGURE PAGE

2-1 Trehalose synthesis and hydrolysis (Adapted from Mellor, 1992) 16

4-1 Growth of CE3 and Tn5 derivative mutants in increasing levels of NaCl over time 44

4-2 Light photomicrographs of bacteroid tissue in strain CE3 52

4-3 Light photomicrographs of bacteroid tissue of strain CE3::Tn514b 55

4-4 Electron micrographs of nodule sections from strain CE3 57

4-5 Electron micrographs of nodule section from strain CE3::Tn514b 59

4-6 Growth of CE3 and Tn5 derivative mutants in YETB and YEMB over time 65

4-7 Growth of CE3 and Tn5 derivative mutants in YEMB of varying pH values over time 66

4-8 Effect of temperature on the accumulation of arabinose in CE3::Tn514b 67

IX Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy

TREHALOSE IS NOT A DETERMINANT OF NODULATION DOMINANCE IN Rhizohium leguminosarum BIOVAR phaseoli

Mary Christina Rwakaikara Silver

April 1994

Chairman: D.H. Hubbell Major Department: Soil and Water Science

There is need for distinctive marker traits to aid in the selection of Rhizobium inoculant strains that are

effective and exhibit nodulation dominance. Trehalose, an

unusual (nonreducing) a,a-glucose disaccharide, has been

documented in Bradyrhizobium and Rhizobium species. Its

role as a predictor of inoculant quality rhizobia has been

suggested. Experiments were conducted to determine any

relationship that might exist among trehalose accumulation,

survival under conditions of environmental stress and

nodulation dominance. Rhizobium leguminosarum biovar phaseoli strain CE3,

resistant to 200 ^ig ml”^ streptomycin, and four derivative

Tn5 mutants, resistant to 200 jig ml’^ each of streptomycin

and kanamycin, were grown in an arabinose mannitol defined

X .

medium formulated to subject the cells to a) elevated salt concentrations, b) varying pH, c) elevated temperature, and d) antibiosis.

In the absence of salt, and for the same dry weight of cells, strain CE3 accumulated 32.12 ]Xg trehalose compared to

4.2 }ig in CE3:;Tn514b. In all five strains, the amount of trehalose decreased with increasing salt concentration.

Strains CE3 and CE3::Tn514b had a 53% and 27% increase,

respectively, in nodule trehalose when the plants were grown

in a medium containing 0.05 M NaCl . Light and electron microscopy indicated that salt-stressed nodules had reduced bacteroid tissue, poly-J5-hydroxybutyrate, volutin granules,

and leghemoglobin

No trehalose was detected in the cell extracts when the

strains were subjected to stressing conditions of pH,

temperature or antibiosis.

Nodule occupancy studies indicated that inoculum ratio

was the main factor determining the percent strain

representation in nodules.

Trehalose had no osmoregulatory role in the test

rhizobia nor did it improve their survival under other

environmental stress conditions which were investigated.

High levels of trehalose in a strain did not confer

nodulation dominance in this study. Definite conclusions

will require use of a trehalose negative Rhizobium

strain (s) .

XI CPIAPTER I INTRODUCTION

The increasing need for expensive chemical nitrogen

fertilizers in agriculture and optimistic expectations from molecular genetics are some of the factors that led to a

reappraisal of biological nitrogen fixation in the 1970s

(Quispel, 1988) . The identification and successful transfer

of the nitrogen fixing {nif) genes from Klebsiella pneumoniae into Escherichia coli in 1971 (Postgate, 1982)

was a great scientific achievement. A possibility was

visualized whereby plants, especially cereals and other

organisms capable of fixing nitrogen, could be genetically

engineered. Genetic manipulations would also be extended to

known nitrogen-fixing systems in order to enhance their

performance (Bauer, lj|90) . However, the transfer of

bacterial genes into the plant genome, per se, could not

confer to those plants the nitrogen-fixing ability

(Postgate, 1987) . In addition, such propositions were met

with skepticism from environmentalists. However, optimism

was expressed by Bauer (1990) after a careful study of

genetically modified organisms.

The contribution of the legume- Rhiz obi urn symbiosis to

the nitrogen cycle has been recognized for over a century

(Hellriegel and Wilfarth, 1888) . In a series of complex

1 .

2 interactions between the host legume plant and the microsymbiont, root nodules are formed that are the centers of dinitrogen fixation.

Host nodulation and subsequent reduction of atmospheric dinitrogen requires the presence of homologous, infective

and efficient strains of Rhizobium or Bradyrhizobium in the

rhizosphere. Often the nodule bacteria for a given legume

are either absent, or present in low numbers, or are

ineffective (Lowendorf, 1980) . In such situations,

inoculation of legume seed or soil with appropriate

Rhizobium strains is a recommended agricultural practice,

designed to maximize symbiotic nitrogen fixation. Despite

improvements in inoculation methods (Boonkerd et al., 1978;

Sparrow and Ham, 1983) and the selection of strains with

high nitrogen-fixing potential (Kishinevsky et al., 1984)

sometimes inoculation does not lead to increased plant

growth and crop yields (Stephens, 1967; Ham et al., 1971;

Zdor and Pueppke, 1988) . From the extensive reviews of

Lowendorf (1980) and Gibson and Jordan (1983) lack of

response to inoculation may be attributed to host plant or

bacterial traits, incompatible interactions between the

symbionts or to environmental stresses.

When the inoculant Rhizobium strain is introduced in

the soil, it is confronted by a general microbiota that has

adapted to adverse ecological factors (Trinick, 1982)

Antagonistic interactions develop between rhizobia and

microbial populations (Chatel and Parker, 1973; Chowdhury, .

Furthermore, soils 1977; Scotti et al . , 1982/ Vidor, 1982). that have supported legumes before may harbor rhizobial populations greater than 1000 cells per gram (Trinick,

1982). These rhizobia may be effective or ineffective.

Consequently, the inoculant strains have to compete with the native ones for growth and nodule sites, a concept advanced by Nicol and Thornton (1941)

Reports suggesting differential strain representation among host nodules include those of Means et al., 1961;

Skrdleta, 1970; Herridge and Roughley, 1975;, Bromfield and

Gareth, 1980; May and Bohlool, 1983; Peterson et al . , 1983;

Zdor and Pueppke, 1988/ and Beattie et al., 1989. Yet the specific factors that are involved in preferential

nodulation are not clear. Lowendorf (1980) , Trinick (1982) and May and Bohlool (1983) outline several reasons for the observed nodulation dominance by inoculum or soil rhizobial strains. The factors range from those related to the partners and their interactions to physiochemical, biological and environmental conditions. Therefore, understanding the phenomenon or phenomena that lead to one strain forming the majority of the nodules on a given host

(strain dominance) is important, especially in efforts to screen for inoculant-quality strains of rhizobia.

The nonreducing disaccharide trehalose (a-D- glucopyranosyl- (1-1) a- glucopyranoside) has been reported to accumulate in strains of Bradyrhizobium japonicum and Rhizobium leguminosarum biovar phaseoli and others 4

(Streeter, 1985; Hoelzle, 1990) . Trehalose also occurs in other organisms such as algae, bacteria, fungi, invertebrates, birds and yeasts (Elbein, 1974) . In fungi, this sugar was associated with periods of reduced growth.

Its synthesis was intense during differentiation and periods of starvation. Generally, trehalose serves mainly as a specialized storage carbohydrate in fungi (Thevelein, 1984/

Earlier, Crowe et al . (1984) had Donnin et al . , 1988). underscored the role of trehalose as a storage compound in fungi. Instead they emphasized its role in maintaining membrane integrity during desiccation. In Escherichia coli

(Larsen et al., 1987) and fungi (Gadd et al., 1984) trehalose is noted as an effective osmoticum.

The functions of trehalose in the legume-Rhizobium

symbiosis are not clearly evident (Streeter, 1985) . He

obtained a negative correlation between trehalose and

acetylene reduction and he doubted whether such observations

reflected cause and effect. He speculated that trehalose played a role in the survival of Bradyrhizohium japonicum in

soils. Later, Hoelzle and Streeter (1990b) showed that,

under low oxygen levels, bacteroids accumulated greater

amounts of trehalose than at higher levels.

Based on the knowledge gained from earlier studies

implicating trehalose accumulation in stress environments,

and realizing that survival of inoculum strains under such

conditions would influence their competitiveness, the

following hypothesis was developed. : .

Trehalose imparts a competitive advantage to Rhizobium or Bradyrhizobium strains by improving survival under environmental stress conditions such as extremes in temperature, moisture, pH, and as well as antibiosis and salinity. Thus, the objectives of the study were as follows

1) To document trehalose accumulation by Rhizobium leguminosarum biovar phaseoli strain CE3 and its Tn5

mutants under stress conditions of extremes of

temperature, pH, osmotic pressure, and of presence of

antibiosis

2) To determine relative nodule occupancy by strain CE3 and

its derivative Tn5 mutants on their homologous host

Phaseolus vulgaris under stress conditions.

3) To assess any relationship that might be apparent between nodule occupancy and trehalose accumulation

traits in the symbiosis under stress conditions. , .

CHAPTER II LITERATURE REVIEW

Biological Nitrogen Fixation

Biological nitrogen fixation is an exclusive attribute of prokaryotes belonging to the eubacteria and archaebacteria kingdoms (Woese, 1987). Some of the diazotrophs, such as Azotobacter vinelandii, Rhodospirillum rubrum, and Xanthobacter autotrophicus occur in a free state while others, like cyanobacteria, Rhizobium spp. and

Bradyrhizobium spp., have symbiotic associations with eukaryotes, especially plants (Sprent and Sprent, 1990)

It is difficult to give an accurate estimate of the amount of nitrogen fixed annually on a global scale.

Delwiche (1977) suggests that about 130 Tg (130 xlO® kg) of nitrogen may be fixed yearly by biological processes,

compared to less than 50 Tg produced by industrial and

atmospheric fixation. The largest quantity of biologically

fixed nitrogen is realized from the legumes in symbiosis with rhizobia.

The Leaume- Rhizobium Symbiosis

The macrosymbionts, the legume plants, belong to the

family Leguminosae (Polhill and Raven, 1981) . All rhizobial

genera belong to the family Rhizobiaceae . The most

important visible characteristic of the legume-Rhizobium

6 . .

interaction is the nodule. The nodulation process and subsequent nitrogen fixation involves an intricate cascade of physical, chemical, biochemical and molecular

interactions between the symbionts (Nap and Bisseling, 1990;

Kjine, 1992; Quispel, 1988; Verma, 1992; Vesper and Bauer,

1987)

Factors Affecting the LeQume-Rhizobium Symbiosis

Jardim-Freire (1984) has listed factors that act either

singly or together to influence the success of dinitrogen

fixation by the legume-Rhizobium association. They include

i) infectiveness and efficiency of Rhizobium strains present

in the inoculum and/or in the soil in relation to the

species and varieties of the legume; ii) competitive ability

of introduced rhizobia in relation to the native rhizobial population; iii) ability of the host to supply the micro-

symbiont's nutritional needs; and iv) environmental

conditions, especially factors limiting plant growth in the

soil that act on both the bacteria and the host. Detailed

discussions of these factors have been reviewed by Lowendorf

(1980), Gibson and Jordan (1983), Jardim-Freire (1984) and

Sprent and Sprent (1990)

Interstrain Competition or Nodulation Dominance

Miller and May (1991) reviewed the inability to

introduce Rhizobium strains in fields already inhabited by

native rhizobia. Often, this has led to poor or no response with to inoculation. Roughley et al . (1976) were confronted

this problem when trying to introduce more efficient strains 8 of Rhizohium trifolii in the presence of naturalized populations in southeastern Australia. Both simple mixtures of rhizobia and complex indigenous rhizobial populations are

distributed unequally in nodules (Bottomley, 1992) . Such reports evoke the concept of interstrain competition for

nodule sites advanced by Nicol and Thornton (1941) . Since then aspects of interstrain competition have been demonstrated and reviewed (Vincent and Waters, 1954;

Peterson et al 1983; Amarger, 1984; Robinson, 1969; . ,

Dowling and Broughton, 1986; Cregan et al., 1989; Triplett,

1990; Bottomley, 1992) . Integrating the vast literature on competition, Giller and Wilson (1993) reaffirmed that various rhizobial strains exhibit different abilities to compete for nodule occupancy. Secondly, the relative

success in achieving nodule occupancy was affected by

environmental conditions, and by host plant species and

cultivar. Additionally, the initial population size, its

distribution in soil and competition from other organisms,

are critical factors in nodulation dominance.

Hubbell (personal communication) is opposed to the

terminology 'interstrain competition' or 'strain competitive

ability' . He considers the terms inaccurate as they do not

describe the observations or phenomena. He proposes the

terms 'dominant' and 'subordinate' in reference to the

relative ability of a strain to form nodules in the presence

of other strains in a mixture. The rationale for the

alternative terms is the obvious lack of a limiting 9 factor (s) for which strains could possibly be competing.

Hubbell (personal communication) also questions the concept of infection sites. Such a concept implies existence of limited discrete, highly localized sites on the root hair which are chemically or physically predetermined as highly amenable to infection prior to exposure to rhizobia. He notes that experimental evidence for such sites is lacking.

Weaver and Frederick (1974) found that the ratio of a

strain in an inoculum was crucial in determining its ultimate nodulation success. Giller and Wilson (1993), commenting on reports that deal with inoculum size, noted the apparent inconsistency in the findings. Some reports

(Weaver and Frederick, 1974; Singleton and Tavares, 1986) emphasized cell numbers of the indigenous rhizobial population. On the contrary, Moawad et al. (1984) detected no significant difference in the resident populations of various serogroups of Bradyrhizobium japonicum in bulk soil

or rhizosphere of host and nonhost plants. Yet, strains of

USDA 123 were always the most successful in competing for nodule occupancy. Furthermore, Giller and Wilson (1993)

quote Sylvester-Bradley et al. (in press) as having been

able to isolate rhizobial strains that give inoculation

response in a tropical pasture legume, kudzu (Pueraria

phaseoloide ) . This occurs even when inoculated with as few

as 5x10^ per seed, in soils already containing 3.5x10^

compatible rhizobia g“^ of soil. ,

Notable from rhizobial population data is the wide range in numbers given, from <10-10^ g of soil (Bottomley, 10^-10^ 1992) . Within a single field, the numbers vary from g"^ soil (Wollum and Cassel, 1984) . In addition, a great diversity exists among the strains. The method employed to enumerate rhizobia (most probable number technique)

(Vincent, 1970) is unable to discriminate between strains.

Application of other techniques, such as multilocus enzyme electrophoresis, showed that at least 10-20 strains were capable of nodulating a particular host (Pinero et al.,

1988) .

Although many findings have demonstrated nodulation dominance of one strain over the others, rarely has a specific factor been identified. Rhizobium leguminosarum biovar trifolii strain T24 produces a bacteriocin, trifolitoxin (Triplett and Barta, 1987; Triplett, 1988;

Triplett et al., 1989; Triplett, 1990), which is effective against most Rhizobium leguminosarum biovar trifolii, biovar

viciae, biovar phaseoli, and Rhizobium fredii but not

Rhizobium meliloti, Bradyrhizobium and plant pathogenic

bacteria (Triplett and Barta, 1987) . Unfortunately, strain

T24 induces ineffective nodules on clover. Early efforts to transfer trifolitoxin genes into effective Rhizobium by conjugation of recombinant pTFXl were not practical. Such conjugants are usually unstable in the absence of selection

Long et al., 1982). pressure (Lambert et al . , 1987;

Recently, Triplett (1990) employing the marker exchange . .

11 technique which utilizes homologous DNA present in pTFXl, constructed a trifolitoxin-producing exoconjugant TA1::10-

15. The latter is very competitive when coinoculated with a trifolitoxin sensitive reference strain. In addition, strain TAl:: 10-15 is stable. Triplett (1990) was optimistic that similar manipulations could be extended to other

Rhizobium and Bradyrhizobium species.

In practice, use of inocula that contain both the specific Rhizobium and another microorganism which complements the Rhizobium to enhance colonization and nodulation, have been reported. Some success was achieved by Li and Alexander (1986) They used coinoculants of antibiotic-producing Pseudomonas spp. and antibiotic- resistant Rhizobium meliloti strains and antibiotic- producing Bacillus spp. with a resistant Bradyrhizobium

japonicum strain. Four Pseudomonas spp. and one Serratia spp., coinoculated with different Bradyrhizobium japonicum strains, significantly altered nodule occupancy of soybean grown in sterile vermiculite deficient in iron (Fuhrmann and

Wollum, 1989) The Pseudomonas species used were

fluorescent and siderophore-producing strains. Other

species of Pseudomonas were effective even when iron was not

limiting. Hodgson et al. (1985) utilized bacteriocin- resistant strains of Rhizobium trifolii in coinocula with bacteriocin-producing strains of Rhizobium trifolii to alter the normal nodulation pattern of Trifolium subterraneum. 12

Other organisms have been insignificant as coinoculant strains. They include nonsymbiotic nitrogen-fixing strains of Azospirillum, Azotobacter, and Beijerinckia in various combinations with Rhizobium (Apte and Ishwaran, 1974;

Plazinski and Rolfe, 1985; Singh and Subba Rao, 1979) . Both positive and negative effects on nodulation and the quantity of nitrogen fixed were reported. All coinoculant organisms are considered to produce plant growth hormones and the observed responses are believed to be a result of changes in the sites available for nodule initiation. The coinoculation of vesicular-arbuscular mycorrhiza with

Rhizobium is another system known to enhance nitrogen fixation in legumes. This is particularly so in tropical legumes growing in soils with high phosphorus-fixing

(Ganry al Mosse, 1976). capacity et . , 1985;

The principle of coinoculation is based on the production of a chemical by the coinoculant

microorganism (s) . The compound subsequently modifies the nodule occupancy of the Rhizobium spp. Direct use of fungicide or antibiotics, either added to the soil or coated on the seed along with the Rhizobium strain (s) resistant to the compound has been tested. Lennox and Alexander (1981) obtained a greater percentage of the nodules on Phaseolus

vulgaris formed by a thiram resistant strain of Rhizobium phaseoli if seeds were treated with thiram. Similarly,

Jones and Giddens (1983) inoculated thiram treated soybean seed with a thiram resistant mutant of USDA 110. They were . .

13 successful in increasing the proportion of thiram resistant isolates in the soybean nodules. The use of streptomycin and a streptomycin-resistant Rhizohium meliloti strain led to an increase in the number of nodules on alfalfa roots in a nonsterile soil. This was due to the suppression of other bacterial cells in the rhizosphere (Li and Alexander, 1986)

Another aspect related to nodulation dominance is

'nodulation blocking' whereby a strain unable to nodulate a particular cultivar of pea completely suppressed nodulation by a superior one (Winarno and Lie, 1979) Therefore, even nonsymbiotic strains of rhizobia which may be abundant in

soil (Segovia et al., 1991) may exert a very specific effect on the competition outcome. Another mechanism demonstrated

is 'restricted nodulation' (Cregan et al., 1989).

Attempting to overcome the dominance of Bradyrhizobium

japonicum serocluster 123, Cregan et al. (1989) exploited the ability of two soybean genotypes to restrict nodulation by USDA 123 serocluster. Their results demonstrated host participation in restricting nodulation and in reducing

dominance. It seems to be a specific phenomenon which can

effectively discriminate between Bradyrhizobium japonicum

strains which are serologically related.

Trehalose

Physical Properties

Trehalose (1-0-alpha-D-glucopyranosyl-alpha-D-

glucopyranoside) naturally occurs as a-D-glucose

disaccharide, with an approximate molecular weight of 342 . . . . ,

mole'^. Its chemical formula is C 12 H 22 O 11 . Some reports have claimed detection of other isomers, alpha-beta or beta-beta trehalose (Cook et al., 1984; Parrish, 1987). Trehalose is a sweet sugar (Birch, 1979), soluble in dilute ethanol and water. It crystallizes in cold 80% ethanol as a dihydrate and has a melting point of 96-97°C (Birch, 1963) Trehalose is a nonreducing sugar and yields alpha-glucose on hydrolysis

Occurrence

Trehalose was first isolated in 1832 (Mellor, 1992) from ergot. Since then it has been discovered in a wide variety of organisms, including bacteria and cyanobacteria, red algae, liverworts, some lower vascular plants, fungi, yeasts, insects, crustaceans, nematodes and annelids

(reviewed by Elbein, 1974) The disaccharide is also found in pteridophyte lesser clubmosses, in leaves of several eusporangiate ferns and in ripening fruits of several members of the Apiaceae (Kandler and Hopf, 1980) In a few cases, where trehalose has been detected in vascular plants, it exceeds sucrose in concentration. It replaces sucrose as

the translocated sugar (Lewis, 1984; Arnold, 1968) .

Trehalose has not been reproducibly identified in

angiosperms (Gussin, 1972; Brocklebank and Henedry, 1989) .

Yet the sugar is common in plant symbioses (Lopez et al .

1983, 1984; Streeter and Salminen, 1988) .

Trehalose dimycolate is formed by the diesterification of mycolic acids found is the cell walls of Mycobacterium . . .

15 tuberculosis and other related bacteria. The dimycolate is also known as the cord factor and it is responsible for the antigenic reactions of tubercles in infected mammalian tissue. Different species of bacteria produce trehalose dimycolate of varying chain lengths. They are all capable of eliciting the same toxic response in infected mammals

(Silva et al., 1988). Other trehalose containing derivatives include glycolipids, oligosaccharides, and sulfolipids. They all have been isolated from mycobacteria and related genera (Elbein, 1974; Saadat and Ballou, 1983;

Ristau and Wagner, 1983; Tomiyasu et al . , 1986; Shimikata et al 1985). However, their function is unclear. . ,

Trehalose Biosynthesis

The pathway for trehalose synthesis was first elucidated using yeast extracts (Cabib and Leloir, 1958)

It is a two-step process. Uridine diphosphoglucose (UDP) and glucose-6-phosphate combine to yield trehalose-6- phosphate (T-6-P) and UDP. Then T-6-P is cleaved to form trehalose and inorganic phosphate (Fig. 2-1) The enzymes involved are T-6-P synthetase (EC 2.4.1.15) and T-6-P phosphatase (EC 3.1.2.12), respectively. The synthetase requires Mg^"" for activity and usually utilizes uridine diphosphoglucose. However, Paschoalin et al. (1989) reported a trehalase synthase from yeast that used adenine diphosphoglucose. This enzyme also needed Trehalose synthesis may be regulated by end product inhibition from either T-6-P or trehalose (Elbein, 1974) 16

Glucose + GIucose-6-pliosphate

Phosphotrehalase' Reversible reaction

Treha!ose-6-phosphate

Trehalose-6-phosphatase' Reversible reaction

2x Glucose ^ Mixture of non-specific saccharidases^ Trehalose

Trehalase^ •

Trehalose phosphorylase^ Reversible reaction

Glucose + Glucose-l-phosphate

fungi, animals: ^Euglena; '‘Detected in Occurrence: ‘Trehalose upuakc in bacteria; ^Plants, Frankia, a source of artifacts.

Figure 2-1. Trehalose synthesis and hydrolysis (Adapted from Mellor, 1992). .. .

Trehalose Catabolism

Trehalose can be hydrolyzed in several ways (Mellor,

1992) As a result there has been some confusion in the literature, especially with regard to prokaryotes (Crabbe et

The chief enzyme responsible for the al . , 1969). degradation of trehalose is trehalase (EC 3.21.28 alpha-

alpha-trehalose-l-D-glucohydralase) . It was first

demonstrated by Bourguelot in 1893 (Hoelzle, 1990) . On hydrolysis trehalose yields alpha-glucose and beta-glucose

molecules (Defaye et al., 1983/ Nakano et al . , 1989) .

A second enzyme, trehalose phosphorylase (EC 2. 4. 1.1), also catalyzes the degradation of trehalose. The products are alpha-glucose and beta-glucose-l-phosphate (Marechal and

Belocopitow, 1972) . Trehalose phosphorylase, though rare in nature, has been detected at low levels in Bradyrhizobium

japonicum bacteroids (Salminen and Streeter, 1986) . A similar enzyme is found in the basidiomycete fungus

Flammulina velutipes Its action on trehalose produces alpha-glucose and glucose-l-phosphate instead of alpha- glucose and beta-glucose-l-phosphate (Kitamoto et al.,

1988) This enzyme has a higher specific activity, both in mycelia and fruiting bodies, compared to trehalase activity.

Trehalose phosphorylase is capable of synthesizing trehalose

from the catabolic products.

Phosphotrehalase is another catabolic enzyme of trehalose. It breaks down trehalose to glucose and alpha-

glucose-l-phosphate . Phosphotrehalase was found in Bacillus . . .

18 popilliae (Bhumiratana et al., 1974) and in soybean

(Salminen and Streeter, 1986) Products similar to those of phosphotrehalase mediated T-6-P hydrolysis were detected in

Escherichia coli and Salmonella typhimurium extracts and were attributed to trehalose hydrolase.

Trehalase enzyme is widely distributed in many

organisms. It is found in Escherichia coli though its

function in the organism is unclear (Boos et al., 1987).

However, Escherichia coli takes up trehalose after phosphorylation via a phosphotransferase mediated transport

system. Hydrolysis follows immediately to glucose and

alpha-glucose-6-P (Boos et al., 1987; Giaever et al., 1988).

A similar situation exists in Salmonella typhimurium (Postma

et al., 1986) and Bacillus popilliae (Bhumiratana et al.,

1981) .

Rhizobia can grow on (Glenn and Dilworth, 1981) and

accumulate (Streeter, 1985) trehalose, though the mechanism

is not well understood. The accumulation of trehalose in

the bacterial cytoplasm diminishes the presence of trehalase

there (Mellor, 1988) Soluble trehalase was reported in

cultured Frankia (Lopez and Torrey, 1985) There are also

reports of trehalase in Streptomyces hygroscopicus (Hey-

Furguson et al., 1973/ Hey and Elbein, 1968) and

Streptomyces antibioticus (Brana et al . , 1986).

Several fungal genera have trehalase. These include

Aspergillus niger (Ng et al., 1974), Aspergillus oryzae

(Horikoshi and Ikeda, 1966), Aurobasidium pullulans (Catley . . ,

languginosa (Prasad and and Kelley, 1975) , Humicola

Maheshwari, 1978) , Mucor rouxii (van Laere and Siegers,

1987) and Phacomyces blakesleenus (van Laere and Hendrix,

1983) . Among the yeasts that have trehalase are

Schizosaccharomyces pombe (Inoue and Shimoda, 1981) and

Saccharomyces cerevisiae (Londesborough and Varimo, 1984) .

There are two trehalases in baker's yeast; one is found in the vacuole and is highly glycosylated. It has an acid

(4.0) pH optimum and is inhibited by acetate. The second trehalase has a neutral pH optimum and is suppressed by zinc

Many membrane associated trehalases have been reported from diverse animal sources. They include the rabbit kidney

(Sacktor, 1968), the labial glands of ants (Paulsen, 1971), bee (Lefebvre and Huber, 1970) moth (Gussin and Wyatt,

1965), cockroach (Gilby et al . , 1967), and shrimp (Hand and

Carpenter, 1986) In Euglena gracilis trahalase is absent and the catabolism of trehalose is accomplished by trehalose

phosphorylase (Marechal and Belocopitow, 1972) .

Trehalase occupies an unusual position in so far as where it is located. Its substrate is missing in higher

plants except in symbiotic organs (Mellor, 1992) . Trehalase activity has been reported in sugar cane (Glasziou and

Gayler, 1969; Alexander, 1973; Fleischmacher et al . , 1980).

Pollen also contains trehalase, which is similar to the one found in soybean nodules, with a broad pH optimum of 5 and above. Trehalase activity was also reported in a number of 20

other plant genera (Veluthambi et al., 1981) . A low activity was observed in Phaseolus primary leaves and it was practically undetectable in Cascuta.

It is noteworthy that trehalase can exist in organisms that synthesize or accumulate trehalose as well as those which do not. In the latter case trehalase is a probable defense mechanism against trehalose toxicity if the sugar were consumed or absorbed.

Functions of Trehalose

Several trehalose functions have been suggested for

different organisms. It is also possible for trehalose to participate in more than one function in one organism.

Trehalose is considered to be a storage compound in fungi.

The sugar is accumulated during the life cycle and in the

reproductive stages. It is mobilized during spore

germination (Thevelein, 1984) . Trehalose is also stored in

yeasts (Arnold, 1979) and in Streptomyces griseus (McBride

and Ensign, 1990) . However, in Phycomyces the obvious role

of trehalose as a reserve carbohydrate for germinating

spores was contradicted by several observations given by van

Laere et al. (1987) . Where trehalose occurs, it is utilized

during reduced growth, whether in response to environmental

stress or due to the developmental stage of the fungus.

Pane)c and Barnardes (1983) showed that both glycogen and

trehalose were present in Saccharomyces cerevisiae. While

trehalose is needed to trigger sporulation and the glycogen

is necessary for sporulation. McBride and Ensign (1990) , . . .

21 reported that spores of Streptomyces griseus contained both trehalose and trehalase. The hydrolysis of the sugar does not occur until spore germination is initiated (McBride and

Ensign, 1990; Inoue and Shimoda, 1981) The enzyme is separated from the substrate in the dormant stage (Inoue and

Shimoda, 1981) . Trehalase only became accessible to trehalose upon germination.

In bacteria, trehalose has similar functions to those noted for fungi. It is a major storage form of carbon

(Schmiz et al., 1985), energy source (Lopez et al., 1983)

and an agent of desiccation tolerance (Martin et al., 1986).

The disaccharide also has a structural role in some bacteria. For example, trehalose is esterified to fatty

acids, forming trehalose dimycolate. The latter is

incorporated into the cell wall of species of Mycobacterium

and Corynebacterium.

Trehalose is the major carbohydrate in the insect

hemolymph and plays an important role in carbohydrate metabolism (Wyatt, 1967) The sugar provides the main

direct energy source for flight in Phormia regina (Clegg and

Evans, 1961) , Periplanta americana (Polacek and Kubista,

1960) and in Musca domestica (Rockstein and Srivastava,

1967; Srivastava and Rockstein, 1969) Trehalose is the preferred metabolite in active life of Tenebrio molitor

(Dutrieu and Gourdoux, 1971) . The sugar also serves as an

important food reservoir (Neetles et al., 1971) and is

converted to glucose during periods of starvation. , .

Trehalose decreased (40-60%) in both sexes of aging

Callosobrunchus maculatus This implies that trehalose is an energy source for metabolism in aging bruchids (Puri and

Sharma, 1984) .

In a variety of anhydrobiotic organisms such as fungi, nematodes, brine shrimps and yeast, trehalose stabilizes the bilayer structure of the dry membrane. The sugar replaces the water of hydration surrounding the lipids in the bilayer

Crowe and Crowe, 1984). The trehalose- (Clegg et al . , 1982; lipid interaction that effects membrane stability was

studied by Lee et al. (1989) .

The ability of organisms to grow under conditions of elevated osmotic pressure attracted attention from different disciplines: physiology and biochemistry (Brown, 1976;

Jennings, 1983; Munns et al., 1983), food science (Onishi,

1963) and industrial application (Ben-Amotz and Avron,

1963). Walled organisms such as algae (Mei]cle et al., 1988) operate a system of turgor regulation while wall-less ones

show volume control (Reed et al., 1984) . Most organisms that grow under high osmotic pressure, generate the internal osmotic potential by accumulating one or more low molecular mass compounds. These may be organic or inorganic solutes.

Osmotic adjustment of intracellular organic solutes has been reported for a wide range of microorganisms. Included are bacteria (Means, 1975), cyanobacteria (Reed et al., 1984;

(Brown and Simpson, 1972; Mackay et al . , 1984), fungi

Jennings, 1983; Gadd et al., 1984). ,

All osmoresponsive solutes (osmolytes) fit the definition of a compatible solute proposed by Brown and

Simpson (1972) . These substances fall into three general classes (Yancey et al., 1982): polyols (sugars, sugar acid alcohols and glycerol) , amino acids and amino derivatives (glutamate, proline and betaines, gamma-

aminobutyric acid and taurine) , and urea and methylamines

(trimethylamine-N-oxide) . Examples of osmolytes are trehalose, sucrose, glycine, glutamate, quaternary ammonium compounds such as glycine betaine, glycerol, glucosylglycerol (0-alpha-D-glucoparanosyl- (1-2) -glycerol) and ions such as K”" and Na"^.

All species of cyanobacteria tested have the ability to

osmoregulate (Reed et al., 1984/ Stal and Reed, 1987) . Many of them accumulate trehalose as the principal osmoticum.

Other solutes that act as osmoprotectants in cyanobacteria are sucrose, glucosyl-glycerol and glycine betaine. These compounds may occur singly or in combination in osmoregulating cyanobacteria (Mackay et al., 1983). There

is a relationship between salt tolerance and the type of osmoticum accumulated. This relationship could be used to classify cyanobacteria. For example, species from soil and

fresh water tend to accumulate saccharides like trehalose.

More salt-tolerant species contain glycine betaine, and glucosyl-glycerol (Mackay et al., 1983). However, there are exceptions to these trends (Mackay et al., 1984; Reed et

Stal and Reed, 1987). al . , 1984; , ,

Teinp0ratur0 inodifi0S th0 ratio of trohaloso to gPuicosyl“glyc 0 irol in rosponso to salt strass in Spirulins

1985). At 37°C 31% of tha total platensis (Warr at al . , was trahalosa. In trahalosa plus glucosyl-glycarol , contrast, at 20°C, trahalosa accountad for only 9% of tha total osmoticum prasant. Tha highest trehalose concentration in the cited study was noted at high temperature and low salt level.

Several studies have documented various Escherichia salt coli strains with higher levels of trehalose under high

al 1986; Larsen et al . , 1987; concentrations (Strom et . ,

Munro et al . Giaever et al . , 1988; Reed et al . , 1987;

1989). Mutants lacking the enzyme necessary to synthesize

UDP-glucose were consequently unable to produce trehalose (Strom et al 1986). This and hence were osmosensitive . ,

phenomenon was successfully used to screen for other et al 1988). trehalose negative mutants (Giaevar . ,

Trehalose synthase activity increased in Escherichia coli

when assayed in vitro under high osmotic potential (Giaever

. et al . , 1988 )

Rod et al. (1988) screened over 20 strains of

Escherichia coli for increased trehalose accumulation. They time found that strains that had been in culture for a long to had mutated. The mutations rendered them unresponsive

osmotic stress since they were unable to accumulate

trehalose. Escherichia coli strain K12, which has been in

since 1922, does not accumulate trehalose and . ,

consequently it is osmosensit ive . However, some descendants of this strain have recovered this capability. Studies by

Larsen et al. (1987) and Dinnbeir et al . (1988) have reported that Escherichia coli accumulates glutamate in addition to trehalose when under osmotic pressure.

Glutamate is accumulated rapidly after the initiation of osmotic stress but is immediately replaced by trehalose over a period of one or two hours.

In rhizobia. Smith and Smith (1989) and Hoelzle (1990) observed increased trehalose accumulation in some Rhizohium strains. Smith and Smith (1989) also identified a dipeptide

(N-acetylglutaminyl-glutamine amide, NAGGN) which has osmoregulatory properties. Other researchers (Hua et al.,

1982/ Botsford, 1984) have documented the presence of glutamate in rhizobia in response to high levels of salt.

Furthermore, Breedveld et al. (1990) and Dylan (1990) have reported the accumulation of cyclic beta (1-2) glucans in

Rhizobium meliloti under hypoosmotic conditions.

Many fungi can grow in media of increased osmotic potential because they can regulate their internal solute concentration. This is achieved by increased production of

glycerol (Brown, 1978), erythritol (Gadd et al . , 1984), and arabitol (Adler and Gustaffson, 1980/ Nobre and da Costa,

1985) Other carbohydrates, including trehalose, have been documented (Edgley and Brown, 1983) especially in stationary phase cells. In addition, trehalose has been . . . , ,

26 reported in Euglena gracilis (Dwyer, 1986) and in the

mosquito Culex tarsalis (Garrett and Bradley, 1987) .

Some organisms that have cryoprotectant s can withstand temperatures below 0°C. Known agents include sorbitol, glycerol, glucose, glycogen, myo-inositol, erythritol and trehalose (Lee and Lewis, 1985; Baust and Lee, 1983) There are examples in which trehalose acts either singly or in combination with other compounds as a cryoprotectant.

Trehalose is documented in alpine beetles (Bakken, 1985) gall fly (Eurosta solidagins) larvae (Lee and Lewis, 1985)

al sunflower larvae of Xestia c-nigrum (Goto et . , 1986),

common lacewing Chrysoperla moth (Rojas et al . , 1989), and cornea (Vannier, 1986). Additionally, Niederer et al.

(1992) reported a positive correlation between trehalose content of mycorrhizas and their ability to survive frost or desiccation as determined by the electrolyte leaching method

Trehalose in Plant-microbe Interactions

Mellor (1992) contends that only trehalose tolerant plants closely associate with trehalose producing microorganisms, since the sugar is toxic to many plants.

Trehalose has been reported in most of the known symbioses:

1) In cyanobacteria, the disaccharide is considered to be

an osmoprotectant (Reed et al., 1984). There are scanty

data on the mechanism of uptake and/or utilization of

trehalose in cyanobacteria as well as in symbiotic

organs these organisms have limited ability 2) In actinomycetes ,

to metabolize a wide range of carbohydrates (Stowers et

al Consequently, metabolic al . , 1986; Tisa et . , 1983).

studies have focused on organic acids as sources of

carbon (Akkermanns et al . , 1981). However, Frankia

accumulate glycogen (Bension and Eveleigh, 1979) and

trehalose (Lopez et al . , 1983, 1984). The amount of

trehalose in Frankia is inversely correlated with

nitrogen fixation. Lopez and Torrey (1985) attributed

this to reduced trehalose accumulation during the

process of nitrogen reduction. Actinorrhizal nodules

contain sufficient amounts of sucrose and fructose to

sustain nitrogen fixation. Therefore, the role of

trehalose in Frankia was not evident.

3) In rhizobial associations, Streeter and Salminen (1988) found that trehalose and sucrose were the dominant

carbohydrates (20 mg gm""' fresh weight) in soybean

nodules. The sugar has also been reported in Phaseolus

sativum, Arachis hypogeae Medicago vulgaris , Pisum ,

sativa, Trifolium repens and Lotus corniculatus

(Streeter, 1985). The actual synthesis of trehalose

occurs in the bacteroids (Salminen and Streeter, 1986;

Streeter, 1985). However, a big proportion of trehalose

is released to the host cytoplasm (Streeter, 1987). The

infected regions of the nodules contain trehalase

(Salminen and Streeter, 1986; Streeter, 1982). The

enzyme is considered to be of host origin because of its . .

location. On the other hand Bradyrhizobium bacteroids

contain little (Salminen and Streeter, 1986) or no

(Mellor, 1988; Hoelzle and Streeter, 1990a) trehalase.

In contrast, Rhizobium and Agrobacterium possess

trehalase activity (Hoelzle and Streeter, 1990a).

Furthermore, Hoelzle and Streeter (1990b) observed

increased trehalose accumulation in nodules of Phaseolus

Yulgsr is which were exposed to 1-8 rather than 21-s

oxygen prevalent symbioses amongst 4) Mycorrhiza are the most vascular plants (Newman and Reddell, 1987). Detached

ascomycete ectomycorrhizas convert glucose into

trehalose and mannitol (Lewis and Harley, 1965; Harley leaks and Smith, 1983; Martin et al., 1988). Trehalose

from the fungus to the host (Lewis and Harley, 1965; also 1989). Schubert et al . (1992) Niederer et al . ,

reported that trehalose was the main soluble

carbohydrate in sporocarps of vesicular-arbuscular

mycorrhiza {Glomus versiforme) and in roots of Tagetes mosseae tenui folia and Glycine max infected with Glomus amounts or The disaccharide was either present in trace

undetectable in nonmycorrhizal roots. The trehalose

content increased proportionately with the degree of mycorrhization but was depressed by phosphorus

fertilization or light deprivation. Keen and Williams (1969) 5) In parasitic interactions. observed trehalose in tissues of Brass ica oleracaea . . .

infected with Plasmodiophora brassicae A similar

occurrence was noted on the leaves of Senecio squalidus

infected by Albugo tragoponis

In spite of trehalose having been documented in plant-

microbe interactions, no definite role has been attributed

to the sugar. Trehalose is toxic to many plants (Mellor,

1992) and there are early reports of microbial toxins such

as rhizobitoxin . Relating rhizobitoxin to trehalose, Mellor

(1992) speculated that the disaccharide could be an

incidental compound. Its major role is to prevent stable

interactions with susceptible (nontrehalase producing)

hosts. After integrating information about the functions of

trehalose in nonsymbiotic systems and its symbiotic status,

Mellor (1992) concluded that there was no )cnown active role

of the sugar in symbioses. Trehalose may be investigated as

a possible protectant of membranes against proteins, or in

general biological structures, against water stress.

The Common Bean {Phaseolus vulgaris L.)

The common bean (Phaseolus vulgaris L.) is a frequently

cultivated food legume in most agricultural areas of the

world, like Central and South America, United Kingdom

Graham and Rosas, 1977) and in East (Taylor et al . , 1983;

Africa (Abebe, 1987; Taylor et al., 1983; Sengooba, 1987)

It is a major source of dietary protein in the developing

countries. Bean production in the latter countries is

carried out by peasant farmers on small holdings. In most . ,

30 cases they are unable to afford expensive inputs such as

. fertilizers (Aguirre and Miranda, 1973/ Kileo et al . , 1987)

Yields in some of these areas have been lower than the

potential yields (Graham and Rosas, 1977) .

The ability of field beans (Phaseolus vulgaris L.) to support Rhizohium phaseoli and subsequently benefit from symbiotically fixed nitrogen has been defined as the nis

gene (nitrogen fixation supportive trait) by Rennie (1981) .

Bean response to Rhizohium inoculation is variable. For example, Graham and Rosas (1977) quote reports ranging from those of decreased yields in the case where nitrogen was not limiting (Pessanha et al., 1970), to substantial increases

(Pillay and Mamet, 1972; Habish and Ishanga, 1974) . Failure of applied inoculum to increase seed yield has been

attributed to many causes and interactions (Graham, 1981) .

Some of the factors are pH, nutritional and water status, competition or dominance from the indigenous rhizobia and

host cultivar differences (Graham, 1981) . Consequently, the effect of different Rhizohium phaseoli strains on nitrogen fixation and on yield of cultivars has received attention

extremes (Graham et al only in relation to environmental .

1982) To correct this anomaly, CIAT (Centro Internacional de Agriculture Tropical, Cali, Colombia) coordinated an international bean inoculation trial (IBIT) in seven South

American countries, UK, USA, and Canada, between 1979-1981.

Nineteen strains of Rhizohium phaseoli and two bean cultivars (Aurora and Kentwood) were evaluated under 31 phytotron conditions prior to field experimentation. The

isotope dilution method is preferable to the Acetylene

Reduction Assay (ARA) since Rennie and Kemp (1981, 1983) found no reliable and consistent relationship between acetylene reducing activity and actual nitrogen fixed under either phytotron or field conditions. They attributed some of the negative results to the use of ARA, whereas reports using method indicate that the bean crop can supply 50% of its nitrogen requirements from nitrogen fixation

(Westermann et al., 1981). CHAPTER III MATERIALS AND METHODS

Experiment la: Accumulation of Trehalose bv Rhizobium leQuminosarum Biovar nbasGoli. Strain CF.^ and its TnS Derivative Mutants Cultured Under Varying Conditions

Rhizohium leguminosarum biovar phaseoli strain CEB and its TnS derivative mutants were obtained as agar cultures on trypsin medium from Jocylen Milner (University of Wisconsin,

Madison). Strain CEB is resistant to 200 pg ml'^ of streptomycin and the mutants are resistant to both of 200 ml _/| j- 0 p-ji^omycin and kanamycin at a concentration pg of each antibiotic. The strains were verified as rhizobia by routine laboratory and nodulation tests (Vincent, 1970;

Somasegaran and Hoben, 1985).

Screening for Trehalose Negative Mutants

Plates of yeast extract mannitol agar (YEMA), (Appendix (Sigma A) containing 200 pg ml"'' streptomycin sulphate

Chemical Co. St. Louis, MO) and varying concentrations of equivalent to 0.400, NaCl 0, 0.05, 0.1 and 0.2 M excess

2.925, 5.850, 11.700, 22.400, 25.100, 46.800, g L"^ respectively, were streaked with cells of strain CEB in duplicate for each salt level. Twenty four CE2::Tn5 mutants were grown on YEMA containing NaCl at a concentration of isolate an osmosensitive 11.700 g L~^ excess in order to tre) negative. strain which would likely be trehalose (

22 .

Mutant strains of varying salt tolerance were obtained by growing the mutant cells on YEMA of different NaCl concentrations; a) 0.400, b) 29.250 and c) 58.500 g

Quantification of Trehalose and Other Saccharides in Bacterial Cells Grown Under Different Salt Concentration

The parent strain, CE3, and the derivative mutants,

CE3::Tn5 12b (able to grow on YEMA containing 11.700 gm NaCl

L”'' but growth limited by 23.400 g NaCl L^), CE3::Tn5 18a

(able to grow on 29.250 gm NaCl L"'' but not on a medium

L''' and CE3::Tn5 14b and 16b (able containing 58.500 g NaCl )

L”'' were selected for further to grow on 58.500 g NaCl )

study. The bacterial cells were cultured for 96 h,

harvested, processed and analyzed for trehalose and other

saccharides following the methods described by Streeter

(1985) with the following modifications:

1) The defined medium (Manhart and Wong, 1979) contained

sodium chloride at levels of 0, 0.05, 0.1 and 0.2 M

NaCl

2) Centrifugation was at 15,000 x g.

3) The extraction of saccharides was carried out at 4°C,

and the supernatant was stored at 4°C .

4) A Shimadzu CR 601 chromatograph, fitted with a J and W Scientific column, 30 m long, with an internal diameter

(ID) of 30 mm, packed with DB5 to a thickness of 0.25

uM, was used for carbohydrate analysis.

program for the analysis was as follows; 5 ) The temperature

oven and injection port temperatures were both 250°C. .

5) The temperature program for the analysis was as follows;

oven and injection port temperatures were both 250°C.

The initial temperature was 180°C; hold time was 3 min,

at a rate of temperature increase of 10°C min'^, to a

final temperature of 280°C, followed by a hold time of 1

min, giving a total of 14 min per run.

6) A few samples were also analyzed by gas chromatography/

mass spectrometry (GS/MS) using a Varian-Saturn II 3400

gas chromatograph/mass spectrometer (GS/MS) (Appendix

B) .

Determination of trehalose in cells grown under pH, temperature, and antibiotic stress conditions was performed as described above with the following modifications:

1) The pH values for the defined medium were 5.0, 6.0, 7.0

and 8.0.

2) Incubation temperatures were 10, 20, 30, and 40°C.

3) Sterile chloramphenicol at the rate of 6.5 jig ml"^ final

concentration was incorporated into the medium prior to

inoculation

These experiments were performed twice.

Analysis of Nodules for Trehalose Content

Seeds of Phaseolus vulgaris L. cultivar Midnight (Idaho

Bean Seed Co., Twin Falls, ID) were surface sterilized by soalcing in 95% ethanol for one min, followed by two min in

3% hydrogen peroxide. This was followed by eight rinses in sterile water. The seeds were imbibed in sterile water for . . .

10-12 h, then transferred aseptically to 2% water agar plates. The plates containing the seeds were wrapped in aluminum foil and incubated for 36 h for the seed to germinate. Healthy seedlings with straight radicles

(approximately 2-cm long) were selected for planting in sterile plastic growth pouches. Each pouch was initially supplied with 10 ml of half-strength Jensen's medium

(Vincent, 1970) and sterilized by autoclaving at 120°C at a pressure of 15 kg cm^ for 20 min as recommended by the suppliers (Vaughan's Seed Co., Downers Grove, IL) One seedling was planted per pouch. The next day, 2 ml of two day-old YEM broth cultures of strains; CE3 (4.1 XIO® cfu ml"

^) 10® : : , CE3::Tn5 12b (4.2 X cfu ml"^) , CE3 Tn5 14b, (4.4 X

10® cfu ml'^) , CE3;:Tn5 16b (4.3 XIO® cfu ml"^) and CE3::Tn5

18a (4.1 X 10® cfu ml"’-) were used to inoculate 15 seedlings per treatment. Uninoculated controls with and without added nitrogen were included in one of the trials. Nitrogen was supplied as KNO3 (analytical grade) at the rate of 0.05 mg plant"’ as suggested by Vincent (1970) Plants were supplied with full strength sterile Jensen's medium once a week and sterile deionized water as required. The treatments were arranged in a Randomized Complete Block Design, replicated three times. The plants were maintained in a growth chamber (Model CEL 37-14, Warren Sherer, Marshall,

MI) The growth conditions were photoperiod, 16 h, temperature, 29°C and light intensity, 230 jimol m^ .

36 s"^. The dark period was 8 h and temperature 23°C. Plants

were harvested 18 d after planting (DAP) . Preliminary trials indicated that longer periods led to nodule degeneration.

Nodules were detached from plants of the same treatment and combined. They were processed and analyzed for

carbohydrates following the method of Streeter (1980) . The crude extracts were not separated into acidic, basic or neutral fractions. Streeter and Bostler (1976) found no significant difference in the gas liquid chromatographic analysis of crude extracts and those fractionated using ion exchange resins.

A subsequent experiment in the same growth chamber was conducted with two strains, CE3, and CE3::Tn5 14b, in which

NaCl at a concentration of 2.925 g L“^ was applied (10 ml planfM once a week to half the treatments. At 18 DAP the nodules were harvested, and the fresh weight immediately determined. The nodules were then ground in 20 ml of 95% ethanol and centrifuged under the same conditions mentioned earlier. The supernatant was collected and the residue resuspended in 80% ethanol; the process was repeated four times and the supernatants pooled. It was dried under a blower in the hood. The dry residue was dissolved in 3 ml of sterile deionized water, a few drops of chloroform (99% pure) were added. The solution was stored frozen until chromatographic analysis was performed (Streeter, 1980)

Twelve nodules from each treatment were dissected and . .

37 processed for observation by both light and electron microscopy

Fixation

The nodules were fixed for 2 h in 2.5% glutaraldehyde in 0.05 M potassium phosphate buffer, pH 6.8, washed and postfixed for 1.5 h in 1% osmium tetroxide in the same buffer. Then the nodules were dehydrated up to 95% ethanol, embedded in LR white resin (Bio-Rad, Cambridge, MA) and polymerized overnight at 60°C.

Sectioning

Sections, approximately 60 nm and 0.5-1 |im, of whole nodules were cut with a diamond knife on a LKB ultratome III

(Leica, Inc., Deerfield, IL) .

Light Microscopy

Thick sections were heat fixed on glass slides and stained with 0.05 toluidine blue in 1% sodium borate for 5 minutes for observation of nodule structure. The presence of poly-i5-hydroxybutyrate (PHB) granules was detected by staining for 15 min with 0.3% Sudan black B in ethylene glycol followed by poststaining with 0.5% aqueous safranin 0.

Electron Microscopy

Thin sections were picked up on Formvar-coated nickel grids and stained for 15 min with 2% aqueous uranyl acetate and for 5 min with 0.5% aqueous lead acetate, pH 12 (Fahmy,

1967) The sections were observed with a Zeiss transmission . .

38 electron microscope (Carl Zeiss, Inc., Thornwood, NY) at

60kV.

Dry Matter Yield and Nitrogen Content of Bean Plants

After removing the nodules the remainder (roots and shoot) were weighed fresh and later oven dried at 65°C for 3 days. The dry weight was then determined. The plant samples were ground to a fine powder for total nitrogen determination

A small sample (0.2000 g) was placed into a digestion tube, mixed with a sodium catalyst and 3 ml of concentrated sulfuric acid were added. The mixture was vortexed, after which the tubes were placed on a heating block at 360°C in a fume hood for several hours. After cooling, deionized water was added to a volume of 50 ml. Total nitrogen was determined using a rapid flow analyzer (RFA)

Experiment lb: Growth of CE3 and Tn5 Mutants in YEMB of Varying Salt Concentrations

Broth inocula of CE3 and of Tn5 mutants were prepared as described before. Aliquots of 2 ml broth culture (two day old) were used to inoculate 150 ml YEMB plus the appropriate antibiotics, contained in separate 300 ml-flasks with side arms. The media were modified by adding different amounts of sodium chloride as follows: 0.400, 2.925, 5.850 and 11.700 g L”^, giving broth of 0, 0.05, 0.1, and 0.2 M

excess NaCl . The growth of the cells was monitored by determining absorbance periodically using a spectrophotometer (Spectronic 20, Bausch and Lomb, 39

Rochester, NY) set at 600 nm. Growth curves were generated

for visual qualitative comparisons.

Experiment 2a: Growth of Rhizobium leQuminosarum Biovar phaseoli Strain CE3 and Tn5 Derivative Mutants on Trehalose Medium

The strains were inoculated in yeast extract trehalose broth (YETB) and also streaked on yeast extract trehalose

agar (YETA) . Similarly, the strains were grown on mannitol

agar and in broth for comparison. Growth in broth was monitored by determining absorbance as described before.

The different strains were streaked on YEMA and YETA and

growth observed qualitatively.

Experiment 2b: Growth of CE3 and Tn5 Derivative Mutants in Yeast Extract Mannitol Broth of Varying pH

Inocula of CE3 and Tn5 mutants were prepared as previously described. Two ml aliquots of broth from CE3 or

from each of the five mutants were used to inoculate 150 ml

of YEMB. The media were contained in separate 300 ml-flasks with side arms. The broth was adjusted to different pH values of 5, 6, 7, and 8. Growth was monitored by recording

absorbance at varying time intervals using the same

spectrophotometer, set at the same conditions as before.

Experiment 3: Nodulation Dominance Trials

Seeds of Phaseolus vulgaris L. bean cultivar Midnight were pregerminated as previously described. Two seedlings were planted per pouch (CYG Seed growth pouch. Mega International

of Minneapolis, MN) . Twenty plants (10 pouches) were used 40 the study, inoculated as single strains or in mixtures

CE3:CE3::Tn5 14b (1:1), CE3::14b (2:1), and CE3::Tn5 14b

(1:2). The broth cultures of the two strains were grown separately in 250 ml flasks and were used as inocula when absorbance reached 1.0 at 600 nm wavelength (late log phase cells). The experiment was run as a Randomized Complete

Block Design replicated three times. Half of the treatments were supplied with 10 ml Jensen' medium containing sodium chloride at a rate of 0.05 M NaCl, two days after inoculation. The treatment was repeated after a week. All plants were watered with Jensen's medium once a week and sterile deionized water as required. The experiment was run in the growth chamber, as previously described, and terminated after three weeks.

At the end of the growth period, nodules from each treatment (at least 50) were harvested and surface sterilized as described for bean seeds. Each nodule from a treatment was then placed in a well on a sterile "microtest

III ™ " tissue culture plate (Becton and Dickinson Co.

Lincoln Park, NJ). A drop of sterile deionized water was

added to each nodule within the well, crushed and the

exudate thoroughly mixed. A multi~inoculator device Immusine Laboratories, Inc., Long Island, NY), ( Clonemaster ,

after sterilization by flaming, was dipped in the wells such

that one "tooth" dipped in only one well. The device was

then used to inoculate YEMA plates; one containing 200 ug

of streptomycin and the other containing both : .

41 streptomycin and kanamycin at 200 ug ml"^ of each. The plates were incubated at 28°C and, after three days, the number of plates showing growth was recorded for each treatment. The percent nodule occupancy of the strain was calculated as follows:

Percent No. nodules formed by mixed inoculum Nodule = ^— ^ Occupancy No. nodules formed by pure inoculum

The remaining plants (roots and shoots) were oven dried

under the same conditions mentioned before. The dry weights

obtained were used to compute the competitive ability of the

individual strains as described by Amarger (1981) as

follows

ym Ca = X 100 yp where ca = competitive ability ym = dry matter yield of mixed strains yp = dry matter yield of pure inoculum

Statistical Analysis

Statistical analyses were performed using SAS software

(SAS Institute, 1988). The ANOVA tables are given in the

Appendix C. Unless otherwise stated, Duncan's Multiple

Range Test was used to separate means of significant

treatments .

CHAPTER IV RESULTS

Characterization of Strains Used in the Study

Microscopic examination showed that strain CE3 and the

Tn5 mutants were Gram negative rods. They were confirmed

rhizobia because they formed numerous small nodules on

Phaseolus vulgaris cultivar Midnight, the host plant.

When grown on YEMA containing bromothymol blue, the medium turned from green to yellow, indicating that they were fast- growing rhizobia.

The growth of Strain CE3 was limited by 0.2 M sodium

chloride. No osmosensitive mutant could be detected, thus

screening for a tre negative mutant was presumed unsuccessful. However, five strains were selected for the

study based on their tolerance to varying salt

concentrations. Strain CE3 and CE3::Tn512b were able to grow on 0.1 M but growth was limited by 0.2 M excess NaCl,

CE3::Tn514b and CE3::Tn516b were able to grow at 1.0 M excess NaCl, and CE3:: TnSlSa could only grow at 0.5 M excess

NaCl

Accumulation of Trehalose by Rhizobium leQuminosarum Biovar phaseoli Grown Under Different Levels of Salt Concentration

Rhizobium leguminosarum biovar phaseoli strain CE3 and

its derivative Tn5 mutants (CE3 : : Tn512b, CE3::Tn514b,

CE3::Tn516b, CE3::Tn518a) accumulated trehalose when

42 43 cultured in defined medium. The amount of trehalose accumulated decreased to undetectable levels at the higher salt concentrations (Table 4-1). Analysis of variance

(ANOVA) revealed significant differences between strain and salt treatments. In addition, the interaction between salt and strain was significant (Appendix C-1). Strain CE3 accumulated more trehalose than the other strains (Table 4-

. 2 )

Arabinose was detected in all the strains used and at all levels of salt. Significant differences between strains were detected by the F test (Appendix C-2).

Effect of Sodium Chloride on the Growth of Rhizobium leauminosarum Biovar phaseoli Strain CE3 and its Tn5 Mutants

When sodium chloride was added to yeast extract mannitol broth (YEMB) the growth of strain CE3 and its derivative Tn5 mutants was unaffected at 0 and 0.05 M NaCl excess. The media had an osmolarity of approximately 33 and

97 mMOs kg”\ respectively. At 0.1 and 0.2 M sodium chloride (192 and 352 mMOs kg"\ respectively) growth was

limited (Fig. 4-1). However, strain CE3::Tn514b grew fairly well under high salt conditions.

Carbohydrate Content in Nodules

Analysis of the crude nodule extract showed the presence of carbohydrates, mainly disaccharides, and monosaccharides. The most abundant sugar was sucrose.

Other sugars included trehalose, glucose and fructose.

Statistical analysis (ANOVA) showed no significant

differences in the amount of saccharides accumulated, among .

Absorbance

Time (h) Time (h)

Figure 4-1. Growth of CE3 and Tn5 derivative mutants in increasing levels of NaCl over time. A-CE3,

B=CE3 : :Tn512b, C=CE3 : : Tn5 14b and D==CE3 : : Tn5 16b .

Table 4-1. Trehalose content of Rhizobium leguminosarum biovar phaseoli strain CE3 and its Tn5 derivative mutants grown under increasing salt concentrations

Salt concentration (NaCl M)

0 0.05 0.1 0.2

strain |j.g trehalose /mg dry weight cells

CE3 31 . la 8.8a 3.0a 0.0a

CE3: :Tn512b 4 . 4c 0 . Oc 1.6b 0.2a

CE3 : : Tn514b 3.7c 3.9b 0 . Ob 0 . Oa

CE3 : : Tn516b 4.1c 0.8c 0 . Ob 0 . Oa

CE3 : : Tn518a 9.7b 4.8b 0.0b 0 . Oa

Means within a column followed by the same letter are not significantly different, P=0.05, (Duncan's Multiple Range

Test) .

Table 4-2. Accumulation of trehalose and arabinose by Rhizobium leguminosarum biovar phaseoli strain CE3 and its derivative Tn5 mutants grown in basal medium.

Trehalose Arabinose strain \ig/mg dry cell

CE3 11 . Oa 1.2b

CE3 : : Tn512b 1 . 5c 4 . 3a

CE3 : : Tn514b 1.9c 2 . 6ab

CE3 : : Tn516b 3.6b 1.1b CE3: :Tn518a 3.6b 1.9b

Means within a column followed by the same letter are not significantly different, P=0.05 (Duncan's Multiple Range

Test) . 46

Table 4-3. Arabinose content in Rhizohium leguminosarum biovar phaseoli strain CE3 and its Tn5 mutants grown under increasing salt concentrations.

Salt concentration (NaCl M)

0 0.05 0.1 0.2

strain Jig arabinose /mg dry cells

CE3 1 . 9a 1 . Ob 1.0b 0.9b

CE3: :Tn512b 0 . 6b 1.6b 4 . la 10.9a

CE3: :Tn514b 2.7a 2.1a 3 . lab 2.5b

CE3: :Tn516b 1.7a 0 . 6c 1.3b 0.7b

CE3: :Tn518a 1 . 8a 1 . 9a 1 . 6ab 2.2b

Means within a column followed by the same letter are not significantly different, P=0.05 (Duncan's Multiple Range

Test) . i H^ 1

biovar 1 in ro ro 0 CO • DAP. 0 cr\ r- • . • 0 • fa CO 0 18 c/3 in CO in rH

CD 0 0 m 00 rH CO leguiuinosdrum plants u C • • • • • 3 0 0 (N rH 0 T— cn 0 e 'S’ LD N*

bean LT) Q) • CM r- 0 • cn • • tH 0 • ro CM on fa — T! c/3 1 1 r- 0 Rhizobiutn 0 0 C

Cfi 0 O' CO r- r- ro mutants u C • • • • * \ 0 cn 00 rH in O' [Ti CT\ 0 CM by 0 0 0. g CN CN rH ro ro

formed CO 0 • • 1— T— CO • • derivative ttj . ro rH fa LT) ro CN 0 !/3 CO (N ro rH (N 4-) CD w 4J nodules 0 W rH 0 ro ro • • • its (-1 fd r- x: 0 • • CO rH ro VD fa (D 0 ro (N fO (N rH i-i 0 rH 00 0 0 of rH rH rH and Eh E

CE3 r- 1 ( ID VD ro content CQ tH rH r' CD i fa • • • • • — rH rH c/3 1 0 ° 0 !

Carbohy

X3 X 0 X! 0 CO

. rH rH rH LD rH in in 4 c c — c c 4 c Eh Eh •H 0 •• ro ro ro ^]^0 ro fO +J fa fa fa fa fa W D u u u u

biovar

vulgaris SE 25.5 15.0 36.4 13.6 4.5 ± ± ± ± ±

Fructose

201.9 338.4 269.7 228.7 222.2

mean

leguminosarum Phaseolus

on 6.0 2.5 3.9 3.9 SE 24.5

± ± ± ± ±

Rhizobium mutants

239.9a 232.4a 145.6b 199.3a 138.6b

nodules)

Glucose

mean

by Tn5

(|lg/g

SE formed 51.7 30.9 43.8 37.4 ±28.4

derivative ± ± ± ±

Sugars

Trehalose

nodules 108.0 120.7 102.7 96.3 98.3 mean its

DAP. of r- i— 00 OO and w 0) w 00 rH CM rH • • 18 rH CO • • • o o t— I 1 rH nP CE3 o content Q) C +1 +1 +1 +1 +1 (0 O CTi

(— (Ts rH Midnight M \ c CM O' strain O O' m rH O' CM rH rH p e CD • • • • • e kD LT) r- CO O'

Carbohydrate

phaseoli

cultivar

XI n b m (N 00 rH rH rH rH

4-5. to IT) uo lO c c c G c H E-i Eh •H 03 M 00 oo 00 00 00 Table 4-1 w w W W w to o O u U u .. .

49 the strains used. In one trial a significant difference was observed in the glucose content (Tables 4-4 and 4-5)

Sucrose content in nodules formed by strain CE3 decreased when plants were supplied with 20 ml per pouch of

Jensen's medium containing sodium chloride at 0.05 M (97

mMOs kg"^) . However, a similar treatment had little effect on strain CE3::Tn514b, although the sucrose content was much lower than in CE3 whether salt was added or not. Trehalose was detected in nodules of plants either receiving or not receiving sodium chloride. The differences observed between treatments were not significant. Nonetheless, in both strains an increase in trehalose content was observed.

Strain CE3 was more responsive than the mutant (Table 4-6)

Cytological differences in nodules from salt and nonsalt treated plants were revealed by differential staining techniques and observations under both light and electron microscopes. Nodules formed by strain CE3 and

CE3::Tn514b under non-saline conditions had a large bacteroid area which stained deeply with toluidine blue as observed under the light microscope. On the contrary, nodules from salt- stressed plants of the same strain had a greatly reduced bacteroid area of (Fig. 4-2 and 4-3) When nodule sections of the same strain were stained with Sudan black B, it was observed that unstressed nodules had extensive black stained area suggesting the presence of poly-f)-hydroxybutyrate granules. Salt-stressed nodules had very little stained area which indicated less PHB. .

23.8 88.1

0.05 ± ± by derivative different nodules

791.3 457.7 (Fisher's

mean formed

its

|ig/g

plants and SE statistically M) 22.3 128.9

trehalose P=0.05 CE3 (NaCl O ± ±

stressed

strain 517.7 361.3 not

mean different,

are salt concentration CN

phaseoli SE 0.23

of O letter 05

. ± +1 Sugars 0 statistically nodules) to salt 2.16a nodules CN same biovar LO

mean CN CM the in

(mg/g are 00 SE o 0.29 2 by

leguminosarum o content LSD).

sucrose ± and o +1 1 followed iH 2.23b CT»

protected mean IT) mutants. rH

Carbohydrate column

Rhizobium superscripts

Tn5 a

b LSD) (Fisher's

rH LO within 4-6. c with c H •H • • protected • • L OO n Table 4-) W w Means Means CO o u

P=0.05 . . .

Electron photomicrographs (Fig. 4-4 and 4-5) show PHB

granules of varying sizes within the bacteroids. Strain

CE3::Tn514b had more granules than strain CE3. The effect

of salt was clearly shown by the reduced bacteroid area and

PHB granules. In addition, small black volutin granules

were more prevalent in the nodules from nonsalinized plants

than salinized ones. Based on previous electron photomicrographs the dark stained material on the periphery

of the bacteroids was interpreted to indicate the presence

of leghemoglobin . Again, this was much reduced in the salt-

stressed nodules.

Dry Matter Yield and Nitrogen Content of Bean Plants (Phaseolus vulgaris Cultivar Midnight) Inoculated with Rhizohium leQuminosarum Biovar phaseoli Strain CE3 and Tn5 Derivative Mutants 18 DAP

Dry matter yield (shoot and roots) was unaffected by

inoculation of bean seedlings by strain CE3 or the mutants

(Table 4.7 and 4.8) The nitrogen content of root and

shoot tissue was similar in both the nitrogen control and

inoculated plants. Results obtained in a separate trial,

when plants were salt stressed, did not reveal differences between the treatments. However, there was a reduction in

dry matter yield of plants inoculated with strain CE3 when

grown under saline conditions (Tables 4-7 and 4-8)

Similarly, no differences were observed in the nitrogen

content of these plants (Table 4-9) Figure 4-2. Light photomicrographs of bacteroid tissue in strain CE3,

A and B; nodule sections from non-salt treated plants stained with toluidine blue; the dark areas show the bacteroid area.

C nodule section shows black areas which represent poly-B-hydroxybutyrate granules.

D, E, and F. represent corresponding situations in nodule sections from salt-treated plants. 53 .

Figure 4-3, Light photomicrographs of bacteroid tissue of strain

CE3 : :Tn514b.

A and B nodule sections from nonsalinized plants stained with toluidine blue. Black color shows the bacteroid areas

C nodule sections stained with sudan black and the dark areas show the extent of poly-B-hydroxybutyrate granules.

D, E, and F represent corresponding situations in nodule sections from salinized plants. 55 Figure 4-4. Electron micrographs of nodule sections from strain CE3.

A shows the abundance of PHB granules which appear white. V= volutin granules appearing as tiny black granules. The section was from non-salt treated plants.

B shows reduced amount of PHB granules but volutin granules, though reduced, are more evident. The section was from salt treated plants. 57 nodule Figure 4-5 Electron micrographs of sections from strain

CE3 : :Tn514b.

A Nodule section from non- salt treated plants shows abundant PHB granules and large dark area of volutin granules (v).

B Correponding situations in sections from salt treated plants; PHB and volutin granules are greatly reduced. 59 60

Table 4-7. Dry matter yield of plants inoculated with Rhizobium leguminosarum biovar phaseoli strain CE3 and its Tn5 mutants 18 DAP.

Dry matter yi eld (mg/plant) First trial Second trial

strain mean SE mean SE + 331.33 + 18 .02 CE3 400 . 00 11.59 + + 335 . 00 6 . 02 CE3; :Tn512b 383 . 67 4.38 + 424 .00 ± 7 . 73 320 . 00 16.84 CE3 : :Tn514b + 435 .67 ± 21.49 335 . 00 10.24 CE3 : :Tn516b + 22.93 398 . 67 ± 19 . 94 317.67 CE3 : :Tn518a

Table 4-8. Nitrogen content of plants inoculated with Rhizobium leguminosarum biovar phaseoli strain CE3 and its derivative Tn5 mutants 18 DAP.

Nitrogen content (mg N/ plant) First trial Second trial

strain mean SE mean SE 1.17 CE3 13.77 ±0.90 9.71 ± 12.29 ±0.19 12.60 ± 1.57 CE3 : :Tn512b 16.29 ±0.88 11.86 ±0.18 CE3 : :Tn514b 14.14 ± 1.01 10.12 ±0.59 CE3 : :Tn516b 14.67 ±0.39 10.33 ±0.40 CE3 : ;Tnl8a ±0.27 N(-) ND ND 10.20 ND ND 11.18 ± 0.38 N( + )

ND=not determined. .

Table 4-9. Nitrogen content of plants inoculated with Rhizobium leguminosarum biovar phaseoli strain CE3 and its derivative Tn5 mutants grown under salt stress conditions.

Salt concentration (NaCl, M)

0 0.05 nitrogen content (mg N /plant) strain mean SE mean SE

CE3 11.64 ± 1.09 10.08 ±0.55 CE3: :Tn514b 11.19 ±0.66 9.76 ±0.51

Growth of Rhizobium leguminosarum Biovar phaseoli Strain CE3 and its Tn5 Mutants on Trehalose and Mannitol

Strain CE3 and its Tn5 mutants grew on solid yeast

extract trehalose agar (YETA) . The growth on YETA and YEMA plates and in YET and YEM broth (Fig. 4-6) was comparable.

Effect of pH on the Accumulation of Trehalose by Rhizobium lecruminosarum Biovar yhaseoli Strain CE3 and its Derivative Mutants

Trehalose was not detected in strain CE3 and its derivative mutants at pH values of pH 5, 6, 7 and 8.

However, arabinose was detected as shown in Table 4-10.

Analysis of variance (Appendix C-3) showed significant differences between strains and pH and their interaction effects (Table 4-10) At all pH values, strain CE3 accumulated more arabinose than the others. There were no differences among the mutants. 1 1 1i —t III

P phaseoli C

conditions. H lO CN IT) t— 00 0) c/5 • • • • • M o o O O o 0) M-l +1 +1 +1 +1 +1 4-1 CO •H pH a b biovar >1

t— G m b b b b p (t3 r~t oo CO o O' G • • • • • o •H under CO o o o o W ^>1 u r- +1 +1 +1 +1 +1 P b X o a G tn g mutants O G (C3 b b b b P cO 00 a\ 00 CT^ b> (0 G 0) • • • • • e f— rH t“H 00 Rhizobium P o -P -p Tn5 G -p (D \ 0) 4J w CN r—i — 00 • • r— c CO • • • o o o o o o by o o +1 6 +1 +1 +1 +1 m derivative 0) K w CO d o 0) — G (0 Arabinose b -P •H CTV G b b b b 4-) W • r- 00 b (CJ o cd O • • • • » p g t-H CN CN 1 oo its b < 0) b cn of (D G and W CN CO O o CO • • • • • o p:; O o o o o I— rH

Accumulation d G -P e -H G G strain G to b b b b pH H S fC5 00 CN o o O < • • • • o w E CO CN 00 00 00 nJ G

b b b fd G O CN 00 H G rH — rH pH b G 4-10. LO ID LO LO -P Q G G G G •H G E-t Eh EH Eh 2: •H IT) • • (fl W O p 00 00 00 00 OO G • (0 o Table p W w W W W CO CJ o U u U

of pH 5 6, 7, When the cells were grown in YEM broth , growth (Fig. 4-7). At pH and 8, there were differences in

(24 h) lag phase, after which 5, the cells exhibited a long the cells multiplied rapidly and then entered the decline phase was shorter but phase. At pH 6 (Fig. 4-10), the lag there were strain differences in growth response. Growth was monitored by changes in optical density of the broth.

Growth curves of both strain CE3 and CE3::Tn514b showed multiphase growth cycles. A similar observation was also CE3 and noted at pH 7, while at pH 8 only strains

CE3:tTn514b were able to grow after lag phase (Fig. 4—7).

Effect of Chloramphenicol on the Accumulat ion of Trehalose haseol i strain CE3 and in Rhizobium 1 Ran minosarum Biovar p its Tn5 Mutants

When chloramphenicol was added to the medium at the

ml”'' no trehalose was detected. However, rate of 6.25 /ug ,

arabinose was present in small quantities (Table 4-11). Temperature Effects on Rhizobium leauminosarum—Biova r phasBoli Strain CE3 and CE3: .Tn5 14b and Strain CE3 and CE3::Tn5 grown in the defined medium

incubated under different temperatures (10, 20, 30, and of trehalose. 40 °C) did not accumulate detectable amounts

However, arabinose was present except at the highest

temperature. At 40 °C, cell growth was curtailed as dry indicated by zero value for arabinose and reduced cell strains. weight. The amount of arabinose decreased in both greater amounts 3-t-j-ain CE3::Tn514b accumulated significantly The than strain CE3, 2.643 versus 1.509 pg mg”'' dry cells. 4

64 computed on three temperatures (10, 20 and 30°C) (Appendix

C-4) revealed a significant strain effect but not a temperature effect. The interaction between strain and temperature was also significant. The relationship existing between temperature and arabinose content was determined using the contrast procedure. It showed a significant linear relationship. When a regression analysis was performed on means for each of the strains, a highly significant negative correlation between temperature and arabinose for CE3;: Tn5 was obtained (R^=0.99) while a weak one of (R^=0.51) was obtained for strain CE3. The results

are summarized (Figure 4-8) .

Table 4-11. Effect of chloramphenicol on trehalose and arabinose content in Rhizobium leguminosarum biovar phaseoli strain CE3 and its mutants.

Arabinose Trehalose

[i.g/ mg dry cells strain mean SE mean SE

CE3 1.9 ± 0.3 ND NA

CE3: :Tn512b 2.3 ± 0.3 ND NA

CE3: : Tn514b 2.2 ± 0.2 ND NA

CE3: :Tn516b 1.8 ± 0.2 ND NA

CE3: : Tn518a 2 . ± 0.0 ND NA

ND= not detected NA= not applicable. . ,

Absorbance

Time (h) Time (h)

Figure 4-6. Growth of CE3 and Tn5 derivative mutants in YETB and YEMB over time. A=CE3, B=CE3 : : Tnl4b

C=CE3 : :Tn516b and D=CE3 : : Tb5 18a .

Absorbance

Time(h)

Figure 4-7. Growth of CE3 and Tn5 derivative mutants in YEMB of varying pH values over time. A=pH 5, 7 8 B^pH 6 , C=pH and D=pH 67

(/-^g/mg)

ARABINOSE

2.3-j

' 12 ' 16 20 24 28 32 TEMPERATURE °C

accumulation of Figure 4-8. Effect of temperature on the arabinose in CE3::Tn514b. 68

Table 4-12. Accumulation of arabinose by strain CE3 and CE3::Tn5 under different temperature regimes.

Arabinose content |ig/mg dry cell CE3 CE3: ;Tn514b

Temperature °C mean SE mean SE 10 1.5 ±0.0 3.0 ±0.0

20 1.4 ±0.3 2.7 ±0.1

30 1.6 ±0.1 2.2 ±0.0 40 ND ND ND ND

ND=not detected

Nodulation Dominance Studies

Since trehalose had been detected in strains CE3 and

CE3::Tn514b under salt stress both in culture and in the nodules, experiments were conducted to determine the role of

trehalose in nodulation under hyperosmolarity . Strains CE3 and CE3::Tn5 were inoculated, either as single strains or mixed in ratios (1;1, 1;2, 2;1), on bean seedlings.

Proportional nodule representation is summarized (Table 4-

13) .

The cell ratio of strains in the inoculum was a major determinant in the nodule strain representation as indicated by the significant F test (Appendix C-5 and C-6) in both trials. However, strain CE3::Tnl4b showed superior nodulation ability as noted by the high number of nodules formed, especially when the strains were present in equal ratios or numbers. 1 . ^

Table 4-13. Effect of inoculum ratios on percent nodule occupancy by strain CE3 and CE3::Tn514b.

FIRST TRIAL Inoculum ratio Percent nodule occupancy CE3 :CE3Tn514b CE3 CE3: :Tn514b

1 : 0 100a Oe 65c 1 : 34c 2:1 56b 39d 1:2 lid 87b 0:1 Oe 100a SECOND TRIAL

1:0 100a Od

7 1 : 21c 8b 2;1 38b 62c 1:2 18c 81b 0:1 Od 100a

Means within a column followed by the same letter are not significantly different P=0.05 (Duncan's Multiple Range Test)

The effect of salt was variable. However, there was an interaction between salt and inoculum ratio and the response is given in Table 4-14. Dry matter yield (mean weight per plant) was variable. In the first trial, no significant effect was observed due to inoculum ratio. However, strain

CE3::Tn514b applied alone was superior to the other inoculum

a significant ratios (Table 4-15) . Surprisingly, salt had stimulatory effect, with a mean weight of 361.33 mg per plant versus 288.73 mg plant”^ for nonsalinized plants. In the second trial salt had no effect; mean weight for plants varied slightly (386.93 mg and 380.45mg) for nonsalinized . .

70 and salinized plants, respectively. Strain CE3 inoculated alone gave better yield than any of the mixed inocula.

Based on dry matter yield, the competitive ability (the ratio of the dry weight given by the mixed inoculum to the dry weight obtained with the single strain inoculum x 100) of the two strains was computed (Table 4-16)

Table 4-14. The effect of sodium chloride on nodule occupancy by strains CE3 and CE3::Tn514b.

FIRST TRIAL PERCENT NODULE OCCUPANCY

NaCl (M) CE3 CE3: :Tnl4b

0 34b 62a 0.05 40a 56b Mean 37 59 SECOND TRIAL PERCENT NODULE OCCUPANCY

0 38a 62b 0.05 32b 68a Mean 35 65

Means within a column followed by the same letter are not significantly different, P=0.05 (Duncan's Multiple Range Test) 71

Table 4-15. Dry matter yield as affected by inoculum ratio.

CE3:CE3: :Tn514b Dry matter yield (mg/plant) Inoculum ratio First trial second trial 1:0 317.2a 538.8a 1:1 310.7a 312.8b 2:1 313.5a 341.4b 1:2 330.0a 358.9b 0:1 353.8a 366.5b

Means within a column followed by the same letter are not significant P=0.05 (Duncan's Multiple Range Test).

Table 4-16. Competitive ability of strains CE3 and CE3::Tn514b in mixed inocula.

FIRST TRIAL

CE3:CE3Tn514b % competitive ability Inoculum ratio CE3 CE3: :Tn514b 1:1 98 88 2:1 99 88 1:2 104 93 SECOND TRIAL

1:1 58 85 2:1 63 93 1:2 66 97 CHAPTER V DISCUSSION

Response of Rhizobium lecruminosarum Biovar vhaseoli to Salt Stress

The hypothesis of this study is that trehalose enhances the survival of Rhizobium strains which accumulate it under stress conditions, thus rendering them dominant in nodule formation. To unequivocally answer the hypothesis, it was necessary to use strains that did not vary appreciably except in their ability to accumulate or synthesize trehalose. Strain CE3, resistant to 200 ^ig ml'^ of streptomycin, and its Tn5 derivative mutants, resistant to

200 |ig ml"^ each of streptomycin and kanamycin, were selected for this study. The mutants were then screened for a trehalose negative mutant (s) for comparative studies with trehalose positive strains. No such mutant was found.

Hoelzle (1990) performed a similar exercise and was unable to identify a trehalose negative Rhizobium mutant.

Generation of Tn5 rhizobial mutants has a possible problem, as there could exist multiple copies of the gene of interest. In that case, when one copy is successfully interrupted, other (s) would still be functional. The occurrence of multiple gene copies has been documented in many species of rhizobia and more than one gene copy may

72 .

73 function simultaneously (Quinto et al., 1982/ Flores et al.,

1987; Appelbaum et al . , 1988).

Boos et al. (1987) and Giaever et al. (1988) identified genes in Escherichia coli that are involved in trehalose metabolism. They are otsA and otsB (osmoregulatory

trehalose synthesis genes) . These genes, when fused with

lacZ gene, form an operon that osmotically induces £>- galactosidase activity. This activity is increased greatly

if Escherichia coli is grown in a medium of elevated osmotic

strength. Mutants in these genes are osmotically sensitive when cultured in a glucose-mineral medium. An attempt to

clone these genes and introduce them in Tn5 mutants of

Rhizohium leguminosarum biovar phaseoli strain USDA 2667,

was unsuccessful (Hoelzle, 1990) . Thus, to date no

rhizobial trehalose negative mutant is available to conduct

strict comparative studies.

Because of the lack of trehalose negative mutant (s)

CE3::Tn5 mutants were screened and grouped on the basis of

their tolerance to different salt levels on YEMA medium.

Strain CE3 and CE3::Tn512b poorly grew on media that had

salt at a concentration of 0.2 M NaCl, strains CE3::Tn514b

and CE3::Tn516b were able to withstand 1.0 M NaCl while

CE3::Tnl8a survived only at 0.5 M NaCl. The five strains

were tested for their ability to accumulate trehalose when

grown in the defined medium (Manhart and Wong, 1979) Their

response to osmotic potential would then be a basis for

further comparative studies. 74

Osmotic potential is a problem for all forms of life and adaptation to osmotic stress (osmoregulation) allows cells to tolerate adverse conditions such as drought or high salinity. A common mechanism of osmoregulation is the accumulation of inorganic or organic solutes or both, in the cytosol. This restores turgor in plants and microbes or controls cell volume in wall-less cells (Le Rudulier, 1984;

Some compounds that accumulate under Yancey et al . , 1982). osmotic stress are: polyols (sugars, sugar alcohols, and (glutamate, glycerol) , amino acids and amino derivatives

proline and betaines t-aminobutyric acid, taurine) , and urea and methylamines (trimethylamine-N-oxide) . Trehalose is one of such compounds and often increases in concentration in cyanobacteria and Escherichia coli when subjected to elevated osmotic potential (Larsen et al., 1987; Giaever et

al . , 1988; Reed et al., 1986).

Some reports indicate that trehalose is accumulated by some Rhizohium strains and its concentration increases as the osmotic potential of the medium increases (Hoelzle,

1990; Smith and Smith, 1989) . Hoelzle (1990) also found that six strains of Rhizohium leguminosarum biovar phaseoli neither showed any difference in internal trehalose concentration nor grew at all under salt stress. Some of them were intolerant of even low salt concentrations (25 mM excess sodium chloride was present in the medium) . Other

species of Rhizohium tested did not show any change in their trehalose content but were able to grow. In Rhizohium 75 meliloti studied by Smith and Smith (1989), another osmolyte

dipeptide, N-acetylglutamylglutamine amide (NAGGN) , was identified. Such observations suggest that trehalose may not be a universal osmoticum in rhizobia. In the present experiments strain CE3 accumulated more trehalose than the mutants. Secondly, the amount of trehalose accumulated decreased with increasing salt levels. Such observations are contrary to those outlined above. However, noting that the cells were incubated for 96 h, it is possible that the cells were harvested in a post-logarithmic phase of growth.

As suggested by Hoelzle and Streeter (1990b) rhizobia accumulate substantial amounts of trehalose when in the log phase. Furthermore, Hoelzle (1990) found that approximately

70 h after inoculation the internal reserves of trehalose

rapidly disappeared from the cells. Judging from their

growth response curves when cultured in yeast extract broth

of differing salt concentrations, growth of these cells was

seriously curtailed at high salt concentrations. The 96 h

incubation period was adopted from Streeter (1985) who used

Rhizobium phaseoli strains grown in defined medium. He

detected trehalose in these strains. However, he did not

subject his strains to salt stress.

Arabinose, which is a constituent of the defined

medium, was also detected. The cells may have a mechanism

for arabinose uptake, replacing trehalose in osmoregulation.

Arabinose was detected at all levels of salt concentration;

it was quantitatively less abundant than trehalose. .

Osmoregulation may be effected in the same organism by more than one compound. For example, Reed et al. (1987) noted that glycerol was the major low molecular weight carbohydrate in salt stressed yeast cells. However, lower levels of arabitol and trehalose were also reported as osmotica, especially in stationary-phase yeast cells.

Both Escherichia coli and Rhizobium species are Gram negative bacteria, while cyanobacteria and rhizobia are diazotrophs. Cyanobacteria and Escherichia coli both accumulate trehalose under high levels of osmotic potential.

Deductively, it was speculated that rhizobial species may accumulate trehalose either in the free or symbiotic state

when exposed to osmotic stress (Hoelzle, 1990) . It is not definite whether the nodule environment is osmotically stressful for bacteroids. Some estimates of nodule solute concentration are higher than many common rhizobial media.

For example, YEMG (Vincent, 1970), a medium often used to grow rhizobia, has an osmotic potential of 100 mOsmol, and well watered soybean nodules have an osmotic potential of

150 mOsmol (Fellows et al., 1987), while cowpea {Vigna

unguiculata) nodules register 220 mOsmol (Khan-Chopra et al., 1984). These latter studies measured water potentials in water stressed nodules and reported a value of 580 mOsmol

Streeter and Salminen (1987) suggested that rhizobial

strains that accumulated trehalose in culture also

synthesized and accumulated the sugar in the nodules. In 77 tti0 6xpGiriinGnt s bGing rGportGd trGhalosG was dGtGctGd in culturG undGr low salt conditions and in thG crudG nodulG

Gxtract. SucrosG which was the most abundant saccharide; glucosG and fructose were also detected in the nodules. The saccharides detected were fewer than those reported by

Streeter (1980) for soybean nodules. It is not clear whether this is due to differences in plant or bacterial

attributes or to both. Another possible reason could be

related to the low light intensity under which the plants

were grown. When the plants were supplied with 20 ml per

plant of Jensen's medium containing sodium chloride at a

concentration of 2.925 gm L there was a shift in the

carbohydrate profile. In strain CEB there was a reduction

in sucrose content while in CEB: :Tn514b no difference was

observed. Since the reduction in sucrose was only apparent

in plants inoculated with CEB and not in the mutant, it may

reflect a strain effect rather than a host effect since only

one cultivar of beans was used. Secondly, in both strains,

glucose nor fructose could be detected, suggesting a

reduction in their synthesis. It is likely that sucrose

hydrolytic enzymes are salt sensitive. However, trehalose

was detectable and increased with the addition of salt.

However, the increase was not statistically significant.

A nonsignificant F test minimizes the role of trehalose

as an osmoticum in bean nodules in this study. This

observation conforms with the reports of Francoise-Fougere

et al. (1991). Analysis of carbohydrates in nodules and 78 bacteroids of Medicago sativa showed that sucrose, pinitol, glucose and ononitol were the major saccharides. Maltose and trehalose occurred in low concentration (Francoise- compounds were the principal Fougere et al . , 1991). Other osmolytes and trehalose remained low when the plants were grown in salt stress conditions. Recently, Streeter (1992) analyzed solutes in the apoplastic region of intact soybean nodules. He detected a variety of solutes that included carbohydrates, amino acids, organic acids and ions. The major solute was allantoic acid. Calculated values for solute concentrations in cortical apoplast were high. This suggested that apoplastic solutes may represent a significant osmotic component in nodule cortex.

The above account reveals that several organic compounds are found in the nodules. They may have similar roles such as osmoregulation or other functions in the symbiosis. Therefore, it is difficult to attribute one role to one particular compound unless the others are artificially blocked. In this study analysis of amino acids and organic acids, compounds mentioned as possible osmolytes, was not done.

The leguxae-Rhizobium association is symbiotic. The plant supplies the microsymbiont with reduced carbon compounds and in return the plants get reduced nitrogen.

The major carbohydrate that enters the nodules is sucrose

(Bach et al., 1958; Streeter and Salminen, 1987). Similar

observations were obtained in the trials being reported. . . .

Sucrose is then cleaved either by alkaline invertase (EC

3.2.1.26) or sucrose synthase (EC 2.4.1.13) as reported by

Morell and Coperland (1984) and Morell and Coperland (1985)

The products of sucrose hydrolysis, glucose and fructose, may be metabolized further in the plant fraction of the nodule

Trehalose, a nonreducing glucose-glucose disaccharide, is present in most legume nodules that have been studied so

far (Streeter and Salminen, 1987) . This sugar is not synthesized in higher plants and may be toxic to plants if it is not hydrolyzed by trehalase. Several lines of evidence suggest that trehalose in nodules is synthesized in bacteroids since the key enzymes for trehalose could only be found in bacteroids (Streeter and Salminen, 1987) . It was difficult to explain why bacteroids contain only a small fraction (20%) of nodular trehalose. Yet trehalase occurs only in the host cytosol. Furthermore, there is no knowledge that indicates how accessible trehalase is to the

substrate. On average the bacteroids only retain 50% of the trehalose they synthesize. However, there are variations

among different species and their age. In the present

report trehalose was detected in nodules in all cases.

However, the nodules were not fractionated and thus it is not possible to make any comparisons in this respect. Another intriguing observation that has been reported

is the negative correlation between trehalose and acetylene

reduction (Streeter, 1985; Streeter and Salminen, 1987) .

Other sugars were not correlated to acetylene reduction.

While Streeter and Salminen (1987) did not think that such a correlation was an artifact, they could not explain the phenomenon. They speculated that trehalose was synthesized in the bacteroids because there was less demand for reduced carbon in the bacteroids. By extrapolation strains of Bradyrhizobium japonicum that synthesize high concentrations of trehalose in culture also synthesize high levels of trehalose in nodules. This leads to lower nitrogenase activity

Nitrogenase activity in the present report was estimated by measuring dry matter yield and determining the total nitrogen content of the plant parts. This approach was adopted because it is simple and since the plants were wholly dependent on fixed nitrogen except for the nitrogen control. No significant difference was obtained between the treatments. The lack of significant response was attributed to the short period (18 d) the plants were allowed to grow.

The short period was necessary because longer periods led to nodule degeneration, in which case no trehalose would have been detected. Trehalose is synthesized in actively nitrogen fixing nodules. Secondly, plants were grown in plastic pouches (5.5 by 6 inches) which also limited the time plants could be allowed to grow. In addition, during this short period of growth the nitrogen supply from the cotyledons appeared sufficient to sustain plant growth.

Furthermore, determination of total nitrogen is not as ,

81 sensitive a method as is acetylene reduction assay in detecting small differences (Somasegaran and Hoben, 1985)

Qualitatively, nodules from plants which were salt treated had low hemoglobin content and no or few PHB granules compared to nodules from nonsalinized plants.

These potentially important observations need further quantitative investigation. Leghemoglobin is a product of the late nodulins, and is abundantly produced in actively nitrogen fixing nodules (Nap and Bisseling, 1990) . Leghemoglobin regulates the supply of oxygen to the bacteroids so that it is available in sufficient amount for oxidative phosphorylation to provide energy and yet is at levels harmless to the nitrogenase enzyme complex (Postgate, might be 1987) . Nitrogen fixation in salt stressed nodules defective if leghemoglobin content is affected. Such deductions agree with the reports of Singleton and Bohlool and Bohlool (1983) . Using a split root system, Singleton

(1983) found that nitrogenase activity was reduced when the nodulated half of the root system was salinized. The

reduction was greater if the whole root was salinized.

Similarly, Bernstein and Ogata (1966) and Zehran and Sprent

(1986) indicate a negative effect on nitrogen fixation and

other related processes by salt stress.

The reduced quantity or absence of PHB granules in

salt- affected nodules would imply reduced carbon supply to

the nodule. Sucrose in strain CE3 was decreased under

saline conditions. This may to lead to reduced carbon 82 supply to the bacteroids since sucrose is the main carbohydrate entering the nodules (Reibach and Streeter,

1983; Kouchi and Yoneyana, 1986) . Streeter (1985) found a positive correlation between sucrose and trehalose (r=0.59).

In the studies under consideration, a correlation between sucrose and trehalose was also noted (r=0.7-0.9) but broke down under salt stress. In the current report, a nonsignificant increase in trehalose amount under salt stress was observed though the amount of sucrose decreased.

The presence of trehalose would suggest that trehalose was synthesized under elevated osmotic potential. Its participation in osmoregulation was diminished by a nonsignificant F test.

Poly-I5-hydroxybutyrate was shown to accumulate in soybean nodules (Gerson et al., 1978). Although PBH represents 50% of the bacteroid volume (Werner and Marshall,

1978), there was no specific role attributed to trehalose.

Clugston (1993) found a positive correlation between strain

PHB production or utilization and bacterial survival in soil. However, earlier reports (Bergersen, 1977) had failed to detect such a relationship. In the current study, it was noted that PHB granules were either absent or present in small quantities in salt treated nodules. As noted earlier, this may be a reflection of either reduced carbon supply or that PHB had been catabolized under the stress environment. . . .

Carbon Utilization by Rhizobium leauminosarum Biovar phaseoli Strain CE3 and its Tn5 Mutants grown on Trehalose and Mannitol

Rhizobia were reported to grow on a variety of carbon sources such as sucrose, maltose, and mannitol (Fred et al., growers) produced an 1932) . Some species of Rhizobium (fast acid reaction and others (slow growers) produced an alkaline reaction. Later, Vincent (1970) also indicated that mannitol was often the preferred source of carbon for routine culture of rhizobia. However, other forms of carbon, such as sucrose and glucose, could also be utilized.

Slow growing species were capable of growing on galactose and arabinose, Elkan (1984) quoted Allen and Allen (1950) and Graham (1964) who reaffirmed the differences between the slow and fast growing groups of rhizobia. They advocated the separation of the genus Rhizobium into two genera. The strains used in this study were fast growers and acid producers

Bradyrhizobium japonicum strains USDA 110, 138, and

61A76 accumulated trehalose and the amount varied depending on the medium used (Streeter, 1985) Irrespective of the type of medium employed, none of the strains could grow on trehalose as the sole carbon source. Such an observation is typical of slow growing Rhizobium species (Glen and

Dilworth; Sadowsky et al., 1983; Stowers and Eaglesham,

1984) It was paradoxical that strains accumulated trehalose and yet could not grow on it. Deductively,

Streeter (1985) speculated that trehalose could be used ,

84 under some special conditions. Contrary to the observations noted for the slow growing rhizobia, fast growers accumulate less trehalose but are able to utilize trehalose even when it is the sole carbon source (Streeter, 1985). The strains used in this study were able to grow when trehalose replaced mannitol both in solid media and in broth.

The observations that some rhizobial strains can grow

on trehalose would suggest that trehalose were either an

energy source or a storage compound in rhizobia. However,

Mellor (1992) and Streeter (1985) underscore such

speculations for reasons discussed in chapter II.

Accumulation of Trehalose Under Varying pH Conditions

Previous reports (Streeter, 1985; Hoelzle and Streeter,

1990b) indicate that trehalose is accumulated in rhizobia in

a free or symbiotic state. Secondly, its quantity increases

under environmental stress conditions such as high

osmolarity (Hoelzle, 1990) or low oxygen tension (Hoelzle

and Streeter, 1990b). In the latter case, the biochemical

mechanisms of increased trehalose accumulation were not

clear. The authors speculated that the increase was due

either to increased synthesis, decreased breakdown or both.

Low oxygen tension was reported to reduce the activity of et al tricarboxylic acid cycle enzymes in bacteria (Gray .

1965); Manhart and Wong, 1979). As a result of the above

event, carbon is available for trehalose synthesis.

The Rhizobium-legume symbiosis is influenced by several

environmental factors, including pH and its associated . .

85 problems, such as aluminum toxicity and calcium deficiency to (Lie, 1971) . The microsymbiont is particularly sensitive acidic conditions. For example, growth of Rhizobium leguminosarum biovar trifolii was restricted at pH <5.5

Acidity was noted as a major cause for the poor nodulation of clover (Thornton and Davey, 1983; Wood and Cooper, 1985)

In addition, low numbers of Rhizobium leguminosarum biovar trifolii were associated with low pH soils which subsequently influenced the nodulation of the host plants.

Yet, multiplication of rhizobia is important in a competitive environment. For example, Lotus rhizobia that are acid tolerant demonstrated superior nodulating ability compared to the acid sensitive ones (Cooper et al., 1985).

The absence of trehalose in the strains under study, when grown under varying pH conditions, is difficult to explain. It is possible that survival or growth under these conditions is achieved by different mechanism (s)

Homospermidine, a polyamine, conferred acid or base tolerance to the fast growing Rhizobium fredii strain P220 when grown in media ranging from pH 4. 0-9. 5 (Fujihara and

Yoneyama, 1993) . On the other hand, the growth of

Bradyrhizobium japonicum, which did not accumulate a

significant amount of homospermidine, was repressed at elevated pH. The same compound also participated in

osmoregulation when strain P220 was grown in a medium of

0.15 M NaCl. .

Chen et al. (1993) have reported that the genes involved in acid tolerance are carried on the Sym plasmid the second largest plasmid, in Rhizobium leguminosarum biovar trifolii. The phenotypic characteristics associated with acid tolerance include production of large amounts of exopolysaccharides, ability to maintain an internal pH near neutrality, reduced permeability to protons and likewise an efficient mechanism for extrusion of protons. These traits were missing in strain P22, which had been cured of its Sym plasmid. Such capabilities were restored on reconstituting the strain with the plasmid.

The accumulation of arabinose in the cells was also noted earlier in experiments related to osmotic stress. As noted by Elkan and Kuykendall (1982), carbohydrate metabolism in Rhizobium is not fully understood. In the defined medium used in these studies, two sources of carbon were supplied. The accumulation of arabinose in these cells could be interpreted as catabolite repression, a phenomenon observed in Escherichia coli (Ronson and Primose, 1979), and in Rhizobium trifolii (Wood and Cooper, 1988)

Vincent (1970) notes that mannitol is a more commonly used carbon source in media for rhizobia, especially the fast growing strains. Arabinose is preferred by the slow growing strains. Accordingly, in the defined medium employed in these experiments, mannitol was the preferred carbon source by strain CE3 and its Tn5 mutants since they are fast growers. Some arabinose that was taken up was not 87 completely utilized. Glenn and Dilworth (1981) studied the uptake and hydrolysis of disaccharides in both the fast and slow growing rhizobia. They found evidence that fast growing strains have two distinct disaccharide uptake systems. One system carries sucrose, maltose and trehalose and the other carries lactose. Since cellobiose affects the uptake of other disaccharides it suggests that it is also transported via the same route. On the contrary, slow growing strains failed to utilize disaccharides and this was attributed to either lack of the disaccharide uptake system or demonstrable hydrolytic enzymes to cleave the disaccharides. According to Duncan and Fraenkel (1979), the catabolic pathway for arabinose in the fast growing strains

yields a-ketoglutarate . In slow growing strains it leads to the production of glycoaldehyde and pyruvate (Pedrosa and

Zancan, 1974) .

The failure to detect trehalose at pH 7 was difficult to explain since it was detected at pH 6.8 in earlier studies. One reason could be the changes that occurred in the medium as the pH changed from acid to neutral and alkaline conditions. First, the defined medium of Manhart and Wong (1979) is complex; it is formulated with many macro and micro elements. The solubility of these compounds changed as the pH changed, as indicated by the presence or absence of precipitates. At low pH values (pH 5 and pH 6),

some elements like iron, zinc, and manganese could have existed in solution at levels toxic to the bacteria or .

88 inhibitory to their enzymatic activities. One may expect reactions similar to those that would occur if aluminum were present at low pH values. At neutral and alkaline pH manganese, iron and other elements may precipitate and become unavailable for bacterial use and metabolism.

Coincidentally, Hoelzle and Streeter (1990a) found that a- glucosidases, which includes trehalase (which breaks down trehalose), operated in the pH range 5. 0-8.5, with optimum activity around pH 7.0. The increased activity of such enzymes may have contributed to the breakdown of trehalose if it were synthesized at all. Subsequent analysis was likely to reveal no trehalose presence.

Cells of strain CE3 and the Tn5 mutants displayed different growth patterns when grown in YEM broth of varying pH. At pH 5, there was a pronounced lag phase ranging from

6-24 hours depending on the strain. Then the cells went through an exponential phase and a decline phase. The long lag phase may be necessary for the few cells that survive the acidic conditions to multiply to a level where turbidity is visible. Wood and Cooper (1988), working with Rhizobium trifolil strains HP3 and BEL1192, found that the cells multiplied at the same rate. The pH of the medium remained unchanged at 5.5 until the cell density was 10”®-10”’ colony forming units (cfu) This critical density coincided with visible turbidity in the medium. Furthermore, the growth of strain BEL1192 exhibited diauxic growth response at pH 4.5.

A fast initial growth rate was replaced by a slow one. As .

89 the cell numbers increased, the pH of the medium also increased. They attributed this to the increased production of alkali by the cells. Secondly, they had used glutamate which not only served as a pH buffer but also as nitrogen and carbon source. When it was exhausted, the buffer capacity was dissipated and the pH of the medium changed drastically. In the experiments being reported, the pH of the medium was not monitored. Usually, fast growing rhizobia acidify YEM and if this occurred in this case the medium would be expected to become more acid. Consequently, the growth that was observed might have come from a few cells that survived the stress environment.

At pH 6, the lag phase was reduced to almost a half that at pH 5. Strains CE3::Tnl6b and 18a continued to grow well. Strain CE3 and CE3::Tn514b had a shorter lag phase but did not attain high values of optical density. The diauxic growth response at pH 7 was unexpected. The multiphase growth would indicate multiple stresses to which the surviving cells had to adjust before growing. It is difficult explain the growth of strain CE3 and CE3::Tn514b

at pH 8 while the others failed to grow at all.

The Response of Rhizobium lecruminosarum Biovar vhaseoli to Chloramphenicol

Microbial antibiotics or toxins frequently limit the

establishment of the legume-Rhizobium symbiosis (Graham,

1963; Lowendorf, 1980; Triplett and Barta, 1987) The

effect of antibiotics on rhizobia in culture media was . .

demonstrated by Graham (1963) . It was reported to occur under field conditions (Lowendorf, 1980) . The problem of second year mortality in Rhizobium trifolii in western Australia was attributed to the presence of a water soluble toxic compound (Lowendorf, 1980)

In addition, a few rhizobial species may produce toxins that are active against other rhizobia. For example, triflotoxin (an antibiotic) is produced by Rhizobium leguminosarum biovar trifolii strain T24. The toxin inhibits the growth of other Trifolium strains and other related biovars of Rhizobium leguminosarum (Triplett and authors Barta, 1987) , Using Tn5 mutagenesis, the latter concluded that triflotoxin production and nodulation were important for nodulation dominance. Another example of rhizobial bacteriocin is found in Rhizobium leguminosarum biovar trifolii strain CB 782 (Hodgoson et al., 1985),

Similarly, they were able to link bacteriocin production with nodulation dominance. However, Triplett and Barta,

(1987) were opposed to their reports since no Tn5 bacteriocin negative mutant (s) were used. Earlier,

Schwinghamer et al. (1973) had suggested the possibility of arming inoculant strains with bacteriocins . This would improve their competitiveness with less desirable soil rhizobia

In this study, it had been postulated that the presence of antibiotics (such as chloramphenicol) in the growth medium of sensitive strains would create a stressful .

91 environment. In order to survive, the rhizobial strains would respond by accumulating trehalose. Trehalose would then be linked to survival mechanism (s) under unfavorable conditions. It would then be hypothesized that the disaccharide conferred a competitive advantage to the trehalose carriers over noncarriers. Subsequently, trehalose accumulating strains would show nodulation dominance when coinoculated with nontrehalose accumulating strains. However, no trehalose was detected. The presence of arabinose may be attributed to an uptake mechanism by the bacteria, as mentioned earlier in the previous section.

Toxins active against rhizobia may come from non- microbial sources. For example, some leguminous seeds have been reported to possess a toxin that is active against

rhizobia. This is critical when the inoculum is applied

directly onto the seed. Materon and Weaver (1984) found a

thermostable, water soluble diffusate from autoclaved and

non-autoclaved seeds that inhibited the growth of Rhizobium

trifolii .

The Effect of Temperature on the Accumulation of Trehalose and Arabinose in Strains CE3 and CE3::Tn514b

The influence temperature has on the growth of rhizobia

in culture media or survival in soil or the symbiotic

attributes, has been investigated (Parker et al., 1977;

Eaglesham et al., 1981; Eaglesham et ad., 1983; Piha and

Munns, 1987) In most cases, temperatures above 30°C were

detrimental. However, some heat or temperature resistant 92 strains were identified (Piha and Munns, 1987; Eaglesham et al., 1981). The latter authors isolated rhizobial strains from different regions of Nigeria which exhibited different degrees of tolerance to high temperatures. Isolates from

Maradi were not affected when cultured in broth at 40°C and retained their symbiotic efficiency, which indicated that curing did not occur. Piha and Munns (1987) showed that most bean rhizobia were affected at high temperatures, forming small nodules with low specific activity.

Generally, strain CIAT 889 is tolerant to high temperatures.

Survival of rhizobia in soils under high temperatures was also linked to the moisture status of the soil (Foulds,

1971; Danso and Alexander, 1974) . Rhizobia were reported to survive better at high temperatures in dry soils.

Unfortunately, in the cited literature no mechanism for survival under elevated temperatures is given.

Microorganisms that survive above 45°C, called thermophiles, have adaptational properties that enable them survive such

conditions (Paul and Clark, 1989) . They possess a high proportion of unsaturated lipids in their membranes which prevents them from melting at high temperatures. In addition, thermophiles synthesize heat resistant enzymes which are not inactivated by extreme temperatures.

Thermophiles at high temperatures exhibit a longer generation time compared to mesophiles. This is because they expend energy to repair the denatured enzymes. .

In the current study, the growth of both strains was curtailed at 40°C as indicated by the reduced dry weight of cells. While trehalose was not detected, arabinose, a constituent of the defined inediuiti, was accumulated. Its amount decreased with increasing temperature. It is impossible to explain the observed relationship between arabinose and temperature. Possibly, as temperature increased from 10°C to 30°C, metabolic activity increased.

This demanded utilization of more carbon materials such as arabinose

The main objective of this study was to relate trehalose to rhizobial survival under stressful environmental conditions. This relationship has not been evident. Hottiger et al. (1987) found that yeast cells of

ScLCcharomyces cerevisiae accumulated a large amount of trehalose when they were transferred from 27°C to 40°C. On reversing the conditions the amount of trehalose rapidly declined. The amount of trehalose phosphate synthase and trehalase also increased simultaneously. Heat shock precipitated both anabolic and catabolic trehalose processes. The biosynthesis of trehalose consumed three molecules of ATP while trehalose hydrolysis yields none.

Deductively (Hottiger et al., 1987) concluded that an energy-consuming futile cycle existed between trehalose and

glucose. Additionally, at 30°C, 1 ^imole of ATP is produced

per gram of dry weight of cells. Yet a sevenfold increase

is produced at superoptimal temperatures. Therefore, the 94 trehalose cycle is a device to counteract ATP overproduction. Two other possible reasons for the futile cycle were given: glucose in the 1) the cycle maintains a constant level of cytosol so that it does not interfere with other

biochemical processes. the cytoplasm 2) the accumulated trehalose would prevent from damage by heat shock or prolonged dehydration due

to prolonged exposure to high temperatures.

Since cells of CE3 and CE3::Tn514b did not accumulate trehalose, it implied that the cells were vulnerable to such effects. This led to reduced cell growth. Nodulation Studies

When two or several Rhizobium strains occur in the presence of a homologous leguminous plant one or more strain(s) form(s) nodules in preference to the others

(Robinson, 1969; 1975; Amarger and Lobreau, 1982; Beattie et Rhizobium species are preferentially selected al . , 1989). by the host plant. The phenomenon is referred to as inter-

strain nodulation competitiveness or lately as nodulation

dominance. The number of nodules formed by a given strain

on a homologous host also depends on rhizobial numbers. As

the relative proportion of a strain in a mixed inoculum is also j_nc]f 0 ased, the number of nodules formed by the strain

increased (Weaver and Frederick, 1974; Sparrow and Ham,

1983) . 95

In the current studies, the inoculum ratio significantly influenced the proportion of nodules formed by either strain CE3 or CE3::Tn514b. As the ratio of either strain increased in the mixed inoculum, the number of nodules also increased. Amarger (1981) and Amarger and

Lobreau (1982) have discussed mathematical models developed to describe the competitive ability of a given rhizobial strain. The models attempted to determine the competitive ability of strains in mixed inocula or in the presence of resident rhizobia in the soil. In most of their experiments, they used either effective or ineffective strains. With such a combination they defined a positive correlation between the percentage of nodules formed and the proportion of the number of effective strains in the inoculum. Similarly, the shoot dry weight increased with the number of effective nodules. Later experiments by

Amarger and Lobreau (1982) revealed a discrepancy in the above model. It was noted that the slope of such a curve was expected to give the competitive ability of a strain under study. This was found to be false. Some strains with the lowest coefficients, indicative of poor compatibility between the strain and the host, were found to form 90% of the nodules. Subsequently, another model was derived that used the logarithmic ratio of strains in the soil or medium and in the inoculum. Using their own data and that of other workers, Amarger and Lobreau (1982) were able to

authenticate their model and derive coefficients (k) that 96 were significant and ranged from 0.285-0.530. Beattie et al. (1989) added another term to the above model, which accommodated the double nodule occupancy factor. They reasoned that double nodule occupancy was important in their experiments. Double nodule occupancy varied from 0-16% in nonsterile soil and 0-40% in sterilized matrices. The method employed in the current report relied upon the differential sensitivities of CE3 and the mutants to streptomycin and streptomycin plus kanamycin. If double nodule occupancy occurred, the method likely underestimated the percent nodules formed by CE3. Detection of CE3 was based on its failure to grow on YEMA containing both antibiotics. Thus, if the two strains occurred in the same nodule only the mutant was likely to grow. As a result this would underestimate the percent of nodules occupied by

strain CE3. Double nodule occupancy was emphasized by Beaty

Kang (1991). However, earlier work et al . (1989) and ^ by Roughley ^ (1976) and Amarger (1981) overlooked the effect of dual strain nodule occupancy. Serological techniques were not used because CE3 and the mutants differed slightly. Therefore CE3 and the Tn5 mutants were unlikely to differ in their antigenic determinants.

To generalize the above model, Amarger and Lobreau

(1982) compared competitive ability regression curves to that obtained for Freundlich adsorption isotherms. Applying the same analogy, the number of nodules formed by a strain varies as a function of the number of rhizobial bacteria in 97 the medium. It was reasoned that the root possesses a certain number of adsorption sites and that for each

Rhizobium strain there is a different isotherm. The isotherm corresponds to the affinity of the strain for such adsorption sites. Differences in competitive ability would then represent a selectivity mechanism in strain adsorption by the root. Such arguments are supported by the observations made by Dazzo et al. (1976). They demonstrated that infective Rhizobium trifolii, when compared with noninfective Rhizobium trifolii or other Rhizobium species, was preferentially adsorbed to clover root hairs.

In spite of the observations and evidence cited above that emphasize the importance of rhizobial cell numbers in determining strain dominance, there is some disagreement

(Hubbell, personal communication; Moawad et al., 1984;

Silvester-Bradley as quoted by Giller and Wilson, 1993) .

First, Hubbell (personal communication) is strongly opposed to the concept of discrete infection sites on the root hair.

However, reviews of the physical and biochemical aspects of nodulation (Dazzo et al., 1976; Vincent, 1980; Hubbell,

1981; Sprent and Sprent, 1990) and molecular genetics

reports (Quispel, 1988; Nap and Bisseling, 1990; Kijne,

1992; Verma, 1992) concur that both the host and the

Rhizobium have a complex communication system which controls

a cascade of events that culminate in dinitrogen fixation.

As suggested by Hubbell (personal communication) the

mechanism (s) involved in nodulation are likely to be 98 influenced by both intrinsic and extrinsic factors which as yet are not fully understood.

Components of a mixture of Bradyrhizobium japonicum serogroups may exist in approximately equal numbers but serogroup 123 always forms the majority of the nodules on the host plant (Moawad et al., 1984). A significantly greater number of rhizobial cells of one strain over the other strains is required before one apparently dominant strain can be suppressed. Weaver and Frederick (1974) suggested that an inoculum rate of 1000 times greater than the normal recommended one should be employed if the inoculum strain were to form 50% of the nodules in the presence of other or indigenous strains that were present in a magnitude of more than 10^ gm"^ of soil.

The major objective of this study was to determine whether the accumulation of trehalose by rhizobial strains plays any role in their nodulation patterns or their competitive ability, especially under stress environment.

Earlier results revealed that strain CE3 accumulated significantly more trehalose both in culture and nodules, either in the presence or absence of sodium chloride, than

CE3::Tn514b. The results obtained from the competition studies showed that the latter strain formed the majority of the nodules whether the plants were treated with salt or not. This implied that trehalose either had no role in nodulation dominance or it had a depressive effect.

Alternately, strain CE3 is more tolerant to salt stress, or 99 the salt level used was not stressful for the double mutant.

Possibly other mechanisms were employed to alleviate the salt problem. In addition, further investigations incorporating the double nodule occupancy factor are needed to avoid underestimation of one strain. Currently, it is impossible to make conclusive remarks in the absence of a trehalose negative mutant. Such a strain would also indicate the critical level of trehalose necessary for a certain process. Nonetheless, a negative relationship between nodulation dominance and trehalose will not be strange. It will be recalled that Streeter (1985) reported a negative correlation between acetylene reduction and trehalose accumulation. To date, no explanation has been offered for the observation.

When the competitive ability of the two strains was computed using the equation given in chapter III, inconsistent results were obtained. In the first trial, the two strains exhibited similar competitive ability and there was no significant difference between the strains in dry matter production. However, in the second trial strain CE3 had low competitive ability but showed a significantly higher dry matter yield. Such observations are difficult to explain. However, one reason could be that the model employed was not suitable. In the first instance, it was mainly developed for paired effective and ineffective

strains. It would appear both strain CE3 and its Tn5 derivative are approximately similar in their effectiveness. .

100

Consequently, the model was not likely to give meaningful computations

It was also surprising that strain CE3::Tn514b retained both its infectivity and nitrogen fixation traits.

Rhizobial mutants have been reported to lose their nodulation potency and nitrogen fixation attributes (Turco et al., 1986). Another phenomenon that was noted was that of mutants reverting to the wild type strains. A small percentage (1-4%) of the nodules contained rhizobia that were sensitive to streptomycin and kanamycin. These were more evident in the first than in the second trial. . .

CHAPTER VI SUMMARY AND CONCLUSIONS

The Rhizobium-legume symbiosis is a major process by which soil nitrogen can be supplied, especially in low input agricultural systems. Several factors limit its potential, including rhizobial nodulation dominance. Benefits from selected efficient Rhizobium inoculant strains are often not realized. The inoculum strain (s) fail to survive and establish so as to form the majority of the nodules. This occurrence is well documented. The exact mechanism by which one or a few strains, present in inocula or in the soil with resident rhizobia, form most of the nodules, is still obscure

Rhizobium strains in free living or symbiotic state accumulate trehalose, a nonreducing a, a-D-glucose disaccharide. Its role in either state is still speculative. It has been suggested that trehalose may play a role in the survival of rhizobia under stressing environmental conditions. This study was initiated to investigate possible relations between trehalose, survival of rhizobia under stress environment, and nodulation dominance

Five closely related Rhizobium strains, CE3, resistant to 200 ^ig ml'^ of streptomycin, and its Tn5 derivative mutants, resistant to both 200 [ig ml"^ each of streptomycin

101 102 and kanamycin, were screened and grouped based on their tolerance to salt stress. Tolerance to hyperosmolarity is often an indication of the ability to accumulate trehalose, as an osmoticum. Strain CE3 and CE3::Tnl2b are tolerant to

0.1, CE3::Tn514b and 16b to 1.0 and CE3::Tnl8a to 0.5 M

NaCl.

The strains were grown in defined medium under the following condition:

1) in defined medium supplemented with 0, 0.05, 0.1, and

0.2 M NaCl

2) in defined medium of pH 5, 6, 7, and 8.

3) strain CE3 and CE3::Tn514b were incubated under

temperature regimes of 10, 20, 30, and 40°C.

4) strain CE3 and CE3::Tn514b were also grown in defined

medium containing chloramphenicol at a rate of 6.5 jig

ml"’- in addition to their marker antibiotics.

After a growth period of 96 h the cells were harvested, trehalose and other sugars were analyzed by gas liquid chromatography. Other experiments were conducted to assess the effects of the above conditions when the strains were raised in YEM broth. The ability of the strains to grow on trehalose was also investigated.

Seedlings of Phaseolus vulgaris cultivar Midnight, supplied with 0.05 M NaCl or not, were inoculated with the above strains and raised in growth pouches in a growth chamber. Trehalose and other carbohydrates formed in the 103 nodules were extracted and analyzed. Dry weight of the plants and total nitrogen content of combined root and shoot samples were determined.

Nodules of salinized and nonsalinized plants were examined by both light and transmission electron microscopy.

Nodulation dominance studies were conducted using the same cultivar of beans inoculated with strains CE3 and

CE3::Tn514b in ratios of 1:0, 1:1, 2:1, and 0:1. The plants were grown as described above. Nodule occupancy was determined by taking advantage of their differential antibiotic resistance traits. Dry weight of plant parts

(roots and shoots) for the different treatments was determined. The ratio of the dry weight obtained from plants inoculated with the mixed inoculum to the weight obtained from the pure single strain inoculum multiplied by

100 gave the competitive ability of each strain.

When the strains were salt stressed the amount of trehalose decreased with increasing levels of salt, which was unexpected. This was attributed to cells having been harvested in the post-logarithmic phase. Arabinose was detected in the cells. Since it was a component of the medium, it is possible that the cells had a mechanism for its uptake which might have played a role in osmoregulation.

Energetically, uptake is a cheaper mechanism than synthesis.

The cells were able to grow in YEM broth supplied with 0.05

M NaCl whereas growth was seriously curtailed at 0.1 and 0.2 104 except for strain CE3::Tnl4b. All strains had shown tolerance to 0.1 M NaCl when grown on YEMA. This could have been a localized modification of the environment around the colonies after a few tolerant cells had grown.

Trehalose was detected in nodules formed by plants supplied either with saline or nonsaline medium. Although the differences observed among treatments were not statistically significant, nodules formed by strain CE3 had a 53% increase in trehalose and CE3::Tn514b had a 27% increase in trehalose when the plants were salinized.

Nonetheless, an insignificant F test and a decrease in the amount of trehalose with increasing salt levels in culture, diminishes the possible role of trehalose as an osmoticum.

Microscopic examination revealed that nodules formed by plants grown in salt media had reduced bacteroid tissues,

PHB granules, volutin and leghemoglobin . These effects were more pronounced in strain CE3 than in CE3::Tn514b.

Trehalose was not detected at any pH values used but arabinose was evident. No explanation could be given since a positive control was lacking; perhaps the enzymes of trehalose biosynthesis were affected by the hydrogen ion concentration and other associated deleterious phenomena, such as toxicity of microelements at different pH values.

Trehalose was neither detected when the cells were grown in a medium containing chloramphenicol nor when grown at raised temperatures. Again a positive control was :

105 lacking. It was deduced that trehalose was not involved in alleviating stress factors under the conditions of the experiments. Probably other mechanisms were employed or, as in other cases, growth was curtailed.

All the strains grew well in YET broth and on YETA.

This suggests that the strains had the enzymes for trehalose hydrolysis. If trehalose were present, it might serve as a carbon source for the strains under study.

The superiority of strain CE3 in accumulating greater amounts of trehalose, both in culture and in nodules, than

CE3::Tn5 was not evident in the nodulation dominance studies. Strain CE3::Tn514b was very competitive, especially when it was present in equal numbers. The most significant factor that determined nodule occupancy was the cell ratio. However, double nodule occupancy may have occurred.

From the aforementioned observations, it is concluded that

1) Trehalose may accumulate in rhizobia both in culture

and in nodules. Its role in osmoregulation was not

evident. It was not detected under adverse conditions

of pH, antibiosis, or temperature but positive controls

were lacking. Failure to detect trehalose does not

necessarily disprove its participation.

2) Trehalose can be a carbon source for rhizobia,

especially fast growers like those used in this study. 106

3) Screening rhizobia for salt tolerance as an indicator

for the trait of trehalose accumulation is not valid in

rhizobia as it is in Escherichia coli. Another method

should be sought. This also implies that it is not a

universal osmoticum in rhizobia.

4) There was no relationship between the amount of trehalose accumulated and strain representation in the

nodules. Strain CE3, which accumulated the highest

amount of trehalose, did not form the majority of the

nodules when coinoculated with CE3::Tn514b, in a 1:1

cell ratio assuming that double nodule occupancy was

minimal. Strain CE3::Tn514b exhibited higher

competitive ability ratings. In the absence of a make trehalose negative mutant (s) , it is impossible to

unequivocal conclusions concerning the critical level

of trehalose required for any given process. APPENDIX A

YEAST EXTRACT MANNITOL MEDIUM

Compound amount amount (g/L) (g/lOOml)

0.50 KH 2 PO 4 10.00 CaClj 4.00 0.20 0.05 Mg04 . 7 H 2 O 2.00 NaCl 4.00 0.10

F eCl3 . 6H20 1.00 0.01 Yeast extract 1.00 Mannitol 3.00 Agar 15.00 Water 1000

pH adjusted to 6.8 unless required otherwise. Agar was omitted if broth were needed.

107 APPENDIX B TREHALOSE MASS SPECTRA 109 Appendix B-2. Trehalose mass spectrum. A= trehalose mass spectrum on extract from CE3::Tn514b cells. B= trehalose spectrum of standard pure trehalose. Ill 112

APPENDIX C ANOVA TABLES

Table C-1. Accumulation of trehalose by Rhizohium leguminosarum biovar phaseoli cultured under varying salt conditions.

ANOVA (General Linear Models Procedure) Source df SS MS F Pr>F corrected 39 2038.29 total Model 19 2011.51 105.87 79.06 0.0001

Strain 4 535.28 133.82 99.94 0.0001

Salt 3 715.67 238.56 178.15 0.0001 Strain*Salt 12 760.55 63.38 47.33 0.0001 Error 20 26.78 1.34

Table C-2. Accumulation of araninose by Rhizobium leguminosarum biovar phaseoli cultured under varying salt conditions.

ANOVA (General Linear Models Procedure) Source df SS MS F Pr>F corrected 39 201.57 total Model 20 189.22 9.46 79.06 0.0001

Block 1 2.52 2.52 4.84 0.0480

Strain 4 56.22 14 .06 27.15 0.0001

Salt 3 23.89 7.96 15.39 0.0001 Strain*Salt 12 111.63 9.30 17.97 0.0001 Error 19 9.83 0.52 113

Table C-3. Accumulation of arabinose by Rhizobium leguminosarum biovar phaseoli strain CE3 and its Tn5 mutants under varying pH.

ANOVA (General Linear Models Procedure ) Source df SS MS F Pr>F Corrected 39 522.31 total

Model 20 508.02 25 . 40 33.77 0.0001

Block 1 0.01 0.01 0.02 0.895 Strain 4 284.91 71.23 94.71 0.0001

pH 3 59.57 19 .86 26.40 0.0001 Strain*pH 12 163.53 13.63 18.12 0.0001

Error 19 14.29 0 . 75

Table C-4. Effect of temperature on arabinose accumulation in strains CE3 and CE3: :Tn514b.

ANOVA (General Linear Models Procedure) Source df SS MS F P>F

Model 5 4.4974 0.8895 20.02 0.001 0.0001 strain 1 3.8578 3 . 8578 85.85 2.01 0.2149 Temperature 2 0 . 1803 0.0902 0.0507 Strain 2 0.4589 0 . 2294 5.11 *temperature

Contrast 1 0.4287 0.4287 9.54 0.0214 linear 114

Table C-5. Percent nodule occupancy by Rhizobium leguminosarum biovar phaseoli strain CE3.

ANOVA (General Linear Models Procedures) Source df SS MS F Pr>F corrected 29 8.083 total Model 11 7.758 0.705 39.020 0.0001 Block 2 0.128 0.064 3.540 0.0507

Salt 1 0.031 0.031 1.700 0.208

Ratio 4 7.288 1.822 100.81 0.0001 0

Salt*Ratio 4 0.310 0.077 4.29 0.0130 Error 18 0.3253 0.0181

Table C-6. Percent nodule occupancy by Rhizobium leguminosarum biovar phaseoli strain CE3: :Tn514b.

ANOVA (General Linear Models Procedures) Source df SS MS F P>F corrected 29 9.218 total Model 11 8.919 0.810 48.81 0.0001

Block 2 0.144 0.072 4.33 0.029

Salt 1 0.027 0.027 1.61 0.221

Ratio 4 8.405 2.101 126.51 0.0001

Ratio*Salt 4 0.343 0.086 5.16 0.006 Error 18 0.299 0.017 115

Table C-7, Dry matter yield (whole plants) as affected by inoculum ratio of Rhizobium leguminosarum biovar phaseoli strain CE3

and CE3 : : Tn514b.

ANOVA (General Linear Models Procedures) Source df SS MS F Pr>F corrected 29 152888.97 total Model 11 71587.70 6507.97 1.44 0.237

Block 2 10442.07 5221.03 1.16 0.337

Ratio 4 7532.47 1883.12 0.42 0.793

Salt 1 39530.70 39530.70 8.75 0.008

Ratio*Salt 4 14082.47 3520.62 0.78 0.553 Error 18 81301.27 4516.74 .

LITERATURE CITED

Abebe, A. 1987. Bean production in Ethiopia. Proc. Workshop on Bean Research in Eastern Africa, held at Mukono, Uganda on June 22-25. CIAT Ser. No. 2.

Adler, L., and Gustaffson, L. 1980. Polyhydric alcohol production and intracellular amino acid pool in relation to halotolerance of the yeast Debaromyces hansenii. Arch. Microbiol. 124:123-130.

Aguirre, J. A., and Miranda, H. 1973. Bean production systems. In: Potentials of field beans and other food legumes in Latin America. (Eds) D. Wall. CIAT series seminar No. 2e. pp 167-187.

Akkermanns, A. D. L., Danell, K. H., and Roelofsen, W. 1981. Enzymes of the tricarboxylic cycle and the malate aspartate shuttle in the nitrogen fixing endophyte of Alnus glutinosa. Physiol. Plant. 53:289-294.

Alexander, A. G. 1973. Studies on trehalase in Saccharum leaf and storage tissues. Plant Cell Physiol. 14:157-168.

Amarger, N. 1981. Selection of Rhizobium strains on their competitive ability for nodulation. Soil Biol. Biochem. 13:481-486.

Amarger, N. 1984. Evaluation of competition in Rhizobium spp. In: Current perspectives in microbial ecology, (eds) M. J. Klug and C.A. Reddy. Washington, D.C. Amer. Soc. Microbiol, pp 300-305.

Amarger, N., and Lobreau, J. P. 1982. Quantitative study of nodulation competitiveness in Rhizobium strains. Appl Environ. Microbiol. 44(3): 583-588.

Angle, J. S., Pugashetti, B. K., and Wagner, G. H. 1981. Fungal effects on Rhizobium japonicum- soybean symbiosis. Agron. J. 73:301-306.

Antonius, L. D., and Sprent, J. L. 1978. Growth and nitrogen fixation of Phaseolus vulgaris L. at two irradiances II Nitrogen Fixation. Ann. Bot. 42:399.

116 . .. .

117

D. V., Idler, K., and Chartain, N. Appelbaum, E. R. , Thompson, 1988. Rhizobium japonicum USDA 191 has two nodD genes that differ in primary structure and function. J. Bacteriol. 170:12-20.

Apte, R. and Ishwaran, V. 1974. Cultures of Rhizobium inoculants with those of Beijerinckia and Azobacter Proc Ind. Natl. Sci. Acad. Pt B. Biol. Sci. 40: 482-485.

Arayangkoon, T., Schomberg, H. H., and Weaver, R. W. 1990. Nodulation and nitrogen fixation of guar at high temperature. Plant Soil. 126: 209-213.

Arnold, W. N. 1968. The selection of sucrose as the translocate of higher plants. J. Theor. Biol. 21:13—20.

Arnold, W. N., 1979. Trehalose accumulation and turnover in Torulopsis glabrata. Curr. Microbiol. 2:109-112.

Bach. M. K., Magee, W. E., and Burris, R. H. 1958. Translocation of photosythetic products to soybean nodules and their role in nitrogen fixation. Plant Physiol. 33:118-124.

D. J., and Rolfe, B. G. 1984. Badenoch— Jones , J-, Flanders, Association of Rhizobium strains with roots of Trifolium repens. Appl. Environ. Microbiol. 49: 1511-1520.

Bakken, H. 1985. Cold hardiness in the alpine beetles, Patrobus septentrionis and Calanthus melanocephalus J. Insect. Physiol. 31: 447-454.

Bauer, W. D. 1990. Congress summary: The organismal perspective. In: Nitrogen fixation: Achievements and objectives. Proc. 8^^ Intern. Congr. on Nitrogen Fixation, Knoxville, Tennessee, USA, May 20-26. (Eds) P. M. Gresshoff, L. Evans Roth, G. Stacey, and W. E. Newton. New York. Chapman and Hall, pp 851-855.

Baust, J. G., and Lee, R. E. 1983. Population differences in anti-freeze cryoprotectant accumulation patterns in an Antarctic insect. Oikos. 40:120-124.

and Handelsman, J. 1989. Beattie, G. A., Clayton, M. K. , Quantitative comparison of the laboratory and field competitiveness of Rhizobium leguminosarum biovar phaseoli. Appl. Environ. Microbiol. 55 ( 11 ): 2755-2761

Ben-Amotz, A., and Avron, M. 1983. Accumulation of metabolites by halotorant algae and its industrial potential. Ann. Rev. Microbiol. 37:95-119. . .

118

D. E. 1979. Ultrastructure of Benson, D. R. , and Eveleigh, nitrogen fixing symbiont of Myrica Pennsylvaniaa (bayberoot) nodules. Bot. Gaz. 140: 515-521.

Bergesen, F. J. 1977. Physiological chemistry of dinitrogen fixation by legumes. In: A Treatise on Dinitrogen Fixation. Section III Biology. (Eds) R. W. F. Hardy and W. S. Silver. New York. John Wiley and Sons, pp 519-555.

Bernstein, L., and Ogata, G. 1966. Effects of salinity on nodulation, nitrogen fixation and growth of soybean and alfalfa. Agron. J. 58:201-203.

Bhagwat, A. A., and Reiser, D. L. 1990. Competitive defective mutants of Bradyrhizobium japonicum. In: Nitrogen fixation: Achievements and objectives. (Eds) P. M. Gresshoff, L. Evans-Roth, G. Stacey, and W. E. Newton. New York. Chapman and Hall, pp 251.

Bhumiratana, A., Anderson, R. L., and Costilow, R. N. 1974. Trehalose metabolism by Bacillus popilliae J. Bacteriol. 119:484-493.

Birch, G. G. 1963. Trehalases. Advanced Carbohydrate Chemistry. 18:201-225.

Birch, G. G. 1979. Chemical structure and sweetness of sugars. In: Developments in food sciences. New York. Elsevier North Holland, Vol. 2, pp 367-372.

Boonkerd, N., Weber, D. F., and Bezdicek, D. F. 1978. Influence of Rhizobium strains and inoculation methods on soybeans grown in rhizobia—populated soil. Agron. J. 70:547-549.

Boos, W., Ehmann, U., Bremmer, E., Middendorf, A., and Postma, P. 1987. Trehalase of Escherichia coli, mapping and cloning of the structural gene and identification of the enzyme as a periplasmic protein induced under high osmolarity conditions. J. Biol. Chem. 62:13212-13218.

Botsford, J. L. 1984. Osmoregulation in Rhizobium meliloti: inhibition of growth by salts. Arch. Microbiol. 137:124- 127

Bottomley, P. J. 1992. Ecology of Bradyrhizobium and Rhizobium. In: Biological nitrogen fixation. (Eds) G. Stacey, R. H. Burris, and H. Evans. London. Chapman and Hall, pp 293-348. .

119

Brana, A. F., Mendez, L., Diaz, L. A., Manzanal, M. B., and Hardisson, H. 1986. Glycogen and trehalose accumulation in Streptomyces antihioticus J. Gen. Microbiol. 132:1319- 1326.

Breedveld, M. W., Zevenhuizen, L. P. T. M., and Zehnder, A. J. B. 1990. Osmotically induced polysaccharide synthesis by Rhizohium meliloti SU-47. J. Gen. Microbiol. 136:2511-2519.

Broklebank, K. J., and Hendry, G. A. F. 1989. Characteristics of plant species which store different types of carbohydrates. New Phytol. 112:255-260.

Bromfield, E. S. P., and Gareth, D. J. 1980. Studies on double strain occupancy of nodules and the competitive ability of Rhizohium trifolii on red and white clover grown in soil and agar. Annal. Appl. Biol. 94:51-59.

Brown, A. D. 1976. Microbial water stress. Bacteriol Rev. 40:803-846.

Brown, A. D. 1978. Compatible solutes and extreme water stress in eukaryotic micro-organisms. Adv. Microbiol. Physiol. 17:181-242.

Brown, A. D., and Simpson, J. R. 1972. Water relations in sugar tolerant yeasts, role of intracellular polyos. J. Gen. Microbiol. 12: 589-591.

Bushby, H. V. A. 1982. Ecology. In: Nitrogen fixation. Vol 2 Rhizohium. (Eds) W. J. Broughton. Oxford. Clarendon Press, pp 35-75.

Bushby, H. V. A., and Marshall, K. C. 1977. Water status of rhizobia in relation to their susceptibility to desiccation and to their protection by montimorillonite . J. Gen. Microbiol. 99:19-27.

Cabib, E., and Leloir, L. F. 1958. The biosynthesis of trehalose in the locust fat body. Biochem. J. 78:531-536.

Callaham, D. A., and Torrey, J. G. 1981. The structural characterization of root hairs in Tri folium repens by Rhizohium. Can. J. Bot. 59:1647-64.

Carlson, T. A., Martin, G. B. and Chelm, B. K. 1987. Differential transportation of the two glutamine synthatase genes of Bradyrhizohium japonicum. J. Bacteriol. 169:5861- 5866.

Catley, B. J., and Kelly, P. J. 1975. Metabolism of trehalose and pullulan during the growth cycle of Aureolohasidium polluans. Biochem. Soc. Trans. 3:1079-1081. 120

Chatel, D. L., and Parker, C. A. 1972. Inhibition of rhizobia by toxic soil-water extracts. Soil Biol. Biochem. 4:289-294.

Chatel, D. L., and Parker, C. L. 1973. Survival of field grown rhizobia over the dry summer period in Western Australia. Soil Biol. Biochem. 5:145-423.

Chen, H., Gartner, E., and Rolfe, B. G. 1993. Involvement of genes on a megaplasmid in the acid-tolerant phenotype of Rhizobium leguminosarum biovar trifolii. Appl. Environ. Microbiol. 59 (4): 1058-1064.

M. Chen, H., Richardson, A. E., Gartner , E. A., Djordjevic, A., Roughley, R. J. and Rolfe, B. G. 1991. Construction of an acid tolerant Rhizobium leguminosarum biovar trifolii strain with enhanced capacity for nitrogen fixation. Appl. Environ. Microbiol. 57(7) 2005-2011.

Chhonkar, P. K., and Subba Rao, N. S. 1966. Fungi associated with legume root nodule and their effect on rhizobia. Can. J. Microbiol. 1253-1261.

Chien, C. T., Maundu, J., Cavaness, J., Dandurand, L. M. and Orser, C. S. 1992. Characterization of salt tolerant and acid sensitive mutants of Rhizobium leguminosarum biovar viciae strain C1204b. FEMS Lett. 90:135-140.

Chowdhury, M. S. 1977. Effects of soil antagonists on the symbiosis. In: Exploiting the legume-Rhizobium symbiosis in tropical agriculture. (Eds) J. M. Vincent, A. S. Whitney and J. Bose. University of Hawaii, Honolulu, College of Tropical Agric. Misc. Publ. 145, pp 385- 411.

Clegg, J. S., and Evans, S. R. 1961. The physiology of blood trehalose and its function during flight in the blowfly. J.

Expl. Bot . 38:771-792.

Clegg, J. S., Seitz, P., Seitz, W., and Hazlewood, C. F. 1982. Cryobiology. 19:306-316.

Clugston, J. R. 1993. Rhizobium spp poly-betahydroxybutyrate production and utilization in relation to soil survival. M. Sc. Thesis. University of Florida, Gainesville.

Cook, S. J., Khan, R., and Brown, J. M. 1984. A simple route to beta beta trehalose via trichloroacetidates . J. Carb. Chem. 3:343-348.

Cooper, J. E., Wood, M., and Bjourson, A. J. 1985. Modulation of Lotus pedunculatus in acid rooting solution by fast and slow growing rhizobia. Soil Biol. Bioch. 17:487- 492. . .

121

J., and Hong, Z. 1989. Enzymes of Copeland, L. , Vella, carbohydrate metabolism in soybean nodules. Phytochem. 28(1): 57-61.

Crabbe, L. H., Gussin, A. E. S., andTyrell, E. A. 1969. 2 Trehalase induction in BbcHIus csreus . Experimentia. - 25 : 193 9 ^ 0 .

Cregan, P. B., and Keyser, H. H. 1987. Influence of Glycine spp. on competitiveness of Bradyrhizobium japonicum and Rhizobium fredii. Appl. Environ. Microbiol. 54 ( 3 ): 803-808

Cregan, P. B., Keyser, H. H., and Sadowsky, M. J. 1989. Host plant effects on nodulation and competitiveness of Bradyrhizobium japonicum serotype strains constituting serocluster 123. Appl. Environ. Microbiol. 55 (10): 2532- 2536.

Crowe, J. H., and Crowe, L. M. 1984. Biological membranes. (Ed) D. Chapman. New York. Academic Press, Vol 5 pp 58-103. Chapman, D. 1984. Crowe, J. M., Crowe, L. M. , and organisms. The P]^ 0 servation of membranes on anhydrobiotic role of trehalose. Science. 223:701-703. of Csonka, L. N. 1989. Physiological and genetic responses bacteria to osmotic stress. Microbiol. Reviews. 53:121-

147 .

Cunningham, S. D., and Munns, D. N. 1984. The correlation between extracellular polysaccharide production and acid 48.1273— tolerance in Rhizobium. Soil Sci. Soc . Amer. J. 1276.

Cussein, A. E. F. 1972. Does trehalose occur in angiospermeae? phytochemistry. 11:1827-1828.

Danso, S.K.A., and Alexander, M. 1974. Survival of two strains of Rhizobium in soil. Soil Sci. Soc. Am. Proc.

38 : 86-89

Danso, S. K. A., Habte, M. and Alexander, M. 1973. Estimating the density of individual bacterial populations introduced into ecosystems. Can. J. Microbiol. 19:1450— 1451.

Dazzo, F.B., Napoli, C.A., andHubbell, D.H. 1976. Adsorption of bacteria to roots as related to host specificity in the Rhizobium clover symbiosis. Appl. Environ. Microbiol. 32:166-171. 122

Defaye, J., Driguez, H., and Hernrissat, B. 1983. Stereochemistry of the hydrolysis of alpha-alpha trehalose by trehalase determined by using a labelled substrate. Carb. Res. 124:265-273.

Delwiche, C. C. 1977. Energy relations in the global nitrogen cycle. AMBIO. 6:106-111.

P. Dinnbeir, U, Limpinsel, E., Schmid, R. , and Bakker, E. 1988. Transient accumulation of potassium glutamate and its replacement by trehalose during adaptation of growing cells of Escherichia coli K-12 to elevated sodium chloride concentrations. Arch. Microbiol. 150:348-357.

Ditta, G., Virts, E., Palamares, A., and Choong-Hyun, K. 1987. The nif A gene of Rhizobium meliloti is oxygen regulated. J. Bacteriol. 169:3217-3223.

Donnin, C., Puglisi, P. P., Vecli, A., and Marmiroli, N. 1988. Germination of Saccharomyces cerevisiae ascospores without trehalose mobilization as revealed by nuclear magnetic resonance spectroscopy. J. Bacteriol. 170:3789- 3791.

Dowling, D. N., and Broughton, W. J. 1986. Competition for nodulation of legumes. Ann. Rev. Microbiol. 40: 131-157.

Duncan, M. J., and Fraenkel, D. G. 1979. a ketoglutarate dehydrogenase mutant of Rhizobium meliloti. J. Bacteriol. 137:415-419.

Dunigan, E. P., Bollich, P. K., Hutchison, R. L., Hicks, P. M., Zaunbrecher, F. C., Scott, S. D., and Mowers, R. P. 1984. Introduction and survival of inoculant strains of Rhizobium japonicum in soil. Agron. J. 76: (1-3) 463-466.

Dutrieu, J. and Gourdoux, L. 1971. Trehalose and Glycogen development in Tenebrio molitor L. C.R. Seane Soc. Biol. 164:1994-1999.

Dwyer, M. R. 1986. The soluble carbohydrates of Euglena increased external gracilis , their identity and response to concentration. Plant Cell Env. 9:485-489.

Dylan, T., Helinski, D. R., and Ditta, G. S. 1990. Hyposmotic adaptation in Rhizobium meliloti requires beta(l- 2) glucan. J. Bacteriol. 172(3): 1400-1408. .

123

Eaglesham, A., Seaman, B., Ahamad, H., Hassouna, S., Ayanaba, A., and Mulongoy, K. 1981. High temperature telerant Current perspectives in nitrogen ' cowpea' rhizobia. In: Gibson, and W.E. Newton. Canberra. fixation . (Eds) A.H. Aust. Acad. Sci., pp 436.

S., and Siegers, L. K. 1983. Eaglesham, A. J. R. , Hassouna, on N fixation by cowpea and soybean. Fertilizer-N effects 2 Agron. J. 75:61-66.

A. D. 1983. Yeast water relations: Edgley, M. , and Brown, physiological changes induced by solute stress in Sscchsroinyces cBrsvisise and SscchsromycGS rouxii. J. Gen. Microbiol. 129:3453-3463.

Elbein, A. D. 1974. The metabolism of alpha-alpha trehalose. Adv. Carb. Chem. Biochem. 30:227-256.

El-Jander, J., and Fahraeus, G. 1971. Adaptation of Rhizobium to sub-arctic environment in Scandinvia. Plant

Soil spec, vol . 129-137.

Elkan, G. H., and Kuykendall, L. D. 1982. Carbohydrate metabolism. In: Nitrogen fixation. Vol 2 Rhizobium. (Eds) W. J. Broughton. Oxford Clarendon Press, pp 147-166.

Elkan G. H. 1984. Taxonomy and metabolism of Rhizobium and its genetic relationship. In: Biological Nitrogen Fixation, ecology, technology and physiology. (Eds) M. Alexander, New York. Plenum Press, pp 1-38.

fer A. S. 1967. Cultural El Essawi, T. M., and Abdel-Ghaf , and symbiotic properties of rhizobia from E^ptian clover {Trifolium alexandrinum) J. Appl. Bacteriol. 30:354-361.

Raper, C. D. Jr., and Fellows, R. J., Patterson, R. P. , Harris, D. 1987. Nodule activity and allocation of photosynthate of soybean during recovery from water stress. Plant Physiol. 84:456-460.

Vattone, M. A., Prado, F. E., and Fleischmacher , O. L., Sampietro, A. R. 1980. Specificity of sugar cane trehalase. Phytochem. 19:37-41.

V., Brom, S., Martinez, E., Pinero, D., Flores, M. , Gonzalez, and, Palacios, R. 1987. Romero, D., Davila, G. , Reiterated DNA sequences in Rhizobium and Agrobacterium spp. J. Bacteriol. 169:5782-5788.

Foulds, W. 1971. Effect of drought on three species of Rhizobium. Plant Soil. 35:655-667. 124

D., and Streeter, J. G. Francoise-Fougere, F., Le-Rudulier , and amino acid, organic acid 1991. Effects of salt stress carbohydrate composition and Plant Physiol.^‘r^®^°qg!-,558-l2396.1228 123. cytosol of Medicago sativa L. 1932. Root nodule Baldwin, I. L., and McCoy, E. Fred E. B., Fred, and leguminous plants. (Eds) E. B. bacteria University of Wisconsin Baldwin, and E. McCoy, Madison. Press, pp 73-102. Modulation and Wollum II, A. G. 1989. Fuhrmann, J., japonicum strains as competition among Bradyrhizobium and iron availability. influenced by rhizosphere bacteria 112. Biol. Fert. Soils. 7:108 1993. Effects of pH and T^ii-iihara S and Yoneyama, T. polyamine contents rn the soybean osmotii siress on cellular Rhizobium fredil, P220 and Bradyrhrzobiu™ ?hIzobia, Microbiol. 59(4). 1104 1109. japonicum A1017. Appl. Environ.

craJ^es ^;m"o;ic'^"re^fons^es I,' Sri!-!! reiMi!robl!i!" and gas exchange of plants J 1975. water balance in saline environments ""inier- salLe conditions. In: ^-ts and J. Gale. New York. (Eds) A. Poljakoff-Mayber Springer-Varlag, pp 108 185. and Dommergues, Y. R. 1985 G. Wey , J. Qannry F., Diem, H. , improves 1^2Nj fixation by Inoculation with Glomus mosseae 1 Fert.TToT-t- . Sols.Sn S- 1:15-23. field grown soybeans. Biol. Extracellular and Bradley, T. J. 1987. Garrett M. A., trehalose xn the accumulation if proline, serine, and brackish-water mosquitoes. J. haemolymph of osmoconf orming Exp. Biol. 129:2331-2338. The effects T. 4- 1 T T and Wona. M. N. 1978. on poly-B-hydroxybutyrate *^®of°!ge^' darkness and nitrate of Rhizobium in Lup levels' and nitrogen fixing ability 42. 420 423. august ifolius. Physiol. Plant. Kaasen, I., and Strpm, A. R. H. M. Sturvold, O. B., Giaever, , of Bioch4mical and genetic characterization nil Escherichia coll. J. osmoregulatory trehalose synthesis in Bacteriol. 170:2842-2849. .

125

1983. Ecophysiology of rihc^on A H and Jordan, D. C. ^ systems. In: Physiological plant ecology. nitroaen-fixing environment. Iesponses\o^the chemical and biological

B. Os.ond and H. Zielger. 302-390. Lrlin-Heldberg. Springer-Verlag, pp Wyatt, G. R. 1967. Trehalase Gilbv A. R., Wyatt, S. S. and discoidales: activation, from cockroach Blaberas and properties of the solnbnltatlon Acta Biochem.Pol. properties of the intestinal enzyme. 14:83-100. approaches: and Wilson, K. J. 1993. Past p-iiipr K E in tropical siccUs'and'Failures.rax In: Nitrogen fixation success anu J. Wilson, systems. (Eds) K.^ pE. Giller,uiiiei, auand K. cropping 215 235. Willingford, UK. CAB International, pp 1969. Sugar transport: 7 • rp Gavler. K. R. sugarcane. Planta. 85: olci?renc4 of'trehalaL activity in 299-302 1981. The uptake and i R and Dllworth, M. J. fast and slow growing species ‘^'hyiroiysis “"disaccharides by 129.233 239. of Rhizobium. Arch. Microbiol. Effects of Tsutsui, H., and Shimada, K. 1986. Onto C larval development an photoperiod and low temperature on conigrun L. freezing tolerance of Xestia Ent. Zool. 21:143 1 . (Lepidoptera-.Noctuidae) . Appl. sensitivities of the root- rp;^ham P H 1963. Antibiotic Scl. 16:557-559. nSSule bacteria. Aust. J. Biol. and 1981. Some problems of nodulation r-rahaTTi P H a "biotic iitrogen fixation in Phaseolus vulgaris L., review. Field Crops Res. 4: 93 113. 1977. Growth and development Graham P H. and Rosas, J. C. of iideierminate bush and Rhizobium. J. Agri . vulgaris L. inoculated with 88:503-508. T. and Viteri, S. E., Mackie, F., Vargas, A. P H among 1982. Variation in acid tolerance Palacios a' 5:121-128. s?r“nr;i ihlkobiur, phaseoll. Field Crops Res. 1965. wimpenny, W. J., and Mossman, M. R. Gray, C. T., bacteria II- Regulation of metabolism in facultative and nutrition of Kreb Effects of aerobiosis, anaerobiosis Bloch. Biophys. Acta. cycle enzymes in Escherichia coll.

117 : 33-44. .

126

Gussin, A. E. S., and Wyatt, G. R. 1965. Membrane-bound trehalase from Cecropia silkmoth muscle. Arch. Biochem. Biophys. 112: 626-634.

Ham, G. E., Cardwell, V. B., and Johnson, H. W. 1971. Evaluation of Rhizobium japonicum inoculants in_ soils containing naturalized population of rhizobia. Agron. J. 63: 301-303.

Ham, G. E., Lawn, R. J., and Brown, W. A. 1976. Influence of inoculation, nitrogen fertilizers and photosynthesis on field grown soybeans. In: Symbiotic nitrogen fixation. (Ed) P.S. Nutman. London, Cambridge University Press, pp 239- 253.

Hand, C. S., and Carpenter, J. F. 1986. pH-induced metabolic transition in Artenia embryo is mediated by a novel hysteric trehalase. Science. 232:1535-1537.

Harley, J. L., and Smith, S. E, 1983. Mycorrhizal symbioses. New York. Academic Press.

Hardy, R. W. F., Filner, P. and Hageman, R. H. 1975. Nitrogen input. In: Crop Productivity—Research and Imperatives. (Ed) A. W, A. Brown, T. C. Byerly, M. Gibbs, and A. San Pietro. Yellow Springs, Ohio. Charles F, Kettering Foundation, pp 177-200.

Hellriegel, H., und Wilfarth, H. 1888. Untersuchungen uber die stickstof fnahrung der gramineen and leguminosen. Beelageheit zu der Ztschr. Ver. Rubenzucker-Industrie Deutschen Reichs, pp 234.

Hernandez, B. S., Poth, M. and Focht, D. 0. 1987. Increased effectiveness of competitive Rhizobium strains upon inoculation of Cajanus cajan. Appl. Environ. Microbiol. 53: 2066-2068.

Herridge, D, F. and Roughley, R. J. 1975. Modulation and nitrogen fixation by Lablab purpureus under field conditions. Aust. J. Expl. Anim. Husb. 15: 264-. 269.

Hey, A., and Elbein, A. 0. 1968. Partial purification and properties of trehalase from Streptomyces hygroscopicus J. Bacteriol. 96:105-110,

Hey-Furguson, A., Mitchell, M., and Elbein, A. D. 1973. Trehalose metabolism in germinating spores of Streptomyces hygroscopicus. J. Bacteriol. 116:1084-1085. . .

127

W. P. and Ward, J. S. 1985. Hodgson, A. L. M. , Roberts, Regulated nodulation of Trifolium subterraneum inoculated with bacteriocin-producing strains of Rhizohium trifolii Soil Biol. Biochem. 17:475-478.

Hoelzle, A. I. 1990. Studies on the metabolism and accumulation of a, a, -trehalose in rhizobia and legume nodules. PhD Dissertation. The Ohio State University, Columbus

Hoelzle, I., and Streeter, J. G. 1990a. Stimulation of a- glucosidases from fast-growing rhizobia and Agrohacterium tumefaciens by K"^, NH'',^ and Rb*. Can. J. Microbiol. 25:223-

221 .

Hoelzle, I., and Streeter, J.G. 1990b. Increased accumulation of trehalose in rhizobia cultured under 1% oxygen. Appl. Environ. Microbiol. 56:3213-3215.

Y. 1966. Trehalase in conidia of Horikoshi, K. , and Ikeda, Aspergillus oryzea. J. Bacteriol 91:1883-1887.

Hottiger, T., Schmitz, P., and Wiemken, A. 1987. Heat- induced accumulation and futile cycling of trehalose in Saccharomyces cerevisiae. J. Bacteriol. 169:5518-5522.

Hubbell, D. H. 1981. Legume infection by Rhizobium: A conceptual approach. Bioscience. 31(11): 832-836.

Gresshoff, P. Howith, S. M., Udvardi, M. K. , Day, D. A., and M. 1986. Ammonia transport in free living and symbiotic sp. ANU 289. J. Gen. Microbiol. 132:257-261.

G. M. and Noma, A. T. 1982. Hua, S. S., Tsa, V. T. , Lichens, , Accumulation of amino acids in Rhizobium sp. strain WRIOOI in response to sodium chloride salinity. Appl. Environ. Microbiol. 35: 467-470.

Inoue, H. and Shimoda, C. 1981. Changes in trehalose content and trehalase activity during spore germination in fission yeast, Schizosaccharomyces pombe. Arch. Microbiol. 129:19-

22 .

Jardim-Freire, J. R. 1984. Important limiting factors in soil for the Rhizobium-legume symbiosis. In: Biological nitrogen fixation: Ecology, technology and physiology. (Ed) M. Alexander. New York. Plenum Press, pp 51-73.

Jennings, D. H. 1983. Some aspects of the physiology and biochemistry of marine fungi. Biol. Rev. 58: 423-459. .

128

Jones, R., and Giddens, J. 1983. Introduction of effective N -fixing rhizobial strains into soybean plants by use of fungicide resistance, Agron. J, 76:599-602.

Jordan, D. C. 1982. Transfer of Rhizobium japonicum Buchanan 1980 to Bradyrhizobium gen nov a genus of slow-growing root nodule bacteria from leguminous plants. Int. J. Syst. Bacteriol. 32:136-139,

Kang, U., Somasegaran, P., Heinz, J. H., and Bohlool, B.B. 1991. Symbiotic potential, competitiveness, and serological properties of Bradyrhizobium japonicum to Korean soils. Appl. Environ. Microbiol. 57: 1038-1045.

K. R. and Sinha, S. K. 1984. A Khan-Chopra, R. , Koundal , , simple technique of studying water deficit effects on nitrogen fixation in nodules without influencing the whole plant. Plant Physiol, 76:254-256.

Kandler, 0., and Hopf, H. 1980. Occurrence, metabolism and function of oligosaccharides. In: The biochemistry of 220- plants 3. (Ed) J. Preiss. New York. Acad. Press, pp 270.

P. H. 1969. Translocation of Keen, N. T. , and Williams, sugars into infected cabbage tissue during clubroot development. Plant Physiol. 44:748-754.

Kijne, J. W. 1992. The Rhizobium infection process. In: Biological nitrogen fixation. (Eds) G. Stacey, R. H. and H. J. Evans. New York. Chapman and Hall, pp 349-398

Salema, M. P. 1987. Effect of Kileo, R. O., Evers, G. , and rhizobial inoculation, nitrogen fixation and phosphorus fertilizers on nodulation and growth and yield of climbing fresh bean {Phaseolus vulgaris L.). In: Bean Research. Vol.2. (Eds) M, P. Salema, and A. N. Minjas. Papers pj-0 sented at a Workshop held at Sokoine University of Agriculture, Morogoro, Tanzania October 1-3, pp 210- 216.

Kishinevsky, B., Lobel, R. and Friedman, Y. 1984. Symbiotic evaluation of different peanut p 0 j-£Qrmance and efficiency Rhizobium strains under field conditions. Oleagineux 39:417-421.

Kitamoto, Y., Akashi, H., Tanaka, H., and Mori, N. 1988. Alpha-glucose-l-phosphate formation by a novel trehalose phoshorylase from Flammulina velutipes. FEMS Microbiol. Lett. 55:147-150. . . ,

129

Kouchi, H., and Yoneyama, T. 1986. Metabolism of C- labelled photosynthate in plant cytosol and bacteroids of root nodules of Glycine max Plant Physiol. 68:238-244.

Kouchi, H., and Nakaji. K. 1985. Utilization and metabolism of photoassimilated in soybean. Soil Sci. Plant Nutr. 31:323-334. of La Favre, A. K., and Eaglesham, A. R. J. 1987. Effects high temperatures and starter nitrogen on the growth and nodulation of soybean. Crop. Sc. 27:741—745.

Hunus, F. J., A. R. , Cantrell, M. A., Lambert, G. R. , Harker, R. A. and Evans, H. J. 1987. Russell, S. A. , Haugland, Symbiotic expression of cosmid-borne Bradyrhizobium

. Microbiol. j aponicum hydrogenase genes. Appl Environ. 53:422-428.

Larsen, P. I., Sydnes, L. K., Landfald, B., and Str0m, A. R. 1987. Osmoregulation in E. coli by accumulation of organic osmoiytes, betaines, glutamic acid and trehalose. Arch. Microbiol. 147:1-7.

W. D. 1982. Law, I. J., Yamamoto, Y., Mort, A. J., and Bauer, Nodulation of soybean by Rhizohium j aponicum mutants with altered capsule synthesis. Planta. 154:100-109.

Lawrie, A. C., and Wheeler, C. T. 1973. The supply of photosynthate assimilation to nodules of Pisum sativum L. in relation to fixation of nitrogen. New Phytol. 72:1341-1348. temperature and Lee, R. E., and Lewis, E. A. 1985. Effect of duration or exposure on tissue ice formation in the gall fly, Eurosta solidagenis (Diptera, Tephiritidae ) Cryo. Lett. 6:25-34.

S. K. Mattai, J., Shipley, G. G. Lee, C. W. B., Das-Gupta, , Abdel-Mageed, 0. H., Makriyannis, A., and Griffin, R- G. 1989. Characterization of L phase in trehalose—stabilized dry membranes by solid state NMR and X-ray diffraction. Biochemistry 28:5000-5009. and Lefebvre, Y. A., and Huber, R. E. 1970. Solubilization some properties of trehalase from honeybee, {Apis mellifera) Arch. Biochem. Biophys. 140:514-520. enhancement Lennox, L. B., and Alexander, M. 1981. Fungicide of nitrogen fixation and colonization of Phaseolus vulgaris 41:404- by Rhizobium phaseoli. Appl. Environ. Microbiol. 411. 130

G., Goas, G. and Hamelin, J. Le Rudulier, D. T. , Bernad, , 1984. Osmoregulation in Klebsiella pneumoniae: enhancement of anaerobic growth and nitrogen fixation under stress by J. proline betaine , r-betaine and related compounds. Can. Microbiol. 30:299-305.

Les Copeland, J. V., and Zhenquan, H. 1989. Enzymes of carbohydrate metabolism in soybean nodules. Phytochem. 28: 57-61.

Lewis, D. H. 1984. Storage carbohydrates. In: Vascular plants. London. Cambridge University Press, pp 153-163.

Lewis, D. H., and Harley, J. 1965. Carbohydrate physiology of mycorrhizal roots of beech II: Utilization of exogenous sugars by uninfected and mycorrhizal roots. New Phytol. 64:238-255.

Li, D. M., and Alexander, M. 1986. Bacterial growth rates and competition affect nodulation and root colonization by Rhizobium meliloti. Appl. Environ. Microbiol. 52:807-811.

Lie T. A. 1971. Symbiotic nitrogen fixation under stress conditions. Plant Soil Spec. Vol. pp 117- 127.

Londesborough, J., and Varimo, K. 1984. Characterization of two trehalases in baker's yeast. Biochem. J. 219:511-518.

W. F., and Ausbel, F. M. 1982. Cloning Long, S. R. , Buikama, of Rhizobium meliloti nodulation genes by direct complementation of Nod mutants. Nature ( Lond. ) . 298:485-488.

Torrey, J. G. 1984. Levels Lopez, M. F., Fontaine, M. , and of trehalose and glycogen in Frankia sp HFPAArl3. In: Advances in nitrogen fixation research. (Eds) C.Veeger, and W. Newton. Den Haag. Ni jkof f-Junk, pp 374.

Lopez, M. F., and Torrey, J. G. 1985. Purification and properties of trehalase in Frankia Arl3. Arch. Microbiol. 143:209-215.

Lopez, M. F., Whaling, C. S., and Torrey, J. G. 1983. The polar lipids and the free sugars of Frankia in culture.

Can. J. Bot . 61:2834-2842.

Lowendorf, H. S. 1980. Factors affecting survival of Rhizobium in soil. In: Advances in microbiology. Vol. 4. (Ed) M. Alexander. New York. Plenum Press, pp 87-124.

Mackay, M. A., Norton, R. S., and Borowitzka, L. L. 1984. Organic osmoregulatory solutes in Cyanobacteria. J. Gen. Microbiol. 130:2177-2191. .

131

Manhart, J. R., and Wong, P. P. 1979. Nitrate reductase activities of rhizobia and the correlation between nitrate reduction and nitrogen fixation. Can. J. Microbiol. 25:1169-1174.

Marechal, L. R., and Belocopitow, E. 1972. Metabolism of trehalose in Euglena gracilis 1 Partial purification and some properties of trehalase behaviour. J. Biol. Chem. 247:3223-3228.

Marshall, K. C. 1964. Survival of root nodule bacteria in dry soils exposed to high temperatures. Aust. J. Agric. Res. 15:273-281.

Marshall, K. C. 1967. Electrophoretic properties of fast and slow growing species of Rhizohium. Aust. J. Biol. Sci. 20:429-438.

Marshall, K. C. 1968. Nature of bacterium-clay interactions and its significance in survival of Rhizohium under arid conditions. Trans 9*"^ Int. Congress Soil Sci., Adelaide. 3:275 -280.

Marshall, K. C., and Roberts, F. J. 1963. Influence of fine particle materials on survival of Rhizohium_ trifolii in sandy soils. Nature (Lond.). 198:410-411.

Martin, F., Ramstedt, M., Soederhaell, K., and Ganet, D. 1988. Carbohydrate and amino acid metabolism in the ectomycorrhizal ascomycete Sphaerosporella hrunnea during glucose utilization. Plant Physiol. 86:935-940.

Martin, M. C., Diaz, L. A., Manzanal, M. B., and Hardisson, C. 1986. Role of trehalose in spores of Streptomyces FEMS Microbiol. Lett. 35:49-54.

Materon, L. A., and Weaver, R. W. 1984. Toxicity of Arrowleaf clover seed to Rhizohium trifolii. Agron. J. 76:471-473.

Matsumato, T., Yatazawa, M., and Yamamoto, Y. 1977. Distribution and change in the contents of allantoin_ and allntoic acid on developing nodulating and non-nodulating soybean plants. Plant Cell Physiol. 18:353.

May, S. N., and Bohlool, B. 1983. Competition among strains for nodulation of lentils Lens esculanta. Appl. and Env. Microbiol. 45(3): 960-965.

Mcbride, M. J., and Ensign, J. C. 1990. Regulation of trehalose metabolism in Streptomyces griseus spores. J. Bacteriol. 72(7): 3637-3643. 132

H. W., and Erdman, L. W. 1961. Means, U. M. , Johnson, Competition between bacterial strains effecting nodulation in soybeans. Soil Sci. Soc. Proc. 25:108.

Measures, J. C. 1975. Role of amino acids in osmoregulation in nonhalophilic bacteria. Nature (London) 257:398-400.

Meikle, A. J., Reed, R., and Gadd, G. M. 1988. Osmotic adjustment and accumulation of organic solutes in whole cells and protoplasts of Saccharomyces cerevisiae. J. Gen. Microbiol. 134:3049—3060.

Mellor, R. B. 1988. Distribution of trehalase in soybean root-nodules: Implications for trehalose metabolism. J. Plant Physiol. 133:173-177. determinant in Mellor, R. B. 1992. Is trehalose a symbiotic symbioses between higher plants and microorganisms? Symbiosis. 12:113-129.

Miller, R. H., and May, S. 1991. Legume inoculation: Successes and failures. In: The Rhizosphere and plant growth. (Eds) D.L. Keister and P.B. Cregan. Boston. Kluwer Academic Publishers, pp 123-134.

Moawad, H. A., Ellis, W. R. and Schidt, E. L. 1984. Rhizosphere response as a factor in competition among three serogroups of indigenous Rhizobium japonicum for nodulation 47:607- of field grown soybeans. Appl. Environ. Microbiol. 612.

Morell, M. K., and Coperland, L. 1984. Enzymes of sucrose breakdown in soybean nodules. Alkaline invertase. Plant Physiol. 74:1030-1034.

Coperland, L. 1985. Sucrose synthase in Morell, M. K. , and soybean nodules. Plant Physiol. 78:149-152.

Mosse, B. 1976. Plant growth responses to vesicular- arbiscular mycorrhiza. IX. Interactions between VA mycorrhiza, rock phosphate and symbiotic fixation. New Phytol. 76: 331-342.

Munro, P. M., Gauthier, M. J., Breittmayer, V. A., and Bongiovanni, J. 1989. Influence of osmoregulation pj-Qcesses on starvation survival of Escherichia coli in seawater. Appl. Environ. Microbiol. 55. 2017—2024.

Nakano, M., Brewer, C. F., Kasumi, T., and Hehre, E. J. 1989. Steric course of the hydrolysis of a, a-trehalose and a-D- glucosyl fluoride catalyzed by pig kidney trehalase. Carb. Res. 194: 139-144. 133

1990. Developmental biology of Hap, J. P., and Bisseling, T. The legume root nodule. a plant-prokaryote symbiosis: Science. 250: 948-965. Sharbough, C., and Mangum, C. Neetles, W. C. Jr., Parro, B., Trehalose and other carbohydrates in Anthronomous T 1971 during growth grandis.'Hellothls zea and HeJlothis virescens 17:657-675. knd development. J. Insect. Physiol. and Reddell, P. 1987. The distribution of Newman, E. R., plants. New l^colrhizas amongst the families of vesicular Phytol. 106:745—751. 1974. Changes Smith, J. E. and McIntosh, A. P. A M L during Carbohydrate composition and trehalase activity in continuous batch conidiation of Aspergillus niger in Soc. 63.57-66. culture. Trans. British Mycological 1941. Competition between Nicol, H., and Thornton, H.G. of nodule bacteria related strains 59. . Royal Soc. 130.32 infection of the legume host. Proc Trehalose Pankow, W., and Wiemkem, A. 1989. Niederer, M. , spruce . An indicator synthesis in mycorrhiza of No:^ay 19:13-20. viability. Eur. J. For. Pathol. Pankow, W., and Wiemken, A. 1992. Season! Niederer, M., mycorrhizas of Norway of soluble carbohydrates in changed to frost and spru5e and changes induced by exposure 22.291 299. desiccation. Eur. J. For. Pathol. favouring and da Costa, M. S. 1985. Factors Nobre, M. F., Debaromyce^ accumulation of arabinitol in yeast tL' 31:1061-1064. hansenii. Can. J. Microbiol. the cambial sap and Meier, H. 1967 Trehalose in Oesch, F., 6:1147-1148. of Fagus silvatica L. Phytochem. of ^ n r n and Pate, J. S. 1972. Exploration using nlLigei transport system of a nodulated legume ''^N. Planta. 104:35-39. yeasts. Adances in Food Onishi, H. 1963. Osmophilic Research 12:53-94.

T n Wriaht D.A. 1965. Rhizobial-induced japonicum in toybeL Ihlorosis toxin by Rhizobium chlorosis' 40:931-933. in pure culture. Plant Physiol.

Angle, J. S., and Wagner. 1982. Soil Paaashatti, B. K., japonicum. Scro-orCanisms antagonistic towards Rhizobium Soil Biol. Biochem. 14:45-49. 134

role Panek, A. D., and Bernardes, E. F. 1983. Trehalose. Its in gemination of Sscchsroniyces csrsvisisB. Curr. Genet. - 1 : 393 391 .

Parker, C. A., Trinick, M. J., and Chatel, D. L. 1977. Rhizobia as soil and rhizosphere inhabitants. In: A Sect. 4. Agronomy and r,]f 0 atise on dinitrogen fixation. ecology. (Eds) W. R. F. Hardy, and A. H. Gibson. New York. John Wiley and Sons, pp 311-352.

Parrish, F. W. 1987. Chromatographic procedures for isolating a, B, trehalose formed during the preparation of B- B-trehalose. Carbohydrate Res. 168:129-135.

Paschoalin, V. M. F., Silva, J. T., and Panek, A. D. 1989. Identification of an ADPG-dependent trehalose synthase in 16:81-87. Saccharomyces . Curr. Genet.

S. T. Rainbird, R. M., and Pate, J. S., Atkin, C. A., White, , of Woo, K. C. 1980. Nitrogen nutrition and xylem transport nitrogen in ureide producing grain legumes. Plant Physiol. 65:961-965. and Paul, E. A., and Clark, F. E. 1989. Soil microbiolgy biochemistry. (Eds) E. A. Paul, and F. E. Clark. San Diego. Academic Press, Inc., pp 26-29. labial Paulsen, R. 1971. Characterization of trehalase from glands of ants. Arch. Bioch. Biophys. 142:170-176.

Pedrosa, F. O., and Zancan, G. T. 1974. L-Arabinose metabolism in Rhizobium japonicum. J. Bacteriol. 119:336- 338.

J. C., Brendent, W. B. and Miller, R. Peterson, E. A. , Siros, Rhizobium W. 1983. Evaluation of competitive ability of meliloti strains for nodulation in alfalfa. Can. J. Microbiol. 29: 541-545.

Munns, D. N. 1987. Sensitivity of the Piha, M. I., and _ L. symbiosis to high soil common bean {Phaseolus vulgaris ) temperature. Plant Soil. 98: 183-194. Genetic Pinero, D., Martinez, E., and Slander, R. K. 1988. diversity and relationships amomg isolates of Rhizobium leguminosarum biovar phaseoli. Appl. Environ. Microbiol. 54:2825-2832.

B. G. 1985. Influence of Plazinski, J. , and Rolfe, Azospirillum strains on the nodulation of clover by Rhizobium strains. Appl. Environ. Microbiol. 49:984-984. . . . _

135

and Kubista, V. 1960. Metabolism of the Polacek, I., Bohemoslov. cockroach Periplsnsts swericsns Physiologia 9:288-234. free living J. R- 1981. Microbiolo^ of the Postqate, In: nitrogen fixing bacteria, excluding cyanobacteria. Current perspectives in nitrogen fixation. (Eds) A. Acad. Sci. Gibson, and W. E. Newton. Canberra. Aust. PP 217-228

1982. Genetics. In: The fundamentals of Postgate, J. R. Cambridge nitrogen fixation. (Ed) J. R. Postgate. London. University Press, pp 102-138. (Ed) J. Postgate. Postgate, J. 1987. Nitrogen Fixation. London. Edward Arnold (Publishers) Ltd.

Keizer, H. G. and Koolwijk, P. 1986. Postma, P. W. ^ t.yphiniuriuin. J. Transport o £ -{^j-Qjialose in Ssluionslls Bacteriol 168:1107-1111.

and Kubista, V. 1960. Metabolism of the Polacek, I., Bohemoslov. cockroach PBriplsnsta. amGrica.na.. Physiologia 9:228-234. (Eds). Advances in legume Polhill, R. M., and Raven, P. H. Gardens. 1981. systematics Pt I. Kew. Royal Botanical trehalose . in G., and Sharma. S.P. 1984 Changes Puri maculatus (FABR). levels in the Bruchid, Callosobruchus Insect Sci. Appl. 5:103-105. and Maheswari, R. 1978. Purification and Prasad, A. R. S . , of trehalose from the thermophilic fungus properties 528:162-170. Hannicola Ian uginosa. Biochim. Biophys. Acta. Fernandez, L., Quinto, C., De La Vega, H., Flores, _M., Soberon, G. and Placios, R. 1982. Ballado, T., Rhizobium Reiteration of nitrogen fixation gene sequences in 299:724-726. phaseoli. Nature (Lond. ) interactions in symbiotic Ouisoel A. 1988. Bacteria-plant niLogen fixation. Physiol. Plant. 74:783-790. and Gadd, G. M. 1987. Reed, H. R., Chudek, J. A., Foster, R., significance of glycerol accumulation in osmotic Microbiol. exponentially growing yeasts. Appl. Environ. 53:2119-2123. .

136

Reed, R. H., Borowitzka, L. J., Mackay, M. A., Chudek, J. A., Foster, R., Warr, S. R. C., Moore, D. J., and Stewart, W. D. P. 1986. Organic solute accumulation in osmotically stressed cyanobacteria. FEMS Microbiol. Rev. 39:51-56.

Reed, R. H., Richardson, D. L., Warr, S. R. C., and Stewart, W. D. P. 1984. Carbohydrate accumulation and osmotic stress in cyanobacteria. J. Gen. Microbiol. 130:1-4.

Reibach, P. H., and Streeter, J. G. 1983. Metabolism of of glucose ''^C labelled photosynthate and distribution enzymes of glucose metabolism in soybean nodules. Plant Physiol. 72:634-640.

Rennie, R. J., and Kemp, G. A. 1981. Dinitrogen fixation in Phaseolus vulgaris L. as affected by growth stage and temperature regime. Can. J. Bot. 60:1423—1437.

1983. N -fixation in field Rennie, R. J., and Kemp, G. A. 2 beans quantified by % isotope dilution. II Effect of cultivars of beans. Agron. J. 75: 645-649.

Ristau, E., and Wagner, F. 1983. Formation of novel anionic trehalosetetraesters from Rhodococcus erythrypolis under growth limiting conditions. Biotechnol. Lett. 5.95 100.

Robinson, A. C. 1969. Competition between effective and ineffective strains of Rhizobium trifolii in nodulation of Tri folium subterraneum. Aust. J. Agric. Res. 20:827-841.

Rockstein, M., and Srivastava, P. N. 1967. Trehalose in the flight of the housefly, Musca domestica L. in relation to age. Experimentia 23:636-637.

P. Cunningham, P. R. , and Clark, D. Rod, M. L., Alam, K. Y. , 1988. Accumulation of trehalose by Eschorichia coli K— 12 at high osmotic pressure depends on the presence of amber suppressors. J. Bacteriol. 170:3601—3610.

Rojas, R. R., Riemann, J.G., and Leopold, R. A. 1989. Diapause and overwintering capabilities of the larva of Homoeosoma electellum (Lepidoptera:Pyrallidae. ) . Environ. Entomol. 18:552- 557.

Primrose, S. B. 1979. Effect of glucose Ronson, C. W. , and and polyol metabolism by Rhizobium trifolii J. Bacteriol.

139 : 1075-1078

Herridge, D. F. 1976. Roughley, R. J., Blowes, W. M. , and Nodulation of Trifolium subterraneum by tropical rhizobia in competition with naturalized strains. Soil Biol. Biochem. 8:403-40. . .

137

Saadat, S., and Ballou, G. E. 1983. Pyruvated glycolipids from Mycobacterium smegmatis. J. Biol. Chem. 258:1813 1818. in the Sacktor, B. 1968. Trehalase and transport of glucose mammalian kidney and intestine. Proc. Acad. Sci. USA. 70:1007-1014. 1983. Sadowsky, M. J., Keyser, H. H., and Bohlool, B.B. characteriztion of fast-and slow-growing Biochemical _ rhizobia that nodulate soybeans. Int. J. System. Bacteriol 33: 716-722.

Streeter, J. G. 1988. Enzymes of a,a- Salminen, S. O. , and trehalose metabolism in soybean nodules. Plant Physiol. 81:538-541. Cary, NC, SAS/STAT. 1988. User's guide. SAS Institute Inc. USA. advances in Schmidt, E. L., and Robert, F. M. 1985. Recent research the ecology of Rhizobium. In: Nitrogen fixation progress. Current plant science and biotechnology in agriculture. (Eds) H. J. Evans, P. J. Bottomley, and W. E. Newton. Dordrecht. Martinus Nijhoff, pp 379-386.

K. and Overhoff, B. 1985. Schimtz, K. L., Irrgang, , Trehalose, a reserve disaccharide of Cellulomonas sp dependent DSM20108: Its identification, carbon source accumulation, and degradation during starvation. FEMS Microbiol. Lett. 40:333-337. Occurrence of Schubert, A., Wyss, P., and Wiemken, A. 1992. trehalose in vesicular arbuscular mycorrhizal fungi and mycorrhizal roots. J. Plant. Physiol. 140:41-45. R. Schwinghamer, E. A., Pandhurst, C. E., and Whitefield, P. triiolii. (1973). A phage— like bacteriocin of Rhizobium Can. J. Microbiol. 19:359-362. and L., Sa, M. H., Vargas, M. A. T. , Scotti, M. R. M. M. _ strains Dobereiner, J. 1982. Susceptibility of Rhizobium in to antibiotics. A possible reason for legume failure Cerrado soils. In: Biological nitrogen fixation for Eds) P. H. Graham and S. C. Harris. tropical agriculture. ( Papers presented at workshop held at CIAT, Cali, Colombia, March 9-13.

Segovia, L., Pinero, D., and Martinez-Romero, E. 1992. Genetic structure of a soil population of nonsymbiotic Rhizobium leguminosarum. Appl. Environ. Microbiol. 57(2): 426-433 . .

138

Sen, D., and Weaver, R. W. 1984. A basis of different rates of N -fixation by some strains of Rhizobium in peanut and cowpea root nodules. Plant Sci. Lett. 34:239—246.

Sengooba, T. 1987. Profile of the bean research programme and its impact on bean production in Uganda. Proc. of a workshop on bean research in eastern Africa held at Mukono- Uganda, 22-25, June. CIAT African Workshop, Series No. 2.

Shimakata, T., Tsubokura, F., Kusaka, T. and Shizukuishi, K. 1985. Mass spectrometric identification of trehalose-6- monomycolate synthesized by the cell free system of Bacterionema matruchotii Arch. Bioch. Biophy. 238:497-508.

Singh, C. S., and Subba Rao, N. S. 1979. Associative effect of Azospirillum brasilense with Rhizobium japonicum on nodulation and yield of soybean. Plant Soil. 53.387—392.

Tavares, J. W. 1986. Inoculation Singleton, P. W. , and response to the number and effectiveness of indigenous Rhizobium populations. Appl. Environ. Microbiol. 51(5): 1013-1018.

O. Filho, S. L. and Silva, C. L., Tincani, I., Branda, , , Faccioli, L. H. 1988. Mouse cachexia induced by trehalose dimycolate from Nocardia asteroides . J. Gen. Microbiol. 134:1629- 1633.

Singleton, P.W. and Bohlool, B.B. 1983. Effect of salinity on functional components of the soyhean-Rhizobium japonicum symbiosis. Crop Sci. 23:815—818. two Skrdleta, V. 1970. Competition for nodule sites between inoculum strains of Rhizobium japonicum as affected by delayed inoculation. Soil Biol. Biochem. 2:167—172.

Smith, L. T. and Smith, G. M. 1989. An Osmoregulatory dipeptide in stressed Rhizobium meliloti. J. Bacteriol. 171:4714-4717.

Hoben, H. J. 1985. Methods in legume Somasegaran, P. , and Rhizobiun technology. NifTAL Project and MIRCEN, University of Hawaii. Paia, Maui, Hawaii.

Somero, G. N. 1986. Protons, osmolytes and fitness of internal milieu for protein function. Amer. J. Physiol. 251:R197-R213 .

139

M. van Brussel, A. A. N., Glushka, Spaink, H. P., Sheeley, D. , P., J., York, W. S., Tak, T., Geiger, O., Kennedy, E. Reinhold, V. N. and Lugteberg, B. J. J. 1991. A novel highly unsaturated fatty acid moiety of lipo-oligosaccharide signals determines host specificity of Rhizobium. Nature (London). 354 :: 125-130 1983. Nodulation, N - Sparrow, S. D., and Ham, G. E. 2 fixation, and seed yield of navy beans as influenced by inoculum rate and inoculant carrier. Agron. J. 75:20-24. fixing Sprent, J. I., and Sprent, P. (Eds) 1990. Nitrogen organisms, pure and applied aspects. (Eds) J. I. Sprent and J. Sprent. London. Chapman and Hall. development Sprent, J. I., and Zehran, H. H. 1988. Infection and functioning of nodules under drought and salinity. In: jiLtrogen fixation by legumes in Mediterranean Agriculture. (Eds) D. P. Beck and L. A. Materon. Dordrecht. ICARDA/Martinus Nijhoff, pp 145-151.

Srivastava, P. N., and Rockstein, M. 1969. The utilization domestica. of trehalose during flight by the housefly Musca J. Insect. Physiol. 15:1181-1186. mass Stal, L. J., and Reed, R. H._ 1987. Low molecular carbohydrate accumulation in cyanobacteria from a marine Ecol. microbial mat in response to salt . FEMS Microbiol. 45:305-312. and other Stephens, D. 1967. The effect of ammonium sulphate fertilizers and inoculation treatments on beans {Phaseolus vulgaris). E.A. Agric. For. J. 32:411-417. Physiological Stowers, M. D., and Eaglesham, A. R. J. 1984. and symbiotic characteristics of fast growing Rhizobium japonicum. Plant Soil. 77:3-14. spp. Stowers, M. D. 1985. Carbon metabolism in Rhizobium Ann. Rev. Microbiol. 39:89-108.

K. and Steele, D. B. 1986. Stowers, M. D., Kuldarini, R. , Intermediary carbon metabolism in Frankia. Arch. Microbiol. 143:319-324. a Streeter, J. G. 1974. Growth of two soybean shoots on gj_ngle root. Effect of nitrogen and dry matter accumulation by shoots and on the rate of nitrogen fixation by nodulated plants. J. Expl. Bot . 25:189-198. 140

Streeter, J. G. 1979. Allontoin and allantoic acid in tissues and stem exudate from field grown soybean plants. Plant Physiol. 63:478-483.

Streeter, J. G. 1980. Carbohydrate in soybean nodules seedlings during the onset of nitrogen fixation. Plant Physiol. 66:471-476.

Streeter, J. G. 1982. Enzymes of sucrose, maltose and a-a- trehalose catabolism in soybean root nodules. Planta. 155:112-115.

Streeter, J. G. 1985. Accumulation of a-a-Trehalose by Rhizobium bacteria and bacteroids. J. Bacterid. 164: 78-84.

Streeter, J. G. 1987. Carbohydrate, organic acid and amino acid composition of bacteroids and cytosol from soybean nodules. Plant Physiol. 85:768-773.

Streeter, J. G. 1992. Analysis of apoplastic solutes in the cortex of soybean nodules. Physiologia Plantarum 84:584- 592.

M. E. 1976. Carbohydrates in Streeter, J. G. , and Bostler, soybean nodules: Identification of compounds and possible relationship to nitrogen fixation. Plant Sci. Lett. 7:321- 329.

S.O. 1988. Carbon metabolism Streeter, J. G. , and Salminen, and the exchange of metabolites between symbionts in legume nodules. In: Physiological limitations and genetic improvement of symbiotic nitrogen fixation. (Eds) F. 0. 11- Gara, S. Maniam, and J. J. Drevon. Dordrecht. Kluwer, pp

20 .

Strcm, A. R., Falkenberg, P., and Landfald, B. 1986. Genetics of osmoregulation in Escherichia coli: Uptake and biosynthesis of organid osmolytes. FEMS Microbiol. Rev. 39:79-86.

Sturtevant, D. B., and Taller, B. J. 1989. Cytokinin production by Bradyrhizobium japonicum. Plant Physiol. 89: 1247-1252.

Taylor, J.D., Day, J.M. and Dudley, C.L. 1983. The effect of Rhizobium inoculation and nitrogen fertiliser on nitrogen fixation and seed yield of dry beans {Phaseolus vulgaris). Ann. Appl. Biol. 103:419-429.

Thevelein, J. M. 1984. Regulation of trehalose mobilization by fungi. Microbiol. Rev. 48:42-59. ,

141

Thornton, F. C., and Davey, C. B. 1983. Acid tolerance of Rhizobium trifolii in culture media. Soil Sci. Soc. Amer. J 47:496-501.

Tisa, L., McBride, M. and Ensign, J. C. 1983. Studies on the morphology of Frankia strains EAN 1 pec, EVllc, CPllc and ACNl ag. Canad. J. Bot. 61:2768-2773.

K. Tomiyasu, I., Yoshinga, J., Kurano, F., Kato, Y. , Kaneda, Imaizumi, S., and Yano, J. 1986. Occurrence of a novel , , in aurantiacus, glycolipid trehalose, 2 3 , 6 -trimycolate Gordona aurantica. FEBS Lett. 203:239-242.

R. S. 1967. Auxin-cytokinin Torrey, J. G. , and Loomis, control of secondary vascular formation in isolated roots of Raphanus. Amer. J. Bot. 54:1098-1106.

Trinick, M. J. 1982. Eost-Rhizobium associations. In: Sydney Nitrogen Fixation in Legumes .( Eds ) J. M. Vincent. Australia. Acad. Press, pp 229-238.

Triplett, E. W. 1988. Isolation of genes involved in nodulation competitiveness from Rhizobium leguminosarum biovar trifolii Proc. Natl. Acad. Sci. USA. 85:3810-3814.

Triplett. E. W. 1990. Construction of a symbiotically effective strain of Rhizobium leguminosarum biovar trifolii. Appl. Environ. Microbiol. 50(1): 98-103.

T. M. 1987. Triflotoxin Triplett, E. W. , and Barta, production and nodulation are necessary for the expression of superior nodulation competitiveness by Rhizobium leguminosarum biovar trifolii strain T24 on clover. Plant Physiol. 85: 335-342.

S. L. Tate, M. E., Triplett, E. W., Lethbridge, B., Midland, , and Sims, J. J. 1988, Cloning of genes from Rhizobium leguminosarum biovar trifolii T24 responsible for production of trifolitoxin, an anti-rhizobial peptide involved in nodulation competitiveness. In: Nitrogen fixation: Hundred years after. (Eds) H. Bothe, F.J, de Bruijin, W.E. Newton, Stuttgart. Gustav Fischer, pp 788.

Turco, R. F., Moorman, T. B., and Bezdicek, D. B. 1986. Effectiveness and competitiveness of spontaneous antibiotic- resistant mutants of Rhizobium leguminosarum and Rhizobium japonicum. Soil Biol. Biochem. 18:259-262.

Udvardi, M. K., and Kahn, M, L. 1992. Evolution of the {Brady) Rhizobium legume symbiosis: Why do bacteroids fix nitrogen? Symbiosis, 14:87-101. 142

Overloop, K. Verbeke, M. , and van Laere, A., Francois, A., , van Gerven, L. 1987. Relation between germination, trehalose and the status of water in Phacomyces blakesleenus gporGs measured by proton~NRM. J. Gen. Microbiol. 133.239~ 245. van Laere, A., and Hendrix, P. 1983. Cyclic- AMP-dependent Yxiro activation of trehalase from dormant PhacomycGS hlakesleanus spores. J. Gen. Microbiol. 129:3287-3290. van Laere, A., and Siegers, L. K. 1987. Trehalose breakdown 41:247- in germinating spores of Mucor rouxii. FEMS Lett. 252. the Vannier, G. 1986. Increase of supercooling capacity in overwintering adults of the lacewing Crysoperla carnea (Insecta, Neuroptera) Neuroptera. Int. 4:71-82.

Veluthambi, K., Mahadevan, S., and Maheswari, R. 1981. Trehalose toxicity in Cuscuta reflexa Cell wall synthesis is inhibited upon trehalose feeding. Garb. Res. 70:686—688.

Verma, D. P. S. 1992. Signals in root nodule organogenesis and endocytosis of Rhizoblum. The Plant Cell. 4. 373 382.

Vesper, S. J., and Bauer, W. D. 1987. Role of pili (fimbriae) in attachment of Bradyrhizobium japonicum to soybean roots. Appl. Environ. Microbiol. 52. 134 141. symbiosis. Vidor, C. 1982. Microbial constraints to legume fixation for tropical agriculture. In- ' Biological nitrogen at a (Eds) P. H. Graham, and S. C. Harris. Papers presented workshop held at CIAT, Cali Colombia. March 9-13, pp 183- 194. of Vincent, J. M. 1970. A manual for the practical study Biological Programme, root-nodule bacteria . International Handbook No. 15. Oxford. Blackwell Scientific Publications, Ltd.

Vincent, J. M. 1980. Factors controlling the legume- Rhizobium symbiosis. In: Nitrogen fixation. Vol II. (Eds) University W. E. Newton, and W. H. Orme-Johnson . Baltimore. Press, pp 103- 163. nodule Vincent, J. M., and Waters, L. M. 1954. The root bacteria as factors in clover establishment in the red basaltic soils of Lismore district. New South Wales II. Survival and success of inocula in laboratory trials. Aust. J. Agric. Res. 5:61-67. 143

Warr, S. R. C., Reed. R. H., and Stewart, W. D, P. 1984. Osmotic adjustment of cyanobacteria: the effects of NaCl, KCl, and glycine betaine on glutamine synthase activity in a marine and halotolerant strain. J. Gen. Microbiol. 130:2169- 2175.

Waterbury, J. B., Calloway, C. D., and Turner, R. D. 1983. A cellulolytic fixing bacterium from the gland of Deshayes in

shipworm (Bivalvia:teredinidae) . Science. 221:140-143.

Weaver, R. W., and Frederick, L. R. 1974. Effect of inoculum rate on competitive nodulation on Glycine max L. Merr II Field Studies. Agron. J. 66:233-236.

Werner, D., and Morschel, E. 1978. Differentiation of nodules of Glycine max ultra-structural studies of plant cells and bacteroids. Planta. 14:169-173.

Westermann, D. T., Kleinkopft, G. E., Porter, L. K., and Leggett, G. E. 1981. Nitrogen sources for bean production. Agron. J. 73: 660-664.

Williams, L. E., DeJong, T. M., and Phillips, D. A. 1982. Effect of changes in shoot carbon-exchange rate on soybean root-nodule activity. Plant Physiol. 69:432-436.

Wilson, J. R. 1970. Response to salinity in Glycine :VI Some effects of a range of short term salt stresses on the growth, nodulation and nitrogen fixation of Glycine wightii.

Aust . J. Agric. Res. 21:571-582.

Wilkins, J. 1967. The effect of high temperatures on certain root-nodule bacteria. Aust. J. Agr. Res. 18:289-296.

Woese, C. R. 1987. Bacterial evolution. Microbiol. Rev. 51:221-271.

Winarno, R., and Lie, T. A. 1979. Competition between Rhizobium strains in nodule formation, interaction between nodulating and non-nodulating strains. Plant Soil. 51: 135-142.

Wollum, A. G., and Cassel, D. K. 1984. Spatial variability of Rhizobium japonicum in two North Carolina soils. Soil Sci. Soc. Amer. J. 48:1082-1086.

Wong, P. P., and Evans, H. J. 1971. Poly-B-hydroxybutyrate utilization by soybean {Glycine max Merr) nodules and assessment of nitrogenase activity. Plant Physiol. 47:750- 755. 144

clover and M., and cooper, J. E. 1985. Screening wood, aluminum. Soil Lotus rhizobia for tolerance of acidity and Biol. Biochem. 16:493-497. and and cooper, J. E. 1988. Acidity, aluminium, wood, M., mechanisms multiplication of Rhizobium trifolii: Possible 20: 95-99. of aluminium toxicity. Soil Biol. Biochem.

1967. The biochemistry of sugars and Wyatt G. R. 927-930 polysaccharides in insects. Adv. Insect. Physiol. 4:

C., Bowlus, R . D . , and Yancey, P. H., Clark, M. E., Hand, S. water stress: Evolution of Somero, G. N. 1982. Living with osmolyte systems. Science. 217:1214-1222. of sodium H., and Sprent, J. I. 1986. Effects Zehran, hair infection and chloride and polyethylene glycol on root nodulation of Vida faba L. plants by Rhizobium leguminosarum. Planta. 167:303 309.

Pueppke, S. G. 1988. Early infection Zdor, R E., and japonicum competition for nodulation by Bradyrhizobium and 1996-2002. 123. App. Environ. Microbiol. 54: BIOGRAPHICAL SKETCH

on December, Mary Christina Rwakaikara Silver was born born of Mr. and Mrs. Rwakaikara 24, 1948. She is the first of Hoima District, Uganda. in 1972, and an She obtained a B.Sc. (Agric) Hons, University. M.Sc. (Agric.) in 1981, from Makerere in the Department of She worked as a graduate assistant Master's research and Soil Science while conducting her students. teaching soil microbiology to undergraduate Works (Ltd) as an agronomist. Later, she joined Kakira Sugar in the Department She continued to teach soil microbiology on a part time basis. of Soil Science at Makerere University of Soil Science as a In 1986, she joined the Department research related to lecturer, charged to teach and conduct

soil microbiology. member of the In 1988, she was elected an executive member of the African Soil Science Society. She is a Biological Nitrogen Association of the African Society for of East Africa for Fixation (AABNF) and Soil Science Society joining the University which she acted as treasurer before

of Florida for a Ph.D. program. beloved Matthew Emmanuel Aijuka is her cherished and

son.

145 I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy.

Professor of Soil and Water Science

I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy.

Jetry b\ Sartain, Cochairman Professor of Soil and Water Science

/f^. V O^J David Sylvia Professor of Soil and Water Science

Kenneth L. Buhr Assistant Professor of Agronomy I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy.

?d,J ^Robert Stall Professor of Plant Pathology

This dissertation was submitted to the Graduate Faculty of the College of Agriculture and to the Graduate School and was accepted as partial fulfillment of the requirements for the degree of Doctor of Philosophy.

April 1994

Dean, College of Agriculture

Dean, Graduate School