UNIVERZA V LJUBLJANI

TABLE OF CONTENTS / KAZALO VSEBINE

p. / s.

Key words documentation with thesis abstract 3

List of figures / kazalo slik 6

List of tables / kazalo tabel 8

Abbreviations and symbols / Okrajsave in simboli 9

CHAPTER I: THESIS INTRODUCTION AND LITERATURE REVIEW 10

CHAPTER II: EFFECT OF PLANT GROWTH PARAMETERS ON NODULATION OF THE SPLIT-ROOT SYSTEM OF SOYBEAN (Glycine max L., Merr., cv. D-68) BY Bradyrhizobium japonicum strain USDA 110.

Abstract 17

Introduction 19

Material and methods 21

Results 27

Discussion 38

CHAPTER III: PHOTOSYNTHATE PARTITIONING AND AUTOREGULATION OF SOYBEAN (Glycine max L., Merr) NODULE DEVELOPMENT.

Abstract 40

Introduction 42

Material and methods 44

Results 49

Discussion 60

CHAPTER IV:RELATIONSHIP BETWEEN COMPETITION PATTERN AND THE RATES OF NODULE FORMATION BY THE TWO STRAINS OF Bradyrhizobium japonicum ON A SPLIT-ROOT SYSTEM OF SOYBEAN (Glycine max, L., Merr.)

Abstract 63

Introduction 65

Material and methods 67

Results 72

Discussion 78

POVZETEK 81

ACKNOWLEDGEMENTS 89

REFERENCES 90

APPENDICES 100

CHAPTER I

THESIS INTRODUCTION AND LITERATURE REVIEW

The symbiotic association between rhizobia and leguminous plants is formed through a complex sequence of interactions between the partners (for review see 21, 68) and culminates in the establishment of nitrogen fixing nodules. On the basis of present knowledge, three distinct stages of these interactions are recognized:

1. The preinfection stage, comprises the interactions in the rhizosphere/rhizoplane, controlling the attachment of bacteria to the root surface and initiation of infection. These interactions include: a) proliferation of rhizobia in the rhizosphere of their respective hosts and chemotaxis towards defined regions of the roots, stimulated by nutrients and phenolic compounds in the root exudate (7, 45); b) root colonization and adsorption to the root surface in a non specific (67, 83) or specific manner, mediated by plant lectin and bacterial surface polysaccharide molecules (5, 15, 41, reviewed in 25, 29); c) induction of bacterial nodulation (nod) genes by flavonoids released from legume seeds and roots (28, 29, 47, 57, 62, reviewed in 21, 68). Common nodABC genes are involved in generating extracellular factors, inducing root hair curling, branching, initiation of the infection threads and proliferation of centers of cortical cell division in the host plant. Another set of nod genes is responsible for host specific nodulation (hsn genes) and determines the host specificity among rhizobia. Expression of common and hsn nod genes is controlled by the interaction between the regulatory nod D gene and plant flavonoids (see 74 for compilation and 21, 68 for review on nod genes).

2. The infection stage encompasses the interactions within the roots, controlling nodule initiation and subsequent nodule development until the onset of nitrogen fixation. Nodule formation involves coordinated expression of rhizobial nod genes and plant symbiotic genes coding for nodule specific proteins, termed nodulins (12, 36, 37, reviewed in 21, 39). Even before the infection thread penetrates the root cortex, cortical cells begin to divide (79), giving rise to an initial nodule meristem. Further nodule ontogeny includes enlargement of the nodule tissue and differentiation of the vascular system connecting nodule meristem to the root vascular system. Infection thread ramifies intercellulary in the nodule meristem tissue, eventually penetrates the walls of adjacent plant cells and rhizobia multiplying in the infection thread are released into the host cytoplasm. Within the plant cells rhizobia become enclosed in plant derived peribacteroid membrane and differentiate into nitrogen fixing bacteroids. Infection and nodule initiation and development may vary in different legume species (for reviews see 25, 29, 63, 68, 76).

3. The nitrogen fixation stage encompasses interactions within the nodules associated with the onset and efficiency of nitrogen fixation. The synthesis of nitrogenase, the key enzyme converting atmospheric nitrogen into ammonia, is encoded by rhizobial nif or fix genes (63). Ammonia is then assimilated by plant enzymes and exported from nodules in the form of amides or ureides, depending on legume species (for reviews see 29, 32, 64, 65). Rhizobial strain and environmental factors affecting plant growth and nitrogen assimilation determine the efficiency of nitrogen fixation in nodules ((2, 3, 29, 51, 71, 72, 84).

A distinct physiological phenomenon, occurring in the infection stage, which appears to be strictly host controlled is a process termed autoregulation (reviewed in 21, 39, 68). By autoregulation the host plant controls the number of nodules formed on the roots and prevents overnodulation. Autoregulation has attracted considerable research attention over the past 50 years, since the process appears responsible for competition between Rhizobium strains for nodule occupancy on the common host. However, the mechanism(s) of autoregulation remain(s) as yet unknown.

In his classical experiments with red clover Nutman (58 - 60) postulated physiological homology of lateral roots and nodules and envisaged the sites of lateral root initials as predetermined foci of infection (58). Studying the effect of delayed inoculation on nodulation (59), he concluded that number of nodules is determined by the number of these preinfection foci and therefore the increased nodule number obtained with delayed inoculation could be attributed to increased number of preinfection foci on a larger root system. Increased nodulation (60), following nodule and root tip excision, led Nutman to the proposal of the regulatory mechanism based on production of inhibitors by nodule and root meristems. According to his hypothesis, a significant increase in nodule number following the excision of root tips and nodules formed by the effective strain was due to the removal of the source of inhibitor. Consequently, ephemeral nodules formed by the ineffective strain do not produce considerable amount of inhibitor since their excision did not stimulate subsequent nodulation. Nutman (59) has also pointed out that only certain regions of the roots are susceptible to infection by rhizobia. Localized and transient susceptibility of legume roots to infection has been extensively studied by Bhuvaneshvari et al. (9, 11) and shown to be widespread among common legumes (10).

Recent studies on autoregulation employing time separated inoculations (i.e. early inoculation followed by delayed inoculation at various time intervals) on the intact or split - roots of soybean and alfalfa (see 21, 39, for review) and clover (69) have demonstrated systemic mechanism of the autoregulatory response. Two concepts of regulatory substances with regard to the site of their production were generated:

a) - a shoot derived inhibitor, production of which is induced by primary (early) inoculation and its effects manifested as a suppression of nodule development in the secondary (delayed) inoculated root portion (19, 61); b) - an inhibitor is produced at the site of primary inoculation - i.e. in the root - and then transported to developmentally younger regions of the root (19, 66) or, as the case may be, across the split-root system (e.g. 49).

Delves et al. (27) demonstrated nodulation may be controlled by shoot and root factors.

Kosslak (46), using a split root system, has tested regulation of nodulation over a variety of soybean cultivar/B. japonicum strain combinations. She observed suppression of nodule development on delayed inoculated root side even when a less competitive or ineffective strain was used as a primary inoculum. Maximum autoregulatory response was obtained with 4 - 7 day delay of secondary inoculation. Preexposure of young seedlings to a less competitive or ineffective (nod+, fix-) strain for 6 to 72 hours, indicated that, first nodules, formed by a less competitive strain inhibited nodulation by a more competitive strain (48).

Experiments of Pierce and Bauer (66) and Malik and Bauer (52) indicated autoregulatory response in young tap root between 6 - 15 hours after primary inoculation.

Microscopic analysis of nodule initiation and development in soybean by Calvert et al. (24) revealed that, regulation takes place during transition of infections into nodule primordia. Infections were defined as centers of subepidermal cortical cell divisions with associated infection threads. Cortical cell division centers without infection threads in the adjacent root hairs were called pseudo infections. On a single tap root they could identify up to 50 fold greater number of cortical division centers than the average number of mature nodules ultimately formed on the root. Similar observations were reported by Mathews et al. (55). Microscopic analysis of double inoculation experiments (15 h apart) by Calvert et al. (24) showed that, 3 days after the second inoculation there was no evidence of suppression of either number or maturation of late induced infections. However, 7 days after the second inoculation development of infections in younger root regions was arrested at stages after nodule meristem formation but before the emergence of nodule primordia, which appear as bumps on the root surface.

Experiments with host and Rhizobium mutants have demonstrated, that nodulation phenotype, as well as, the intensity of autoregulatory response is controlled by plant and bacterial genotype (11, 55, 61) their interaction (42) and inoculant titer (66, 77). Furthermore, nodulation is affected by several physiological and environmental factors such as nitrate levels (53, 61) and light (49, 54). Malik et al. (54) observed inhibitory (nonphotosynthetic) effect of light on number of infections and stimulatory effect (photosynthetic) on subsequent development of those infections. Kosslak and Bohlool (49) demonstrated that, number of nodules on the early/delayed inoculated split-root system was proportional to the amount of light available to soybean plant for photosynthesis.

Direct dependence of nitrogen fixation in nodules on carbohydrate supply from photosynthesis (51, 84), as well as, selective partitioning of current photosynthate to effective and ineffective nodules (71) have been clearly demonstrated. Differential partitioning of labeled photosynthate to roots and nodules in early stages of the infection process has so far not been reported. Similarly, no direct experimental evidence of the nature or transport of the inhibitory substance(s) involved in regulation of nodulation has been obtained. Since in early developmental stages, nodules and bacteria are nutritionally completely dependent on the host plant, selective abortion of infections in soybean, observed by Calvert et al. (24) might be compared to selective abortion or abscission of flowers and fruits in higher plants (17, 86), where the process is under nutritional and hormonal control.

Exogenously supplied phytohormones (IAA, GA3, ABA, CCC) were shown to inhibit nodulation depending on nitrogen supply to the plant (see ref 39 for review). Production of phytohormones by rhizobia has also been demonstrated (44, 75). Bauer et al. (6) showed that cytokinins can induce cortical cell divisions in the absence of rhizobia. The role of phytohormones in regulation of nodulation, however, remains unclear.

By contrast to autoregulation, which is generally recognized as a plant controlled response, determinants of the interstrain competition pattern on the common host are generally associated with the rhizobial strain attributes. It has been proposed (1) that, relative numbers of cells of homologous of rhizobia in soil or in the inoculum mixture determine relative numbers of nodules formed by competing strains. Whereas environmental factors affecting persistence (16,88) and performance (82) of rhizobia may influence relative numbers of rhizobial strains in soil, when the strains are added to roots in equal cell numbers, speed of nodule formation by competing strains may become the key determinant of the proportion of nodules occupied by each strain. Smith and Wollum (73) and McDermot and Graham (56) examined the relationship between nodulation rate and competitiveness of several strains and obtained somewhat inconclusive results.

Motility and chemotaxis, though not essential for infection, may provide competitive advantage to a strain colonizing the root surface (7, 23). The rate of infection and nodule formation may be influenced by the response of different strains to plant symbiotic signals in the root exudate (40). Plant and rhizobial symbiotic signals acting in concert and affecting nodule initiation and development may vary considerably with plant and rhizobial genotype or with the combination of the two (34, 40, 57, 81). How the early host/strain molecular interactions affect interstrain competition is unknown.

Since the genetic and biochemical basis of autoregulation and interstrain competition remains unknown, further hystochemical and physiological studies may provide indirect evidence for the underlying mechanisms.

We used sequential inoculation of the split-root systems with the same strain or, simultaneous inoculation with two strains of different competitiveness, in soybean plants, grown under different environmental conditions to determine: 1) in what stage(s) of the infection process regulation of nodulation takes place; 2) how the host plant growth potential and root infectible area affect nodule initiation and development; 3) whether selective partitioning of current photosynthate to developing nodules and roots provides a regulatory mechanism, controlling nodule number and mass per plant; 4) whether the outcome of interstrain competition is determined during the early stages of root infection and nodule initiation or, during the process of nodule development.

MATERIAL AND METHODS

Growth systems : Two variants of a split-root procedure, described by Singleton (70) were used:

1) A split-root growth system, shown in Figure II-1 was used in the time course experiment. Two PVC columns, supported by PVC couplers were taped together. Bottoms of the columns were sealed with Parafilm and columns filled with dry horticultural vermiculite. Plastic elbows (90 deg. angle, 1/2" diam.) with a planting hole drilled in the angle center were used to direct split roots into columns. Drainage tubes were inserted 0.5 cm above the bottom in each column. The PVC parts were sterilized prior to assembly with 2.5% sodium hypoclorite and rinsed with H₂O.

2) A split-root growth system shown in Figure II-2 was used in all other experiments described in this chapter. Two growth pouches (Northrup King Co.) were stapled together at the top. Planting troughs were separated from the wicks along the existing perforation and the wicks were shortened by 1.5 cm by folding. A vertical cut 1.5 cm long (Fig. II-2 -D) was made through both bags in the middle of the top edge of the pouches. A single trough (Fig. II-2 -B) was passed through the cut so that one half of the trough was in each bag. A strip (half of the second trough - Fig. II-2 -A) was inserted into each bag to make a capillary connection between the trough (Fig. 11-2 -B) and the main wick (Fig. 11-2 -C) and at the same time direct root growth into the bags.

Planting procedure: Seeds of soybean (Glycine max L., Merr.) cv. D-68 (T.E. Carter, Dept. of Crop Science, NC State University at Raleigh) were surface sterilized with 2.5 % NaClO for 5 minutes, rinsed 7 times with sterile H₂O, imbibed for 4 hours and then sown hillum down in moist, sterile horticultural vermiculite. Approximately 48 hours after sowing, uniform seedlings (1.5 to 2.5 cm radicle) were selected, the tips of the radicles cut off and seedlings planted into the center of the trough or into a hole in the elbows packed with wet vermiculite. Elbows were planted in a layer (approx. 5 cm deep) of sterile vermiculite. After planting, pouches and elbows were covered with transparent polyethylene film to provide sufficient humidity for lateral root growth under growth room conditions. Three to 5 days after planting lateral roots had grown a few cm into pouches or emerged from the elbows. At that point, seedlings were selected for uniformity and roots trimmed to the same number of lateral roots per side; strip connectors between the trough and the wick were removed and the elbows were attached to the PVC columns containing vermiculite moistened with 50 ml N-free plant nutrient solution (PNS). Tops of the columns were sealed with Parafilm. Seedlings in pouches received 30 ml PNS at planting and subsequently, PNS level was maintained at 1-3 cm from the bottom of the pouch with half strength PNS. Plants in PVC columns were irrigated every other day with half strength PNS via microtubing inserted through the top Parafilm cover. Concentrations of nutrients in PNS were: 0.58 mM CaSO₄·2H₂O, 0.5 mM K₂HPO₄, 0.25 mM MgSO₄·7H₂O; and concentrations of micronutrients (added as preformulated Hawaiian Horticulture Mix) were 51 μM Mg, 97 μM S, 40 μM B, 0.6 μM Co, 2.9 μM Cu, 33.3 μM Fe, 10 μM Mn, 0.5 μM Mo, 9.4 μM Zn.

Plants in pouches were grown in the growth room under photosynthetically active radiation (PAR) 300-400 μE/m²/sec, 18 h photoperiod and temperature range 22 - 28°C. Plants in columns were grown in the greenhouse under PAR 1350-1600 μE/m²/sec, approximately 13 h photoperiod and temperature range 18 - 37°C.

Inoculation treatments: Bradyrhizobium japonicum strain USDA 110 (TAL 102, obtained from the Niftal Project collection) as peat based inoculant was used in all the experiments described in this chapter. Peat inoculant was suspended in N-free PNS so that 2.5 x 10⁷ cells was applied per root side - in 2 ml of inoculum per pouch and in 30 ml of inoculum per PVC column. Cell density in PNS was determined by drop plate method (43).

Half root portions were first inoculated (early inoculation) 8 days after planting, when the roots had grown to the bottom of the pouches or columns and at least one trifoliate leaf had emerged on the shoot. Inoculation of the other root half was delayed as indicated in the results for the individual experiments, following the same inoculation procedure as for early inoculation.

Harvest: Plants were harvested as indicated for the individual experiments in the results section. Shoots, roots and nodules were separated and dried at 65°C prior to weighing.

Measurements of leaf area and root length: Length and width of individual leaves was measured at 24 h intervals and leaf area calculated as a surface of elypse. Net increase in length of tap and lateral roots was measured at the same intervals as leaf area. Each time, root tips were marked with the pen on the surface of the pouches, using different pen colors for successive markings.

Counts of nodule primordia and early nodules: Emerging nodule primordia and nodules in pouches were counted on the colony counter.

Statistical analysis: Data were analyzed by Duncan’s multiple range test, paired t-test, Tukey’s HSD test and by regression where applicable, using SYSTAT statistical package (87). At least four replicates per treatment were included in the analysis.

Determination of nitrogenase activity: To detect the beginning of N₂ fixation, extra plants for each treatment were tested for nitrogenase activity at 24 h intervals from 7 to 10 days after inoculation. Half root systems were placed into 100 ml test tubes. Tubes were injected through a serum stopper with 5 ml acetylene and ethylene production was determined by gas chromatography (Varian 940 GC).

RESULTS

Time course of inoculation delay and suppression of nodulation on delayed inoculated root side.

Inoculation delay for 16 hours or more significantly reduced nodule weight on delayed inoculated (D) side compared to early inoculated (E) side, whereas nodule numbers on D side were not significantly affected until 64 hour inoculation delay (Figure II-4). In the uninoculated/delayed inoculated treatment nodule numbers increased progressively with inoculation delay, while nodule weight at harvest decreased slightly with inoculation delay. Shoot weight in the uninoculated/delayed inoculated treatment was also reduced compared to early/delayed inoculated treatment (Figure II-5 B), while root weight was not significantly affected by the inoculation treatments (Figure II-5 A).

Correlation between leaf area, root length and nodulation

Nodule scores 3 weeks after delayed inoculation provide no insight into the pattern of nodule development on the early and delayed inoculated root side. Therefore, another experiment was set up to monitor nodule emergence and subsequent development during 2 weeks after inoculation. Nodule numbers were related to root infectible area (root length) and leaf area from the time of inoculation until emergence of first visible nodules (nodule primordia). Number of nodules per plant at the time of their emergence (5 days after inoculation) and at harvest (14 days after inoculation) was better correlated with leaf area than with root length (Table II-1, Figure II-6 A, B). Rate of nodule development was also much higher correlated with leaf area during first 5 days after inoculation than with root length during the same period (Table II-1, Figure II-7 B).

Effect of delayed inoculation on nodulation before and after the start of nitrogen fixation in early nodules.

In plants grown in pouches (Figure II-2), nitrogenase activity (acetylene reduction) in nodules on E side was first detected 10 days after inoculation. To evaluate the effect of nitrogen fixation on regulation of nodulation we used 4 and 14 day inoculation delay and monitored nodule development on the early and delayed inoculated side over 21 days from delayed inoculation.

When D side was inoculated 4 days after E side, nodule numbers on E and D side increased on successive scoring dates (Table II-II) but calculated suppression of nodulation on D side (Table II-II), for each scoring date, diminished significantly. When D side was inoculated 14 days after E side, suppression of nodulation on D side was considerably reduced (Table II-III), compared to treatment where D side was inoculated 4 days after E side (Table II-II).

Removal of E side at the time of D inoculation, significantly increased the number of nodule primordia formed on the remaining root half (D side), compared to total number of nodule primordia on the intact E/D split-root system. However, at harvest, plants with only half root system (D side) had essentially the same number of mature nodules as plants with both root halves attached (Table II-III).

Effect of light intensity on nodulation of the early/delayed inoculated split-root system.

When plants with early/delayed inoculated split-root system, using 4 day inoculation delay, were grown under high (greenhouse) and low (growth room) light intensity, number of nodules per plant was directly proportional to the amount of photosyntheticaly active radiation in the environment (Table II-IV). Furthermore, suppression of nodulation on D side was much more pronounced under low light intensity.

DISCUSSION

Further to close relationship between light intensity and nodulation observed by Kosslak and Bohlool (49) our results show that nodule numbers per plant and plant autoregulatory response are directly proportional to the light intensity in the growth environment (Table II-IV). Correlation between leaf area at the time of inoculation and nodule numbers at harvest (Fig. II-7 A) suggests that, increased nodulation obtained with delayed inoculation (Fig. II-4 A, Du side) is due to higher photosynthetic potential in older plants, rather than to larger infectible root area, as suggested by Nutman (59) for clover. The conclusion that, plant growth potential at the time of inoculation determines the number of nodule primordia and mature nodules per plant, is further supported by the experiment in which plants with only half root system, inoculated later, initiated more nodule primordia and developed essentially the same number of mature nodules as plants of the same age with complete root system inoculated earlier (Table II-III).

Similar total nodule mass on the E/D and U/D inoculated split-root system (Fig. II-4 B) and inverse relationship between nodule numbers on early and delayed inoculated root half (Fig. II-8) suggest a threshold for nodule numbers and nodule mass, depending on plant developmental stage and its growth potential. Reduced nodule mass on D side (Fig. II-4 B) cannot be accounted for by shorter period for nodule development on that side but, more likely, reflects greater partitioning of current photosynthate and dry matter to early initiated nodules at the expense of late initiated nodules. Approximately 16 hour interval between early and delayed inoculation seems sufficient for early initiated nodules to establish a prevailing sink (Fig. II-4 B).

Pierce and Bauer (66), using different experimental protocol from ours, showed that second inoculation 15 hours after the first produced virtually no nodules when the first inoculum dose was optimized for nodule yield. By contrast, our results (Fig. II-4) indicate that plant autoregulatory response affected nodule development (nodule mass) much more than nodule initiation (nodule numbers). The rate of nodule development may vary with plant growth potential (Fig.II-7 B) and nodules produced by the second inoculum apparently develop at a slower rate than those produced by the first (Table II-II). Therefore, observed autoregulatory response may vary considerably with experimental conditions affecting plant growth and with the interval after second inoculation, when nodules are scored.

Nitrogenase activity in nodules on the E side was first detected 10 days after (E) inoculation and in nodules on D side 8 days after (D) inoculation. Little suppression of nodule development on D side observed with 14-day delay of the second inoculation (Table II-II), compared to 4-day delay of second inoculation, suggests that, other factors became involved in regulation of nodule development after the onset of nitrogen fixation in first mature nodules. According to Atkins (3) there is a considerable lag period between the start of nitrogen fixation and substantial N export from nodules. In the absence of mineral nitrogen, N deficiency symptoms are regularly observed in plants at the onset of nitrogen fixation, since cotyledonary N reserve is depleted and leaf N is mobilized for nodule development (4). Nitrogen limitation to plant growth generally results in increased photosynthate and dry matter partitioning to roots (85) and presumably allows for development of additional nodules from nodule primordia initially arrested in further development (Table II-III). Singleton and Stockinger (72) have shown differential allocation of plant dry matter to effective and ineffective nodules. It seems therefore, that carbon partitioning related to nitrogen fixing efficiency of early nodules may ultimately determine nodule number and mass on legume roots.

MATERIALS AND METHODS

Growth system: A variant of a split-root procedure described in material and methods in Chapter II (Figure II-2) was used.

At planting, 25 ml N-free plant nutrient solution (PNS) was added per pouch and subsequently maintained at a level 1 to 3 cm from the bottom of the pouch with half strength PNS. Concentrations of nutrients in PNS were as described previously (Chapter II, materials and methods).

Plants were grown in the growth room under average PAR 300 to 400 μE/m²/sec, 18 h photoperiod and temperature range 23 to 27°C.

Inoculation treatments: Bradyrhizobium japonicum strain USDA 110 (TAL 102, obtained from the NifTAL Project collection) as peat based inoculant was suspended in N-free PNS so that 2.5 x 10⁷ cells was applied per root half in 4 ml of inoculum. Cell density in PNS was determined by drop plate method (43).

Half root portions were first inoculated (early inoculation) 8 days after planting, when the roots had grown to the bottom of the pouches and at least one trifoliate leaf had emerged on the shoot. The other root half was either inoculated 4 days later - early(E)/delayed(D) treatment, or remained uninoculated -early(E)/uninoculated(U) treatment.

¹⁴C labeling and root processing procedure: Plants of both treatments (E/U, E/D) were placed in a sealed clear plastic chamber at 24 hour intervals from l to 12 days after early and delayed inoculation. Tops of the split-root assemblies were sealed with plastic tape and tape caulk (Mapco, Inc., Cleveland, OH) around the stem to minimize direct 14CO2 incorporation by roots and nodules. The ¹⁴CO₂(45-80 μCi/plant, progressively increasing with plant development) was generated by injecting 10 to 20 ml 3.6 N sulfuric acid through a serum stopper into a beaker containing 2 ml NaH¹⁴CO₃ (ICN Biomedicals, Inc.) in 0.1N NaOH. A fan within the chamber was used to circulate ¹⁴CO₂. Plants were allowed to assimilate ¹⁴CO₂ for 70 min and the chamber opened for additional 20 mins for translocation of assimilates. It was shown previously (38) that, root and nodule radioactivity in ¹⁴CO₂ pulsed soybean peaks approx. 90 minutes after the start of ¹⁴CO₂ fixation by the leaf. Plants were then placed on ice, roots separated from the shoot, stained with Eriochrome black T, prepared after Bohlool (13), for 10 to 15 mins, rinsed in half strength PNS, then submerged in 50 ml half strength PNS with 0.1% Thimerosal (Sigma), added as preservative, and stored at 4°C until dissected.

Analysis of labeled tissue: Two plants of each treatment were used for dissection and one for autoradiography. Root halves of one plant/treatment were dissected, immediately one after another, under a dark field microscope (Wild M7 S). Nodule and root meristematic structures (1 to 5 mm root segments) were excised and grouped, based on structure and developmental stage. Classification of root and nodule structures is presented in Figure III-1. The remaining portions of "tap" and lateral roots were grouped separately. Excised and grouped nodule and root structures were air dried, weighed, placed in the scintillation vials, rehydrated with 0.30 ml H₂O, solubilized in 1 - 2 ml Soluene 350 (Packard) at 45°C in a water bath overnight and then suspended in 15 ml of scintillation cocktail (Hionic-fluor, Packard). Radioactivity was determined on a Packard 22000A Tri-carb scintillation analyzer.

Roots for autoradiography were stained as described above, stored in half strength PNS overnight at 4°C, then freeze-dried, pressed in a vice as described by Turgeon and Wimmers (80) and autoradiographed with Kodak X-OMAT AR film (Eastman Kodak). Exposure time was 2 to 3 days at room temperature. Identical autoradiographs were obtained from stained and unstained (control) roots, that were freeze-dried immediately after labeling.

Scintillation counts of root and nodule structures on stained and unstained test roots, labeled 7 or 14 days after inoculation and stored over 6 weeks showed that neither staining nor prolonged storage significantly affected distribution of radioactivity in roots.

Determination of nitrogenase activity: Half root systems of extra plants (2 to 4 replicates) for each treatment were placed into 100 ml test tubes at 24 h intervals from 6 - 12 days after E or D inoculation. Tubes were injected through a serum stopper with 5 ml acetylene and ethylene production was determined by gas chromatography (Varian 940 GC).

Statistical analysis: Data were analyzed by Tukey’s HSD test using SYSTAT statistical package (87).

RESULTS

Nodule initiation and development on early and delayed inoculated root half.

Individual nodule developmental stages, defined in Figure III-1, occurred at the same time from inoculation on early (E) and on delayed (D) inoculated root side. First cortical cell division centers (CDs) could be identified at 3 days from inoculation, first nodule primordia at day 5, emerging nodules at day 6 and nodules at day 7 after inoculation. Numbers of symbiotic and root meristematic structures, on the E/U and E/D inoculated split-root systems are presented in appendices III-1 and III-2. There were essentially the same number of CDs on E and D half root systems but significantly less CDs advanced to successive developmental stages on D side compared to E side (Table III-I). Nitrogenase activity (acetylene reduction) was first detected at 10 days after inoculation on E side and 8 days after inoculation on D side. The onset of nitrogenase activity coincided with the occurrence of nitrogen deficiency symptoms in plants and was not related to nodule size or number. No further increase of nodule mass on D side was observed after the onset of nitrogen fixation on E side, where nodule mass increased well into the nitrogen fixation stage (Figure III-7).

Sink intensity of developing nodules and roots

Respiratory loss of ¹⁴C from root and nodule structures was not measured. Based on findings of Gordon et al. (38) that, ¹⁴CO₂ respiration parallels ¹⁴C-sugar content in the sink tissue, it was assumed relative differences in radioactivity of separated (excised) root and nodule structures reflect differences in the import of labeled photosynthate.

Distribution of radioactivity within the root system, presented in Figure III-2, indicates the flow of labeled photosynthate to developing nodules and root tips which appear as major sink for current photosynthate. Based on scintillation counts, sink intensity of individual root and nodule structures was characterized by two parameters: 1) specific radioactivity (SA = dpm / mg dry weight), which indicates the flux of labeled photosynthate to or through the sink tissue; 2) relative specific radioactivi_ty (RSA = % dpm / % dry weight), which indicates partitioning of labeled photosynthate to the sink tissue relative to its size. Specific and relative specific radioactivity of individual nodule and root structures is presented in appendices III-3 to III-8.

Pooled data for nodule and root structures are presented in Figures III-3 and III-4. Both, specific and relative specific radioactivity indicate sink intensity of developing nodules on the E side was 3 - 4 fold that of the roots. By contrast, specific radioactivity of nodules on D side indicates reduced flux of photosynthate to those nodules (Figure III-3) compared to nodules on E side. However, relative to the amount of infected tissue (sink size), early (E) and late (D) nodules show similar sink intensity characterized by relative specific radioactivity (Figure III-4). Following the appearance of first mature nodules, their associated roots became increasingly deprived of labeled photosynthate (Figure III-2 E, Figure III-4).

Current photosynthate and dry matter partitioning within and between the opposite sides of the split roots.

Relative to roots, partitioning of labeled photosynthate to nodules increased exponentially with nodule development. The rate of increase was similar on the E side in both treatments (E/U, E/D), but significantly slower on D side. At the onset of nitrogen fixation on the respective sides, nodules accounted for over 60 of the radioactivity within the E side and for less than 10% of the radioactivity within D side (Figure III-5). In both treatments (E/U, E/D), increased photosynthate partitioning to E side coincided with the development of first mature nodules on that side. Even before first mature nodules developed on D side that side was already deprived of current photosynthate (Figure III-6).

Differences in photosynthate partitioning to roots and nodules on the opposite sides of the split-roots were clearly reflected in dry matter partitioning, which represents an integrated value of photosynthate partitioning over time (Figure III-7). Increased photosynthate partitioning to early nodules (Figure 6) reduced root growth on early and delayed inoculated side as well as nodule development on D side (Figure III-7).

DISCUSSION

Further to the findings of Calvert at al. (24), that autoregulation operates via the arrest of development of cortical cell division centers, the pattern of nodule development on the early and delayed inoculated roots in our experiment (Table III-I) indicates that, autoregulation is not a single step but, rather a continuous process where nodule development can be arrested at any stage of development of cortical cell division centers (CDs) into functional nodules. From equal numbers of CDs on early (E) and delayed (D) inoculated half root system, progressively less CDs advanced into each successive developmental stage on D side than on E side.

Much higher specific and relative specific radioactivity of developing nodules, compared to roots (Fig. III-3, Fig. III-4), indicate that even early nodule structures are much stronger sinks for current photosynthate than roots. Root tips among root structures and mature nodules among root structures were the most intense sinks (Appendices III-3 to III-8). Nodule sink intensity, as measured by specific radioactivity (Fig. III-3), increased with nodule developmental stage and, with the development of advanced nodule structures, sink intensity of early nodule structures, as well as, sink intensity of root structures decreased (App. III-5, App. III-8, Fig. III-3, Fig. III-4).

Current photosynthate and dry matter partitioning to early and late initiated nodules and their associated roots (Figures III-5 to III-8) shows the tremendous cost in terms of carbon for early nodule development and clearly indicates that early nodules develop at the expense of late initiated nodules, as well as, at the expense of root growth. Immediate dependence of nodule numbers per plant and of the intensity of autoregulatory response on the amount of photo synthetically active light, available to soybean plant (49, chapter II - Table II-IV) strongly suggest early nodule development is limited by carbon from photosynthesis. Thus, competition between early and late initiated nodules for a limited amount of photosynthate becomes more vigorous as nodules develop.

Regulation of organ growth and development by photosynthate partitioning, controlled by source limitation and sink demand, is a common process in higher plants (85). Once a potential sink is established, competitive success of that organ depends on the development of an adequate vascular link for the supply of carbon and nutrients, apart from the growth characteristics of that organ, imposed by growth regulators (85, 86). A model postulating translocatable signals acting as growth regulators in the earliest stages of soybean nodule development has been proposed by CaetanoAnolles and Gresshoff (19, 39). Their study (19) also shows that at least some meristematic activity in the early inoculated root is necessary to induce feedback suppression of nodulation in delayed inoculated root.

According to Calvert et al. (24), in soybean, only CDs closely associated with the infection threads develop into nodule primordia, which are characterized by vascular connection between the nodule meristem and root steele (Fig. III-1). In our study, the beginning of selective photosynthate partitioning to early inoculated root side coincided with the onset of nodule development - i.e. with the development of first nodule primordia on that side (Fig. III-6). Thus, vascular connection of potential nodule meristems to the root vascular system - a transition of a cortical cell division center into a nodule primordium, may clearly represent the early determinant for successful development of infection into a nodule. Consequently, due to their developmental and thus competitive advantage as sinks for current photosynthate, first established nodule primordia are also the first to develop into functional nodules, while nodule primordia initiated later are deprived of current photosynthate and their development slowed or completely arrested. Caetano-Anolles et al. (22) showed that excision of first formed nodules allows for development of nodule primordia, that are clustered around early nodules (Fig. III-1 d), and were initially suppressed.

High proportion of CDs that develop into nodules in alfalfa (20) and in supernodulating soybean mutant (55) suggest that, autoregulatory response in different species or genotypes of the same species may be related to the overall pattern of nodule development (e.g. indeterminate type nodules in alfalfa compared to determinate type in soybean) or to some underlying mechanism controlling the rate of transition of CDs into nodule primordia (55). Nodule primordia formed at high rate (55) are likely to be equally competitive as individual sinks and most of them may therefore develop into functional nodules.

MATERIALS AND METHODS

Growth systems: Two variants of a split-root system described by Singleton (70) were used:

1) The growth pouch assembly described in Chapter II (Figure II-2) was used in the growth room experiment.

2) For the greenhouse study, two square pots (0.7 L) were taped together on a tongue depressor, serving as a base. Polyethylene bags were placed inside the pots and filled with dry horticultural vermiculite. Plastic elbows (90 deg. angle, 1/2" diameter) with a 13mm hole, drilled in the center were used to direct split roots into pots.

Planting procedure: Seeds of soybean (Glycine max L., Merr.) cv. Lee were surface sterilized, germinated and planted as described in Chapter II. The top of the pot split-root assembly was covered with aluminum foil. At planting, 30 ml of N-free plant nutrient solution (PNS) was added per pouch and subsequently maintained at a level 1 to 3 cm from the bottom of the pouch with half strength PNS. Plants in pots received 200 ml PNS per pot at planting and additional 200 ml per pot 10 days after inoculation. Concentrations of nutrients in PNS were as described in materials and methods in Chapter II. Plants in pouches were grown in the growth room under average PAR 350 μE/m²/sec, 18 h photoperiod and temperature range 23 to 27°C. Plants in vermiculite were grown in the greenhouse under PAR 1350 - 1600 μE/m²/sec, approximately 13 h photoperiod and temperature range between 13 and 37°C.

Inoculation treatments: Bradyrhizobium japonicum, strains USDA 110 and USDA 38 were obtained from the Niftal Project collection. Six day old YEM broth cultures were diluted with N-free PNS adequately, so that 10⁸ cells was applied per root side - in 1 ml of inoculum per pouch and in 50 ml of inoculum per pot. This presumably ensured similar distribution of bacteria along the roots in the two growth media. Cell density in broth cultures was determined by counts on black polycarbonate filters (Poretics Corp., Livermore, CA) using specific FAs for the two strains and later verified by drop plate counts (43). For mixed inoculations, strains were mixed in the ratio 1:1, according to filter counts. Plate counts indicated ratio of USDA 110 : USDA 38 of 1.0 : 1.1 in vermiculite and 1.0 : 1.3 in pouches.

Plants were inoculated 8 days after planting, when the roots had grown to the bottom of the pouches and pots and at least one trifoliate leaf had emerged on the shoot. Three inoculation treatments were imposed on split roots in each growth medium: 1) one side inoculated with either USDA 110 or USDA 38, other side uninoculated; 2) one side inoculated with USDA 110, other side with USDA 38; 3) one side inoculated with a mixture of USDA 110 and USDA 38, other side uninoculated.

Identification of early nodule structures and rhizobial strains: Sets of plants were harvested at 5, 10 and 21 days after inoculation. For the identification of cortical cell division centers and nodule primordia (for classification see Chapter III, Figure III-1), roots were separated from the shoot, stained with Eriochrome Black T, prepared after Bohlool (13) for 10 to 15 min, quickly rinsed in PBS, then submerged in 50 ml PBS with 0.01 Thimerosal (Sigma) and stored at 4°C until dissected. Double inoculated roots were shaken in 250 ml 0.5 N NaOH with 4g glass beads (75 - 150 μm) for 1 hour and then washed 3 X 15 mins in phosphate buffer (pH 7.1) on a wrist action shaker prior to staining. Prior to dissection these roots were incubated in 1:1: 50 mixture of unconjugated (to FITC) antisera specific for each strain and phosphate buffer saline (PBS) for 1 hour at 37°C with minimum rotary shaking; then rinsed in PBS (pH 7) incubated 1/2 hour in PBS at room temperature and stored in H₂O for dissection.

Nodule primordia and early nodules (10 days after inoculation) were excised under the dissecting microscope, air dried and stored for strain identification. Rhizobia within nodule primordia and nodules were identified with FAs specific for each strain, according to Bohlool (13). Rehydrated nodule primordia were crushed on microscopic slides and nodules in microtiter plates (Immulon2, Dynatech Laboratories, Inc.) and smeared in duplicates. Examples of FA reactions in nodule primordia and in nodules are presented in Figure IV-1. Treatment with NaOH removed most of the rhizobia attached to the root surface (Fig. IV-1) and antiserum treatment efficiently blocked the fluorescence of the remaining cells. Test observations of the intact root segments and of crushed nodule primordia proved that these two treatments virtually eliminated interference of surface attached bacteria with identification of rhizobia within nodule primordia and nodules. Rhizobia within nodule primordia could be first detected 4 to 5 days after inoculation.

Determination of nitrogenase activity: Extra plants for each treatment were tested for nitrogenase activity at 24 h intervals from 7 to 10 days after inoculation. Half root systems were placed into 100 ml test tubes. Tubes were injected through a serum stopper with 5 ml acetylene and ethylene production was determined by gas chromatography (Varian 940 GC).

Statistical analysis: Four to six plants per inoculation treatment on each sampling date were scored for nodulation and nodule occupancy and 2 plants per inoculation treatment were used in acetylene reduction assays. Data were analyzed by Tukey’s HSD test, using SYSTAT statistical package (87).

DISCUSSION

The relationship between nodule initiation rates of the two strains and their competition pattern observed in this study supports the correlation between these two processes proposed by McDermot and Graham (56). Comparison of infection and nodule initiation rates of strains USDA 110 and USDA 38 in the indirect competitive system (Table IV-III, inoculation treatment 110/38) with their nodule occupancy in a direct competitive system (Table IV-II, inoculation treatment 110+38/un) indicates that, superior competitiveness of USDA 110 could be attributed to its superior rate of infection and nodule initiation in a competitive system. In vermiculite, USDA 110 initiated significantly more cortical cell division centers and significantly more nodule primordia, compared to USDA 38 (Table IV-III) and at the same time occupied more nodule primordia than USDA 38 when strains were in direct competition (Table IV-II). In growth pouches, USDA 110 showed only moderate advantage over USDA 38 in the infection and nodule initiation rate (Table IV-III), appeared far less dominant in forming nodule primordia and, consequently, occupied less nodules when strains were in direct competition (Table IV-II). The rates of infection and nodule development by USDA 110 and USDA 38 did not differ significantly in a noncompetitive system (Table IV-III; inoculation treatments 110/un, 38/un).

Occupancy by the two strains, of nodule primordia 5d after inoculation and of functional nodules 21d after inoculation (Table IV-II) indicates that, in both, direct and indirect competitive system, the outcome of competition between the two strains was determined during the earliest stages of infection, where more competitive strain (USDA 110) initiated more potential nodule sites (cortical cell division centers) than USDA 38 (Table IV-III) and also initiated nodules faster, as indicated by greater total number and higher proportion of advanced nodule primordia (data not shown) formed by USDA 110 compared to USDA 38.

Earlier nitrogenase activity in nodules formed by USDA 110 (Table IV-IV) also indicates earlier penetration of the infection thread into plant cells and earlier differentiation into N₂ fixing bacteroids, as a result of faster infection by USDA 110.

When double inoculations, using the same strain but delayed second inoculation, are performed on a single (24) or on a split-root (Chapter III, Table III-I) system, similar numbers of infections (cortical cell division centers) are observed on early and delayed inoculated root region but, further development of infections on the delayed inoculated root region is suppressed. This phenomenon, collectively known as autoregulation is a plant controlled response, which is morphologically detectable within 3 to 7 days after inoculation (24, 77). By contrast, our present results indicate that, the outcome of interstrain competition is determined, to a large extent already at the cortical cell division stage or, at the latest, at nodule primordia stage. Autoregulation may simply favor the development of early initiated nodule primordia regardless to their occupancy. Consequently, initial competition pattern observed in nodule primordia is clearly reflected in nodule numbers and nodule mass produced by the two strains (Table IV-III). Thus, the rate of infection rather than rate of nodule development appears to be determinant of strain’s competitiveness. Halverson and Stacey (40) demonstrated that short term preincubation of rhizobia in soybean root exudate increases nodule initiation rate of a slow to nodulate strain. Inoculation with one strain 4 - 6 hours before a second strain is introduced (33, 48), seems sufficient for a less competitive strain to colonize the roots and initiate infections to a stage where infection by a more competitive strain is suppressed.

At optimal inoculum doses for nodule yield, the attachment of rhizobia to soybean roots is virtually completed within 1 hour after inoculation (83) and motility and chemotaxis appear important determinants of root colonization and nodule initiation (7).

In our study, environmental factors, such as growth medium and light intensity, clearly affected early events in the infection process. Therefore, it seems likely that in soil, which is by far more complex environment, strain response to plant symbiotic signals in the rhizosphere, their motility and chemotaxis, determine the proportion of root colonization and nodule initiation by competing strains, which is then manifested in numbers of functional nodules formed by each strain.

REGULACIJA NODULACIJE IN KOMPETICIJE MED SEVI BAKTERIJE Bradyrhizobium japonicum PRI FORMIRANJU SIMBIOZE S SOJO (Glycine max [L], Merrill).

POVZETEK

Formiranje simbioze med strocnicami in bakterijami rodu Rhizobium poteka preko kompleksnega redosleda genetskih in fizioloskih interakcij med partnerjema, ki kulminirajo v zrelih nodulah, v katerih poteka fiksacija atmosferskega dusika. Gostiteljska rastlina regulira stevilo nodul na koreninah s procesom imenovanim autoregulacija. Doslej je bila autoregulacija nodulacije dokazana pri stevilnih zelnatih strocnicah. Ko je gostiteljska rastlina izpostavljena vecim sevom homolognih Rhizobijev hkrati, ti tekmujejo za omejeno stevilo nodul na skupnem gostitelju. Cimbolj kompetitiven je sev, temvecji delez v skupnem stevilu nodul okupira oz. formira. Kompeticija med indigenimi in introduciranimi sevi Rhizobiuma v praksi predstavlja glavno oviro za uspesno inokulacijo s sevi, ki so sposobni fiksirati vec atmosferskega dusika.

Mehanizmi autoregulacije in kompeticije med sevi so se vedno nepojasnjeni. V nasih studijah proucujemo stadije infekcijskega procesa, v katerih poteka autoregulacija nodulacije in kompeticija med sevi Bradyrhizobium japonicum za nodulacijo na soji, ter mehanizme preko katerih gostiteljska rastlina regulira infekcijo in razvoj nodul.

V literaturi predlagani mehanizem avtoregulacije vkljucuje 2 tipa regulatornih substanc: 1) zgodaj zasnovani nodulni in koreninski meristemi producirajo inhibitor(je), ki neposredno zavira(jo) razvoj kasneje zasnovanih nodul; 2) zgodnje infekcije preko se neidentificiranega molekularnega signala izzovejo produkcijo sistemskega inhibitorja v poganjku, to pa nato deluje na pozne infekcije. Med stevilnimi objavljenimi studijami, ki obravnavajo autoregulacijo, ena sama ugotavlja vlogo fotosintatov pri regulaciji stevila nodul na rastlini. Zato v prvi seriji eksperimentov ugotavljamo vpliv rastnega potenciala gostiteljske rastline na stevilo in maso nodul pri optimalni dozi inokuluma ter opredelimo casovni interval po inokulaciji, v katerem nastopa autoregulacija. Uporabljamo razdeljen koreninski sistem (Figures II-1 in II-2), kjer eno polovico korenin inokuliramo zgodaj (cas 0) ali pustimo neinokulirano, drugo polovico korenin pa inokuliramo z dolocenim casovnim zamikom. Razvoj nodul na obeh polovicah korenin koreliramo z kolicino fotosintetsko aktivne svetlobe (PAR - μE/m²/sec) ter z listno povrsino in dolzino korenin v obdobju od inokulacije do pojava prvih nodul.

V primeru ko je bila polovica korenin inokulirana 1, 2, 4, 8, 16, 32, 64 ali 96 ur za drugo polovico (Figure 11-4), je bilo stevilo nodul na pozno inokulirani polovici korenin znacilno nizje kot na zgodaj inokulirani polovici korenin pri zamiku druge inokulacije za 64 ur ali vec. Masa nodul na pozno inokulirani polovici korenin pa znacilno nizja kot na zgodaj inokulirani polovici korenin ze pri zamiku druge inokulacije za 16 ur ali vec. Stevilo nodul na neinokuliranem/zakasnjeno inokuliranem koreninskem sistemu je narascalo linearno (r=0.97; Figure II-4) s casovnim zamikom inokulacije (1 - 96 ur). Izracunana korelacija med stevilom nodul na rastlino in listno povrsino v casu inokulacije (r=0.98; Figure II-7A, Table II-I) je znacilno visja kot korelacija med stevilom nodul in dolzino korenin v casu inokulacije (r=0.67; Table II-1). Zabelezen je bil hitrejsi razvoj nodul na rastlinah z vecjo listno povrsino v casu inokulacije (Figure II-7B). Stevilo in masa nodul na rastlino sta bila premosorazmerna z intenziteto svetlobe v rastnem okolju (Table II-IV), Stevilo nodul na pozno inokulirani polovici korenin pa obratno sorazmerno s stevilom nodul na zgodaj inokulirani polovici korenin (Figure II-8). Intenziteta svetlobe je tudi znacilno vplivala na razmerje med stevilom nodul na zgodaj inokulirani polovici korenin in stevilom nodul na pozno inokulirani polovici korenin (Table II-IV). Pri 4 - dnevnem intervalu med zgodnjo in pozno inokulacijo razdeljenega koreninskega sistema je bilo stevilo nodul na pozno inokulirani polovici znacilno nizje, kot pri 14 - dnevnem intervalu med inokulacijama (prim. Table II-II in II-III). Nitrogenazno aktivnost v nodulah na zgodaj inpkulirani polovici korenin smo ugotovili (acetilenski test) deseti dan po (zgodnji) inokulaciji. V poskusu, kjer smo zgodaj inokulirano polovico korenin odrezali v casu inokulacije druge polovice korenin (14 dni za zgodnjo inokulacijo) se je na preostali polovici korenin razvilo skoraj 5- krat toliko nodulnih primordijev kot na celotnem zgodaj in pozno inokuliranem razdeljenem koreninsken sistemu kontrolnih rastlin (Table II-III); Po stevilu zrelih (funkcionalnih) nodul, 3 tedne po pozni inokulaciji, pa se rastline s polovico korenin in kontrolne rastline niso razlikovale (Table II-III) .

Navedeni rezultati ka2yejo, da je stevilo zasnovanih nodul (nodulnih primordijev) modno odvisno od rastnega potenciala (e.g. listne povrsine) rastline v casu inokulacije, medtem ko na stevilo zrelih nodul na rastlini odlocilno vpliva intenziteta fotosintetsko aktivne svetlobe. Regulatorni mehanizem (avtoregulacijo), ki uravnava stevilo in maso in nodul na rastlino je mogoce opaziti ze 16 ur po inokulaciji. Nizja stopnja autoregulacije pri 14 dnevni zakasnitvi druge inokulacije (po zacetku fiksacije dusika v zgodnjih nodulah) kot pri 4 dnevni zakasnitvi (pred zacetkom fiksacije dusika v zgodnjih nodulah) kaze na fizioloske spremembe v nodulirani rastlini, glede na omejujoce dejavnike rasti rastline in razvoja nodul. Simptomi pomankanja dusika na rastlinah, ki se pojavijo ob zacetku nitrogenazne aktivnosti v nodulah (10 dni po inokulaciji), kazejo na pomankanje N za rast poganjka na kar rastlina reagira s povecanim dotokom asimilatov v korenine. Ti asimilati skupaj s fiksiranim dusikom omogocajo razvoj vecjega stevila nodul.

V naslednji studiji proucujemo vlogo selektivne porazdelitve asimilatov pri autoregulaciji nodulacije. Spet uporabimo razdeljen koreninski sistem (Figure 11-2) z dvema obravnavanjema:

1) zgodaj inokuliran/neinokuliran koreninski sistem (E/U treatment) kot kontrola;

2) zgodaj/pozno inokuliran koreninski sistem (E/D treatment), kjer drugo polovico korenin inokuliramo 4 dni za prvo.

Po 2 rastlini vsakega obravnanja izpostavimo ¹⁴CO₂ vsakih 24 ur od inokulacije do zacetka fiksacije dusika v prvih nodulah. Porazdelitev radioaktivnih asimilatov v koreninah najprej zasledujemo z autoradiografijo (Figure III-2). Po barvanju korenin z Eriochrome Black T, pod svetlobnim mikroskopom (binokularno lupo) na temnem polju lahko opredelimo razvojne stadije nodul (Figure III-1). Celotni koreninski sistem nato (pod mikroskopom) seciramo in grupiramo nodulno in koreninsko tkivo glede na razvojni stadij in strukturo. Po raztapljanju tkiva in suspenziji v scintilacijskem koktajlu merimo radioaktivnost v razvijajocih se nodulah in koreninah posameznih rastlin na scintilacijskem stevcu.

Kot je pokazala mikroskopska analiza, se na zgodaj in pozno inokulirani polovici korenin zasnuje priblizno enako stevilo zacetnih nodulnih meristemov (cortical cell division centers, Figure III-1), od katerih pa napreduje v nadaljne razvojne stadije na pozno inokulirani polovici korenin znacilno nizji delez kot na zgodaj inokulirani polovici korenin (Table III-1). Distribucija radioaktivnosti v koreninskem sistemu prikazana na autoradiografih (Figure III-2) kaze, da so razvijajoce se nodule in koreninski vrsicki glavni ponori asimilatov. Na osnovi scintilacijskih podatkov, smo intenziteto ponorov opredelili z dvema parametroma: 1) specificna radioaktivnost ((dpm (disintegrations per minute) / mg suhe teze], ki kaze dotok radioaktivnih asimilatov v ponorno tkivo (sink tissue);

2) relativna specificna radioaktivnost ( dpm / o suhe teze tkiva), ki kale porazdelitev radioaktivnih asimilatov glede na delez ponornega tkiva v skupni masi noduliranih korenin.

Specificna in relativna specificna radioaktivnost sta bili 3 do 4 - krat visji v nodulah kot v koreninah (Figure III-3, Figure III4). Specificna radioaktivnost v nodulah je narascala sorazmerno z razvojnim stadijem in je bila znacilno visja v nodulah na zgodaj inokulirani polovici korenin kot v nodulah na pozno inokulirani polovici korenin (Figure III-3). Glede na delez nodulnega tkiva v masi zgodaj in pozno inokuliranih korenin pa kazejo zgodaj in pozno zasnovane nodule podobno intenziteto ponora oz. relativno specificno radioaktivnost (Figure III-4, E/D treatment). Z razvojem prvih zrelih nodul postanejo korenine prikrajsane za radioaktivne asimilate (Figure III-2, Figure III-4) . Ob zacetku fiksacije dusika na zgodnji oz. pozno inokulirani polovici korenin je odpadlo na zgodnje nodule preko 60%, na pozne nodule pa manj kot lot skupne izmerjene radioaktivnosti v zgodaj oziroma pozno inokulirani polovici korenin (Figure III-5). Znacilno vecji dotok radioaktivnih asimilatov v zgodaj inokulirano polovico korenin na racun preostale polovice korenin (neinokulirane ali pozno inokulirane) je bil opazen socasno z pojavom prvih zrelih nodul na zgodaj inokulirani polovici korenin (Figure III-6). Razlike v porazdelitvi radioaktivnih asimilatov v nodulah in koreninah na obeh polovicah razdeljenega koreninskega sistema so se jasno odrazale v tezi suhe snovi korenin in nodul (Figure III-7). Teza suhe snovi namrec predstavlja integralno vrednost porazdelitve asimilatov v nekem casovnem obdobju. Posledica vecjega dotoka asimilatov v zgodnje nodule je bila zmanjsana rast korenin na obeh polovicah koreninskega sistema in znacilno nizja masa nodul na pozno inokulirani polovici korenin kot na zgodaj inokulirani polovici korenin (Figure III-7).

Po razvoju nodul na zgodaj in pozno inokulirani polovici korenin (Table III-1) lahko sklepamo, da je autoregulacija kontinuiran proces, ki poteka v celotnem obdobju razvoja nodul od prvih centrov celicnih delitev v korteksu do funkcionalnih nodul. Porazdelitev radioaktivnih asimilatov v razdeljenem koreninskem sistemu (Figures III-2, III-4, III-6) jasno kaze na kompeticijo med nodulnimi in koreninskimi meristemi za razpolozljivo kolicino asimilatov. Da se zgodnje nodule razvijajo na racun pozno zasnovanih nodul in na racun rasti korenin dokazujeta tako intenziteta ponorov (specificna in relativna specificna radioaktivnost) kot teia nodul in korenin pri obeh obravnavanjih razdeljenega koreninskega sistema. Neposredna odvisnost skupnega stevila nodul na rastlino in stopnje autoregulacije od razvojnega stadija rastline v casu inokulacije in intenzitete fotosintetsko aktivne svetlobe kaze, da so asimilati zelo pomemben omejitveni dejavnik razvoja nodul. Na osnovi nasih rezultatov predlagamo mehanizem autoregulacije, ki predpostavlja da so, zaradi narascajocega dotoka asimilatov v zgodnje nodule, pozno zasnovane nodule prikrajsane za asimilate, zato se njihov nadaljni razvoj znatno upocasni ali popolnoma ustavi.

V zadnjem eksperimentu ugotavljamo vlogo autoregulacije pri kompeticiji med sevi Bradyrhizobium japonicum za stevilo nodul na skupnem gostitelju - soji. v dosedaj publiciranih studijah avtorji ugotavljajo kompetitvnost sevov po stevilu funkcionalnih nodul, ki jih formirajo posamezni sevi, ce so v inokulacijski mesanici sevi (stevilo celic) v enakem stevilcnem razmerju. Cela vrsta studij pa kaze, da so za izid kompeticije odlocilni najzgodnejsi stadiji infekcijskega procesa. Z naso tehniko barvanja (eriochrome black) in blokiranjem fluorescence povrsinsko pritrjenih celic B. japonicum lahko identificiramo seve v zgodnjih nodulnih primordijih z za posamezni sev specificnimi fluorescencnimi protitelesci (Figure IV-1). S primerjavo razmerja sevov v nodulnih primordijih in v funkcionalnih nodulah, ugotavljamo prispevek zgodnjih (infekcija) in poznih (razvoj nodul) stadijev infekcijskega procesa h koncnemu izidu kompeticije med sevoma, v funkcionalnih nodulah.

Iz predhodnih raziskav znano kompetitivni sev (USDA 110) in slabo kompetitivni sev (USDA 38) B. japonicum inokuliramo istocasno, v enakem stevilcnem razmerju celic, na razdeljen koreninski sistem soje v treh variantah: 1) oba seva na eni polovici korenin, druga polovica ostane neinokulirana - neposredni kompetitivni sistem; 2) oba seva vsak na svoji polovici korenin - posredni kompetitivni sistem, kjer rastlina igra vlogo posrednika; 3) en sev na polovici korenin, druga polovica neinokulirana - nekompetitivni sistem. Enak eksperimentalni protokol uporabimo pri rastlinah gojenih v rastlinjaku, v vermikulitu - Figure II-1 in rastlinah gojenih v rastni komori v PE vreckah (growth pouches) - Figure II-2.

Stevilo nodul, ki jih formirata seva (posamezno ali skupaj) ugotavljamo 5 dni po inokulaciji - nodule v primordialnem stadiju, 10 dni po inokulaciji - zacetek fiksacije N₂ v nodulah, in 21 dni po inokulaciji - zrele funkcionalne nodule.

Rastni pogoji (medij in intenziteta svetlobe) so znacilno vplivali na kolicino rastlinske biomase (Table IV-I) in na kompeticijo med sevoma (Table IV-II). V posrednem kompetitivnem sistemu v vermikulitu je sev USDA 110 formiral 85 % vseh nodulnih primordijev in 75 o zrelih nodul, v PE vreckah pa 63 % vseh nodulnih primordijev in 74 % zrelih nodul. V direktnem kompetitivnem sistemu v vermikulitu je sev USDA 110 formiral 70 vseh nodulnih primordijev in 94 % zrelih nodul, v PE vreckah pa le 25 % vseh nodulnih primordijev in 48 % zrelih nodul (Table IV-II). Kljub temu se je sev USDA 110 tudi v PE vreckah izkazal kot bolj kompetitiven, saj sta 75 % vseh nodulnih primordijev in 31 % vseh zrelih nodul okupirala oba seva skupaj. V nekompetitivnem sistemu sta seva zasnovala podobno stevilo infekcij (centrov celicnih delitev v korteksu in nodulnih primordijev) in formirala podobno stevilo nodul (Table IV-III). V posrednem kompetitivnem sistemu v vermikulitu pa je sev USDA 110 zasnoval 3 do 5 - krat toliko infekcij kot sev USDA 38 in formiral 3 - krat toliko nodul kot USDA 38 (Table IV-III). Sev USDA 110 je pokazal tudi zgodnejso in vecjo zacetno aktivnost nitrogenaze v nodulah (Table IV-IV).

Priblizno enako razmerje med sevoma po stevilu nodulnih primordijev in po stevilu nodul, ki jih formirata seva v neposredni kompeticiji, kaze, da je koncni izid kompeticije dolocen ze v najzgodnejsih stadijih infekcijskega procesa. Znacilno vecje stevilo infekcij, ki jih zasnuje bolj kompetitivni sev pa kaze, da na kompetitivnost seva oziroma na izid kompeticije med sevi lahko odlocilno vpliva zacetna interakcija med mikro - in makro - simbiontom. Pri zaporedni inokulaciji koreninskega sistema z istim sevom, zgodnji in pozni inokulum zasnujeta priblizno enako stevilo infekcij - zacetnih nodulnih meristemov, autoregulacija pa nato zavira nadaljni razvoj poznih infekcij. Nasa zadnja studija pa kaze, da je pri socasni inokulaciji s sevi, ki se razlikujejo po kompetitivnosti, izid kompeticije (razmerje med sevi v funkcionalnih nodulah) dolocen ze v zacetnih nodulnih meristemih in

autoregulacija nato pac favorizira razvoj zgodaj zasnovanih infekcij (nodul), ne glede na to kateri sev jih zasnuje.

REFERENCES

1. Amarger N, Lobreau JP (1982) Quantitative Study of Nodulation Competitiveness in Rhizobium Strains. Appl Environ Microbiol 44:583-588.

2. Atkins CA (1984) Efficiencies and inefficiencies in the legume/Rhizobium symbiosis - A Review. Plant and Soil 82:273-284.

3. Atkins CA (1986) The Legume/Rhizobium Symbiosis: Limitations to Maximizing Nitrogen Fixation. Outlook in Agriculture. 15,3:128-134.

4. Atkins CA, Pate JS, Sanford PJ, Dakora FD, Matthews I (1989) Nitrogen nutrition in relation to N-hunger in Cowpea (Vigna unguiculata L. Walp). Plant Physiol 90:1644-1649.

5. Bauer WD (1981) Infection of Legumes by Rhizobia. Ann Rev Plant Physiol 32:407-449.

6. Bauer WD, Bhuvaneswari TV, Calvert HE, Law IJ, Malik NSA, Vesper SJ (1985) Recognition and infection by slowgrowing rhizobia. In HJ Evans, PJ Bottomley, WE Newton, eds, Nitrogen Fixation Research Progress, Martinus Nijhoff, Amsterdam, pp 247-253.

7. Bauer WD, Caetano-Anolles (1991) Chemotaxis, induced gene expression and competitiveness in the rhizosphere. In DL Keister and PB Cregan, eds, The rhizosphere and plant growth, Kluwer Academic Publishers, Nederlands, pp 155-162.

8. Bergersen FJ (1977) Physiological Chemistry of Dinitrogen Fixation by Legumes. In RWF Hardy and WS Silver, eds, A Treatise on Dinitrogen Fixation, section III, Biology, J Willey & Sons, pp 519-555.

9. Bhuvaneswari TV, Turgeon BG, Bauer WD (1980) Early Events in the Infection of Soybean (Glycine max L.Merr) by Rhizobium japonicum. I.Localizaton of Infectible Root Cells. Plant Physol 66:1027-1031.

10. Bhuvaneswari TV, Bhagwat AA, Bauer WD (1981) Transient Susceptibility of Root Cells in Four Common Legumes to Nodulation by Rhizobia. Plant Physiol 68:1144-1149.

11. Bhuvaneswari TV, Lesniak AP, Bauer WD (1988) Efficiency of Nodule Initiation in Cowpea and Soybean. Plant Physiol 86:1210-1215.

12. Bisseling T, Govers F, Gloudemans T, Moerman M, Ten Valde I, Van Kammen A (1985) Identification of PEA genes Involved in the Establishment of an Effective Symbiosis. Proc 2nd Intl Symp on Molecular Genetics of the Bacteria-Plant Interaction, Ithaca, New York, pp 74-77.

13. Bohlool BB (1987) Fluorescence Methods for Study of Rhizobium in Culture and in situ. In GH Elkan, ed, Symbiotic Nitrogen Fixation Technology, Marcel Dekker Inc, New York, Basel, pp 127-147.

14. Bohlool BB, (1989) A unifying concept in rhizobial interstrain competition for nodule occupancy. Proc 12th North Am Symb N Fix Conf, Ames, Iowa, p 5.

15. Bohlool BB, Schmidt EL (1974) Lectins: A Possible Basis for Specificity in the Rhizobium-Legume Root Nodule Symbiosis. Science 4147:269-271.

16. Bohlool BB, Schmidt EL (1973) Persistence and Competition Aspects of Rhizobium japonicum Observed in Soil by Immunofluorescence Microscopy. Proc Soil Sc Soc Am 37:561-564.

17. Brun WA, Betts KJ (1984) Source/Sink Relations of Abscising and Nonabscising Soybean Flowers. Plant Physiol 75:187-191.

18. Caetano-Anolles G, Bauer WD (1988) Feedback regulation of nodule formation in alfalfa. Planta 175:546-557.

19. Caetano-Anolles G, Gresshoff PM (1990) Early induction of feedback regulatory responses governing nodulation in soybean. Plant Sci 71:69-81.

20. Caetano-Anolles, Gresshoff PM (1991) Alfalfa controls nodulation during the onset of Rhizobium-induced cortical cell division. Plant Physiol 95:366-373.

21. Caetano-Anolles, Gresshoff PM (1991) Plant Genetic control of nodulation. Annu Rev Microbiol 45:345-82.

22. Caetano-Anolles G, Paparozzi ET, Gresshoff PM (1991) Mature nodules and root tips control nodulation in soybean. J Plant Physiol 137:389-396.

23. Caetano-Anolles G, Wall LG, De Micheli AT, Macchi EM, Bauer WD, Favelukes G (1988) Role of Motility and Chemotaxis in Efficiency of Modulation by Rhizobium meliloti. Plant Physiol 86:1228-1235.

24. Calvert HE, Pence MK, Pierce M, Malik NSA, Bauer WD (1984) Anatomical Analysis of the Development and Distribution of Rhizobium Infections in Soybean Roots. Can J Bot 62:2375-2384.

25. Dart P (1977) Infection and Development of Leguminous Nodules. In RWF Hardy, WS Silver, eds, A Treatise on Dinitrogen Fixation, Section III, Biology, J Willey & Sons, pp 367-448.

26. Dazzo FB, Napoli CA, Hubbell DH (1975) Adsorption of Bacteria to Roots as Related to Host Specificity in the RhizobiumClover Symbiosis. Appl Environ Microbiol 6:166-171.

27. Delves AC, Mathews A, Day AD, Carter AS, Carrol BJ, Gresshoff PM (1986) Regulation of the Soybean-Rhizobium Nodule Symbiosis by Shoot and Root Factors. Plant Physiol 82:588-590.

28. Deshmane N, Stacey G (1989) Identification of Bradyrhizobium nod Genes Involved in Host-Specific Modulation. J Bacteriol 6:166-171.

29. Dixon ROD, Wheeler CT (1986) Nitrogen Fixation in Plants. Blackie, Glasgow, London, Chapman and Hall, New York, pp 15123.

30. Djordjevic MA, Batley RTWJ,Rolfe B. (1987) Clovers Secrete Specific Phenolic Compounds which Stimulate or Repress nod Gene Expression in Rhizobium trifolii. EMBO J 6:1173-79.

31. Djordjevic SP, Ridge RW, Chen H, Redmond JW, Batley M, Rolfe BG (1988) Induction of pathogenic-like response in the legume Macroptilium atropurpureum by a transposoninduced mutant of the fast-growing, broad-host-range Rhizobium strain NGR234. J Bacteriol 170:1848-1857.

32. Emerich DW, Lepo JE, Evans HJ (1983) Nodule Metabolism. In WJ Broughton, ed, Nitrogen fixation vol 3, Legumes, Oxford University Press, pp 213-244.

33. Fernandez-Flouret D and Cleyet-Marel JC (1989) Competition Between Bradyrhizobium japonicum strains for Nodulation: Study of Nodulation Regulation in the Soybean. Symbiosis 7:171-186.

34. Firmin JL, Wilson RE, Rossen L, Johnston AWB (1986) Flavonoid Activation of nod Genes in Rhizobium Reversed by Other Compounds Prersent in Plants. Nature 324:90-92.

35. George MLC, Robert FM (1991) Autoregulatory Response of Phaseolus vulgaris L. to Symbiotic Mutants of Rhizobium leguminosarum bv.Phaseoli. Appl Environ Microbiol 57:2687-2692.

36. Gloudemans T, Bisseling T (1989) Plant gene expression in early stages of Rhizobium-legume symbiosis. Plant Sci 65:1-14.

37. Gloudemans T, De Vries SC, Bussink HJ, Malik NSA, Fransen HJ (1987) Nodulin gene expression during soybean (Glycine max) nodule development. Plant Mol Biol 8:395-403.

38. Gordon AJ, Ryle GJA, Mitchell DF, Powell CE (1985) The Flux of ¹⁴C-Labelled Photosynthate through Soyabean Root Nodules during N₂ Fixation. J Exp Bot 36, 166:756-769.

39. Gresshoff PM, Caetano-Anolles G (1991) Systemic regulation of nodulation in legumes. In Plant Bioitechnology and Development, CRC Press, pp 87-100.

40. Halverson LJ and Stacey G (1984) Host Recognition in the Rhizobium-Soybean Symbiosis. Detection of a protein factor in soybean root exudate which is involved in the nodulation process. Plant Physiol 74:84-89.

41. Halverson LJ and Stacey G (1985) Host Recognition in the Rhizobium-Soybean Symbiosis. Evidence for the involment of lectin in nodulation. Plant Physiol. 77:621-625.

42. Heron DS, Pueppke SG (1987) Regulation of Nodulation in the Soybean - Rhizobium Symbiosis. Strain and cultivar variability. Plant Physiol. 84:1391-1396.

43. Hoben HJ, Somasegaran P (1982) Comparison of the Pour, Spread and Drop Plate Methods for Enumeration of Rhizobium spp. in Inoculants Made from Presterilized Peat. Appl Environ Microbiol 46:1246-1247.

44. Hooper I, Whately B, Iyer VN, (1985) A study of Rhizobium meliloti JJI mutants in indole acetic acid production. In Nitrogen Fixation Research Progress, HJ Evans, PJ Bottomley, WE Newton, eds, Martinus Nijhoff, Amsterdam, p 148.

45. Kape R, Parniske M, Werner D (1991) Chemotaxis and nod Gene Activity of Bradyrhizobium japonicum in Response to Hydroxycinnamic Acids and Isoflavonoids. Appl Environ Microbiol 57:316-319.

46. Kosslak RM, (1984) Interactions between the soybean host (Glycine max) and its microsymbiont (Rhizobium japonicum) which affect competition between strains. PhD thesis. University of Hawaii, Honolulu.

47. Kosslak RM, Bookland R, Barkei J, Paaren HE, Applebaum EA (1988) Induction of Bradyrhizobium japonicum Common nod Genes by Isoflavones Isolated from Glycine max. Proc Natl Acad Sci USA 84:7428-7432.

48. Kosslak RM, Bohlool BB, Dowdle S, Sadowsky NJ (1983) Competition of Rhizobium japonicum strains in Early Stages of Soybean Nodulation. Appl Environ Microbiol 46:870-873.

49. Kosslak RM, Bohlool BB (1984) Suppression of Nodule Development of One Side of a Split-Root System of Soybean Caused by Prior Inoculation of the Other Side. Plant Physiol 75:125-130.

50. La Favre JS, Eaglesham ARJ (1984) Increased Nodulation of "Non-nodulating" (rj-rj) Soybeans by High Dose Inoculation. Plant Soil 80:297-300.

51. Layzell DB, Rainbird RM, Atkins CA, Pate JS (1979) Economy of Photosynthate Use in Nitrogen-fixing Legume Nodules. Plant Physiol 64:888-891.

52. Malik NSA, Bauer WD (1988) When Does the Self-Regulatory Response Elicited in Soybean Root after Inoculation Occur? Plant Physiol. 88:537-539.

53. Malik NSA, Calvert HE, Bauer WD (1987) Nitrate Induced Regulation of Nodule Formation in Soybean. Plant Physiol 84:266-271.

54. Malik NSA, Pence MK, Calvert HE, Bauer WD (1984) Rhizobium Infection and Nodule Development in Soybean are Affected by Exposure of the Cotyledons to Light. Plant Physiol 75:90-94.

55. Mathews A, Carroll BJ, Gresshoff PM (1991) Development of Bradyrhizobium infections in supernodulating and nonnodulating mutants of soybean (Glycine max L., Merrill). Protoplasma 150:40-47.

56. McDermot TR, Graham PH (1990) Competitive Ability and Efficiency in Nodule Formation of Strains of Bradyrhizobium japonicum. Appl Environ Microbiol 56:3035-3039.

57. Myeong-Je Cho, Harper JE (1990) Effect of Inoculation and Nitrogen on Isoflavonoid Concentration in Wild-Type and Nodulation-Mutant Soybean Roots. Plant Physiol 95:435-442.

58. Nutman PS (1948) Physiological Studies on Nodule Formation. I. The Relation between Nodulation and Lateral Root Formation in Red Clover. Annals of Botany 46:81-96.

59. Nutman PS (1949) Physiological Studies on Nodule Formation. II. The Influence of Delayed Inoculation on the Rate of Nodulation in Red Clover. Annals of Botany. 51:261-283.

60. Nutman PS (1952) Studies on the Physiology of Nodule Formation. III. Experiments on the Excision of Root-tips and Nodules. Annals of Botany 1:79-101.

61. Olsson JE, Nakao P, Bohlool BB, Gresshoff PM (1989) Lack of systemic suppression of nodulation in split root systems of supernodulating soybean (Glycine max L., Merr.) mutants. Plant Physiol 90:1347-1352.

62. Peters NK, Long SR (1988) Alfalfa Root Exudates and Compounds which promote or Inhibit Induction of Rhizobium meliloti Nodulation Genes. Plant Physiol 88:396-635.

63. Pankhurst CE (1986) Genetics and molecular biology of symbiotic nitrogen fixation by Rhizobium sp and R.japonicum. In WJ Broughton, ed, Nitrogen Fixation vol 4: Molecular biology, pp 194-222.

64. Pate JS (1977) Functional Biology of Dinitrogen Fixation by Legumes. In RWF Hardy, WS Silver, eds, A Treatise on Dinitrogen Fixation, Section III, Biology, J Willey & Sons, pp 473-517.

65. Pate JS, Atkins CA, (1983) Nitrogen Uptake, Transport and Utilization. In WJ Broughton, ed, Nitrogen fixation vol 3, Legumes, Oxford University Press, pp 245-298.

66. Pierce M, Bauer WD (1983) A Rapid Regulatory Response Governing Nodulation in Soybean. Plant Physiol 73:286-290.

67. Pueppke SG, Hawes MC (1985) Understanding the binding of bacteria to plant surfaces. Trends in Biotechnology 3, 12:310313.

68. Rolfe BG, Gresshoff PM (1988) Genetic Analysis of Legume Nodule Initiation. Ann Rev Plant Physiol Plant Mol Biol 39:297-319.

69. Sargent L, Huang SZ, Rolfe BG, Djordjevic MA (1987) Splitroot assays using Trifolium subterraneum show that Rhizobium infection induces a systemic response that can inhibit nodulation of another invasive Rhizobium strain. Appl Environ Microbiol 53:1611-1619.

70. Singleton PW (1983) A Split-root Growth System for Evaluating the Effect of Salinity on Components of the Soybean - Rhizobium japonicum Symbiosis. Crop Sci 23:259-267.

71. Singleton PW, Van Kessel C (1987) Effect of Localized Nitrogen Availability to Soybean Half-root Systems on Photosynthate Partitioning to Roots and Nodules. Plant Physiol 83:552-556.

72. Singleton PW, Stockinger KR (1983) Compensation Against Ineffective Nodulation in Soybean. Crop Sci 23:69-72.

73. Smith GB, Wollum II AG (1989) Nodulation of Glycine max by six Bradyrhizobium japonicum strains with Different Competitive Abilities. Appl Environ Microbiol 55:1957-1962.

74. Stacey G. (1990) Compilation of the nod, fix and nif Genes of Rhizobia and Information Concerning their Function. In PM Gresshoff, LE Roth, G Stacey and WE Newton, eds, Nitrogen Fixation: Achievements and Objectives, Chapman and Hall, New York, London, p 239.

75. Sturtevant DB, Taller BJ (1989) Cytokinin Production by Bradyrhyzobium japonicum. Plant Physiol 89:1247-1252.

76. Sutton WD (1983) Nodule development and senescence. In WJ Broughton, ed, Nitrogen fixation vol 3, Legumes, Oxford University Press, pp 144-178.

77. Takats ST (1990) Early autoregulation of symbiotic root nodulation in soybeans. Plant physiol 94:865-869.

78. Thies JE, Singleton PW, and Bohlool BB (1991) Modeling Symbiotic Performance of Introduced Rhizobia in the Field by Use of Indices of Indigenous Population Size and Nitrogen Status of the Soil. Appl Environ Microbiol 57:29-37.

79. Turgeon BG, Bauer WD (1982) Early events in the infection of soybean by Rhizobium japonicum. Time course and cytology of the initial infection process. Can J Bot 60:152-161.

80. Turgeon R, Wimmers LE (1988) Different Patterns of Vein Loading of Exogenous ¹⁴C-Sucrose in Leaves of Pisum sativum and Coleus blumei. Plant Physiol 87:179-182.

81. Van Brussel AAN (1990) Symbiotic Signals in Early Stages of the Morphogenesis of Rhizobium-Induced Nodules. Symbiosis 9:135-146.

82. Vargas AAT, Graham PH (1989) Cultivar and pH effects on competition for nodule sites between isolates of Rhizobium in beans. Plant and Soil 117:195-200.

83. Vesper SJ and Bauer WD (1985) Characterization of Rhizobium Attachment to Soybean Roots. Symbiosis 1:139-162.

84. Walsh KB, Vessey KJ, Layzel DB (1987) Carbohydrate Supply and N₂ Fixation in Soybean. The effect of varied daylength and stem girdling. Plant Physiol 85:137-144.

85. Wardlaw IF (1990) The Control of Carbon Partitioning in Plants. Transley Review No.27, New Phytol 116:341-381.

86. Wareing PF, Phillips IDJ (1982) Growth and Differentiation in Plants. Pergamon Press, Oxford, pp 105-137.

87. Wilkinson L (1988) SYSTAT: The System for Statistics. Systat, Inc., Evanston, Illinois

88. Woomer PL, Singleton PW, Bohlool BB (1989) Ecological Indicators of Native Rhizobia in Tropical Soils. Appl Environ Microbiol 54:1112-1116.