Biotic and abiotic factors influencing the development of N 2 -fixing symbioses between rhizobia and the woody legumes Acacia and Prosopis LEENA A. RÄSÄNEN Department of Applied Chemistry and Microbiology Division of Microbiology University of Helsinki Finland ACADEMIC DISSERTATION IN MICROBIOLOGY To be presented, with the permission of the Faculty of Agriculture and Forestry of the University of Helsinki, for public criticism in Auditorium 1041, at Viikki Biocenter 2, (Viikinkaari 5, Helsinki) on December 5 th , 2002, at 12 o’clock noon. Helsinki 2002
Transcript
Biotic and abiotic factors infl uencing the development of N2-fi xing symbioses between rhizobia and
the woody legumes Acacia and Prosopis
LEENA A. RÄSÄNEN
Department of Applied Chemistry and MicrobiologyDivision of Microbiology
University of HelsinkiFinland
ACADEMIC DISSERTATION IN MICROBIOLOGY
To be presented, with the permission of the Faculty of Agriculture and Forestry of the University of Helsinki, for public criticism in Auditorium 1041,
at Viikki Biocenter 2, (Viikinkaari 5, Helsinki) on December 5th, 2002, at 12 o’clock noon.
Helsinki 2002
Supervisor: Docent Kristina Lindström Department of Applied Chemistry and Microbiology University of Helsinki Finland
Reviewers: Prof. Ken Giller Department of Plant Sciences Wageningen University The Netherlands
Dr. David Odee Biotechnology Laboratory Kenya Forestry Research Institute Nairobi, Kenya
Opponent: Docent Marjo Helander Department of Biology University of Turku Finland
ISSN 1239-9469 ISBN 952-10-080-2 (paperback) ISBN 952-10-0821-0 (PDF)
Electronic publication available at http://ethesis.helsinki.fi
Yliopistopaino, Helsinki 2002
Front cover: Root hairs of Acacia senegal stained with methylene blue.Photo: Leena Räsänen.
Oi maailma,elämä sä ihmeellinen!Mi on sun tarkoitukses?Mihin viimein?Se arvoituspa kauan askaretta on aatokseen tuonut.
Aleksis Kivi
To my parents
Contents
List of the original papersThe author’s contribution Abbreviations
1.2. Characterisation of arid and semiarid regions.....................................................3 1.2.1. Dry tropical forests .................................................................................3 1.2.2. Soil types................................................................................................3 1.2.3. Saline soils ............................................................................................6
1.3. Role of N2-fi xing trees in dry tropical ecosystems...............................................6 1.3.1. Trees improve soil fertility ......................................................................6 1.3.2. Signifi cance of N2 fi xation.......................................................................7
1.4. The family Leguminosae (or syn. Fabaceae)........................................................8
1.5. Acacia and Prosopis trees (Mimosoideae) ..........................................................9 1.5.1. Geographical distribution .......................................................................9 1.5.2. Biological and ecological properties .......................................................9 1.5.3. Capacity for N2 fi xation ........................................................................12 1.5.4. Nodulation in the fi eld .........................................................................13 1.5.5. Multipurpose Acacia and Prosopis trees ...............................................14 1.5.6. Problems .............................................................................................15 1.5.7. Systematics of Acacia spp. ....................................................................16 1.5.8. Systematics of Prosopis spp. .................................................................17
1.6. Rhizobia nodulating Acacia and Prosopis trees................................................17 1.6.1. The most common species and genera among tree rhizobia .................18 1.6.2. Occurrence of rhizobia in deep soils ....................................................20
1.7. Development of symbioses between rhizobia and leguminous plants ..............21 1.7.1. Colonisation and attachment of rhizobia on the plant root....................21 1.7.2. Infection and nodulation ......................................................................23 1.7.3. A closer look at the infection process ..................................................24 1.7.4. The structure of nodules ......................................................................28 1.7.5. The symbiosome and the symbiosomal membrane ...............................30 1.7.6. N2 fi xation ............................................................................................30
1.7.7. Altruistic behaviour of N2-fi xing bacteroids...........................................31
1.8. Inhibition of N2-fi xing symbiosis by abiotic factors...........................................32 1.8.1. Abiotic constraints of arid and semiarid regions....................................32 1.8.2. Movement of water in the plant and soil...............................................33 1.8.3. Drought stress and drought tolerance strategies of the plants ...............34 1.8.4. Dehydration postponement in Acacia and Prosopis ..............................34 1.8.5. Salt stress and salt tolerance mechanisms of the plants .........................35 1.8.6. Heat stress and heat tolerance mechanisms of the plants ......................36 1.8.7. Osmotic adjustment and compatible solutes in plants ..........................36 1.8.8. N2 fi xation under stress .........................................................................37 1.8.9. Nodulation and infection under stress ..................................................38
1.9. Abiotic stresses and free-living rhizobia ................................................................40 1.9.1. Osmoadaptation in rhizobia .................................................................42 1.9.2. Role of trehalose during desiccation ....................................................45
1.10. Inhibition of N2-fi xing symbioses by biotic factors..........................................46
2. Aims of the study ....................................................................................................48
3. Materials and methods............................................................................................49
4. Results and discussion.............................................................................................53
4.1. Symbiotic properties of African sinorhizobia ...................................................53 4.1.1. Host specifi city.....................................................................................53 4.1.2. Variation in the effectiveness of the symbiotic association ....................54
4.2. Development of the symbiosis between sinorhizobia and Acacia and Prosopis trees ...............................................................................................55 4.2.1. Infection trough the root hairs...............................................................56 4.2.2. The structure of effective indeterminate nodules ...................................57 4.2.3. Ineffective nodules of A. holosericea and P. africana ............................57
4.3. Tolerance of sinorhizobia to abiotic stresses ....................................................58 4.3.1. Properties of two potential inoculant strains..........................................58 4.3.2. Tolerance of heat ..................................................................................58 4.3.3. Tolerance of salt ...................................................................................59 4.3.4. Roles of glycine betaine and trehalose for sinorhizobia ........................60
4.4. A. senegal-S. arboris symbiosis under heat and drought stress..........................61 4.4.1. Rhizobial population in soil..................................................................61 4.4.2. Adaptation of A. senegal seedlings to heat and drought ........................62 4.4.3. Infection and nodulation under stress ..................................................63
4.5. Role of compatible solutes on the A. senegal-Sinorhizobium symbiosis ...........64 4.5.1. Endogenous glycine betaine in A. senegal ............................................64 4.5.2. Effects of exogenous compatible solutes on A. senegal seedlings .........65 4.5.3. Effects of exogenous compatible solutes on rhizobia ............................66
4.6. Responses of S. arboris populations to changing environmental conditions......66 4.6.1. Alterations in EPS production................................................................67 4.6.2. Changes in cell activity and culturability ..............................................68 4.6.3. Moderately and highly stressed cultures................................................70 4.6.4. Changes in cell morphology .................................................................71
4.7. Is it possible to exploit symbiotic N2 fi xation by inoculation of Acacia and Prosopis seedlings? ..........................................................................................71 4.7.1. Evaluation of the role of inoculation in the fi eld....................................72 4.7.2. Inoculation as a procedure ...................................................................73
List of the original papersThis thesis is based on the following papers, referred to in the text by their Roman numerals. Additionally, some unpublished results are presented.
I Räsänen, L. A., J. I. Sprent and K. Lindström. 2001. Symbiotic properties of sinorhizobia isolated from Acacia and Prosopis nodules in Sudan and Senegal. Plant and Soil 235:193-210.
II Räsänen, L. A. and K. Lindström. 1999. The effect of heat stress on the symbiotic interaction between Sinorhizobium sp. and Acacia senegal. FEMS Microbiology Ecology 28:63-74.
III Räsänen, L. A., S. Saijets, K. Jokinen, and K. Lindström. Evaluation of the roles of two compatible solutes, glycine betaine and trehalose, for the Acacia-Sinorhizobium symbiosis exposed to drought stress. Submitted 2002.
IV Räsänen, L. A., A. M. Elväng, J. Jansson, and K. Lindström. 2001. Effect of heat stress on cell activity and cell morphology of the tropical rhizobium, Sinorhizobium arboris. FEMS Microbiology Ecology 34:267-278.
The author’s contributionI and II Leena Räsänen planned and conducted the experiments, analysed and interpreted the
results and wrote the paper, under supervision of the project leader Kristina Lindström. Janet Sprent acted as an expert on the symbioses between tropical leguminous trees and rhizobia and assisted in writing of paper I.
III Leena Räsänen planned and conducted the experiments analysed and interpreted the results and wrote the paper, under supervision of the project leader Kristina Lindström. Salla Saijets performed growth experiments on bacteria, foliar application of glycine betaine and preliminary drought experiments on plants. Kari Jokinen acted as an expert on the application of plant glycine betaine and assisted in writing.
IV Leena Räsänen planned and conducted the experiments, analysed and interpreted the results and wrote the paper, under supervision of the project leader Kristina Lindström. Annelie Elväng introduced analyses and measurements of luc tagged bacteria. Janet Jansson acted as an expert on the application of bacteria tagged with the marker genes, was the supervisor of Annelie Elväng and assisted in writing.
Abbreviations
ABA abscisic acid AM fungi arbuscular mycorrhizal fungiBD Brown and Dilworth mediumbv. biovarcfu colony forming unitCPS capsular polysaccharidesDMSP b-dimethylsulfonepropionateEPS exopolysaccharidesFAO Food and Agricultural Organizations of the United NationsGFP green fl uorescent proteinGUS ß-glucuronidase Hsps heat shock proteins LPS lipopolysaccharides NAGGN N- acetylglutaminylglutamine amideNAS National Academy of SciencesPEG polyethylene glycol SFDA fl uorochromo 5- (and)-sulfofl uorescein diacetatesHsps small heat shock proteinssubsp. subspeciesUSDA United States Department of AgricultureVBNC viable but nonculturableYEM yeast-mannitol
1
1. INTRODUCTION
1.1. BACKGROUND
Population growth together with recurrent drought periods have lead to deforestation and degradation of many dry ecosystems in the tropics. Activi-ties of the increased human population cause pressure on the environment, resulting in reduction of biological productivity and diversity. The gathering of fuelwood and production of charcoal, which have intensifi ed near large towns, overgrazing, shortening of forest fallow periods, cultivation of range-lands, and land clearance for agriculture and livestock husbandry are the main factors that reduce vegetation cover. Moreover, forced migration due to wars, population shifts or droughts have often pushed people to settle in areas that are alien to them, and therefore agricultural and cropping practices may be more destructive compared to those used by the local people (Kirmse and Norton 1984, Bellefontaine et al. 2000).
Reduction of vegetation and, especially, replacement of perennial trees by annual plants cause decreases in soil fertility, litter, and organic matter, subsequently increasing erosion. Natural, periodic droughts enhance this adverse development (Kirmse and Norton 1984). Land degradation does not only disturb natural plant communities but causes disturbances in plant-microbe symbioses, which are critical factors in helping further plant growth in degraded ecosystems (Requena et al. 2001).
Planting of trees is one important practical step to interrupt or retard this deleterious process. Agroforestry, in which crop plants are grown together with perennial, N2-fi xing trees, has traditionally belonged to the cultivation systems of many tropical areas. These trees were rediscovered in late the 1970’s and 1980’s when the governments and foundations all over the world funded the research of biological N2 fi xation after the energy crisis 1973. The term Multi-Purpose-Trees, including roughly 100 tree species, was invented at that time. Besides being suitable for the production of fuelwood and timber, rapid growth, production of human food and animal fodder, and the ability to improve soils and crops nearby, were their typical features (Smith and Smith 1992). In the beginning, new N2-fi xing tree species were needed for agroforestry purposes. This was important especially in Africa, where the consumption of mineral fertilisers was, and still is low, 2.2-3.9% of the global use (Dakora and Keya 1997). Nowadays, the need of trees with different properties has exploded. Large areas have to be brought back under forest cover in order to reverse the current trend of deforestation, to conserve bio-diversity (Khurana and Singh 2001) and to prevent the development of the global warming.
Acacia and Prosopis trees originating in arid and semiarid regions of the tropics and subtropics, are good candidates for reforestation of dry, marginal
2
or degraded lands with low fertility and/or high salinity. As drought and salt tolerant trees, they can grow in harsh conditions. Moreover, most Acacia and Prosopis species are capable of forming root nodules in symbiosis with root nodule bacteria called rhizobia. In the nodules, rhizobia fi x atmospheric nitrogen (N2) and transform it into a form that is available to the host plant. Legume symbioses have been estimated to contribute 70 million metric tons N per year out of the total global fi xation of 175 million metric tons per year. Approximately one half of N fi xed by the legumes is derived from the tropics and the other half from temperate zones (Brockwell et al. 1995).
In agriculture, crop legumes have been inoculated with appropriate, effective rhizobial inoculants in order to assure formation of N2-fi xing nod-ules and, subsequently to improve plant growth. However, whether the exploitation of the N2 fi xation capacity of leguminous trees, for example, by using inoculation or inoculated seedlings or seeds when planting trees to the fi eld, is largely unknown. More basic knowledge about the symbiosis and ecology of the associations between rhizobia and leguminous trees, such as Acacia and Prosopis spp., is still needed. An understanding of this relationship may facilitate growth of well N2-fi xing plants in fi elds and may help to develop appropriate fi eld transplanting techniques. Consequently, the amount of surviving tree seedlings in the fi eld will increase, which improves the chance to succeed in reforestation.
Table 1.1. Classifi cation of tropical climates adapted by Giller and Wilson (1991).
Climatic Temperature Rainfall Dry Geographic zone month-1 year-1 period distribution
Wet tropic > 18oC > 1 800 mm River basins of Amazon and Congo, lowland of South-East Asia
Wet and dry > 18oC 300 - A dry period Subhumid and tropics 1800 mm of at least semiarid tropics; most two months part of Asia, savannah areas of South America and Asia
Dry tropics < 300 mm Long dry Africa North of 15oN, period of Australia South several months of 15oS
Cool tropics -3oC - +18oC Variability At high altitudes in rainfall
3
1.2. CHARACTERISATION OF ARID AND SEMIARID REGIONS
The major climatic zones in the tropics are mainly distinguished by the amount and distribution of the rainfall throughout the year (Table 1-1). Many semiarid areas are characterised by large variations in rainfall. For example, in a semiarid desert in the Sudan, where the average rainfall is 185.1 mm, the rainfall ranged 1941-1980 between a minimum of 58.5 mm (1960) and a maximum of 442.0 mm (1978) per year (Ahmed 1986).
Arid and semiarid zones cover approximately four-tenths of the earth’s land area, and are inhabited by one tenth of the world’s population. They are mainly located in two subtropical belts. The northern part includes the Sahara and Sahel, Middle East, Southern Asia, and part of North America. The other is in the south, including part of Southern Africa, Australia and the dry zones of Peru, Chile, Argentina and Brazil (Olivares et al. 1988). Accord-ing to FAO (Bellefontaine et al. 2000), 66% of the world’s dry and very dry ecological zones lie within the African continent. Fifty-two percent of the dry deciduous forests are found in Africa, 23% in the Asia-Pacifi c region and 25% in Latin America.
1.2.1. Dry tropical forests
FAO (Bellefontaine et al. 2000) has defi ned dry tropical forests as those areas, where the annual rainfall over the past 10-15 years ranged between 300 and 1 200 mm, and where there were 5-10 dry months a year (rainfall < 30 mm per month). The dry tropical forests include the following vegetation types: dry deciduous forests, thickets, open woodlands, savannahs, and steppes (wooded, tree and shrub). This classifi cation is mainly based on the degree of the crown cover and the height of the woody vegetation (Figure 1-1). Acacia and Prosopis species are characteristic of areas where the trees grow more separately (Table 1-2).
1.2.2. Soil types
Generally, in large areas of tropical and subtropical Africa and South America continuous high temperatures with high rainfalls have caused extreme weath-ering and leaching of soils developed on old rocks. More fertile soils occur in areas that have a relatively recent addition of volcanic ash or alluvium containing minerals, or in climates in which the long dry season has retarded the rate of weathering (Giller and Wilson 1991, Lal 2000).
Although the ability of soils to support plant growth mainly depends on the surface soil horizons, the classifi cation of soils gives background information about the geographical distribution of different soil types, and advances and problems occurring in them (Giller and Wilson 1991, Lal
4
Figure 1-1. Types of woody or partially woody stands of dry tropical areas according to Letouzey (1982) and Bellefontaine et al. (2000).
5
2000). According to the classifi cation of USDA (Soil Survey Staff 1975), Ulti-sols, Oxisols, Alfi sols, Aridisols and Vertisols are the major soil types in arid and semiarid regions. Highly weathered Ultisols and Oxisols are found in parts of West Africa and Australia, which used to be wetter but are now dry. Alfi sols, containing low activity clay similarly to Ultisols and Oxisols, are widely distributed in the semiarid tropics (e.g. Sub-Sahara). Alfi sols and Ulti-sols have low water retention capacity, and are susceptible to soil erosion and compaction. Aridisols, affected by severe drought stress, contain high concentrations of CaCO3 and CaSO4, and cover a large area of Africa (Giller and Wilson 1991, Buresh and Tian 1998, Lal 2000). Vertisols with a high proportion of clay (2:1), are found for example in acid savanna, and cover large areas in India, Ethiopia and Sudan (Giller and Wilson 1991).
About 36% of the tropical soils are dominated by low nutrient reserves. These, usually highly weathered soils, have limited capacity to supply P, K, C, Mg and S (Sanchez and Logan 1992). Defi ciency of N is common, espe-cially in cultivated soils. In Africa, approximately 90% of the mineral N is found in living plants Lack of micronutrients occurs in places. Extensively weathered soils can have such a low pH that Al, and sometimes Mn, may become soluble and, thus toxic to the plants. Vertisols can fi x large amounts of P (Giller and Wilson 1991, Dakora and Keya 1997).
Table 1-2. Acacia and Prosopis species found in dry tropical forests according to FAO’s report (Bellefontaine et al. 2001). Dry areas with a Mediterranean climate and dry zones of Australia and the Pacifi c was not included in the report.
Domain Acacia and Forest type Occurrence Prosopis species
Dry deciduous forest Africa: Sudanian domain P. africana (in legume forests)
Thickets Africa A. ataxacantha (climbing acacia)
Open woodlands “ Zambezian domain Acacia spp.
Savannahs “ Sudania and Sahelian A. seyal, A. senegal, domains A. tortilis ssp. raddiana Asia: central Indian region Acacia spp.
Steppes Africa: Sahelian domain A. tortilis ssp. raddiana, A. senegal, A. ehrenbergiana Central America P. julifl ora Asia: north-west India Prosopis spp., Acacia spp.
6
1.2.3. Saline soils
Saline soils are formed under hot, arid and semiarid conditions due the accu-mulation of salts in the topsoil. They can form naturally or as the result of poorly managed irrigation or clearing from trees (Giller and Wilson 1991). Saline lands (e.g. salt desert) represent about 15% of the arid and semiarid lands of the world, whereas salt-affected cultivated soils represent about 40% of the world’s irrigated lands (Zahran 1997). Saline soils predominantly contain chlorides and sulfates of Na, Ca and Mg. Soils salinized with neutral, soluble salts of Na are termed saline-sodic. There, toxic excess of neutral salts prevents plants from normal growth (Gupta and Abrol 1990).
Many soils in arid and semiarid regions have CaCO3 in the soil profi le. If calcareous soils have an excess of exchangeable Na and high pH, caused by the presence of NaHCO3 and NaCO3, the soils are termed alkali soils. These soils have a poor physical structure, which together with Na toxicity adversely affects plant growth. The elements commonly observed in toxic concentrations include Na, Mo, Bo, and sometimes Se. Alkaline soils contain adequately P and Zn but otherwise they may suffer from low fertility (Gupta and Abrol 1990).
1.3. ROLE OF N2-FIXING TREES IN DRY TROPICAL ECOSYSTEMS
1.3.1. Trees improve soil fertility
Generally, savannah and desert soils have poor fertility. However, soil fertility and growth of annual vegetation are frequently superior under tree crowns compared to that of areas distant from the trees. The studies on African tropical semiarid savannah and on warm desert ecosystems in the Sonoran Desert, California, USA, showed that organic matter and N were concen-trated in the soil surface, in the savannah 0-10 cm and in the desert 0-30 cm (Barth and Klemmedson 1978, Bernhard-Reversat 1982, Virginia and Jarrel 1983, Belsky et al. 1989). Moreover, N and C concentrations were signifi -cantly higher in soils under tree or bush canopy than in open grassland or on dunes. Soil Ca, K, P (Virginia and Jarrel 1983, Belsky et al. 1989) and Mg
(Virginia and Jarrel 1983) also were highest in surface soil beneath the trees. The situation was similar with soil organic C, N, P and K in A. senegal planta-tions in Senegal. N and K contents increased in surface soils as the planta-tion aged (Deans et al. 1999). Five to 30-year-old P. julifl ora plantations founded in degraded alkaline soils in semiarid India also improved levels of organic C, N, P, Ca and Mg. In addition, growth of the Prosopis stands restored alkaline soils by decreasing soil pH, electrical conductivity (salinity) and exchangeable Na levels (Bhojvaid and Timmer 1998). By reducing solar radiation and soil temperatures and by improving the moisture status of the
7
soil the tree canopies created a more favourable microclimate compared to that of open fi elds (Belsky et al. 1989, Belsky 1994, Bhojvaid and Timmer 1998). Microbial biomass was also higher in soils from the canopy than in soils from the root or grassland zones (Belsky et al. 1989).
The origin of N and/or organic C accumulated under the tree canopy is not well understood. For example, it is not known whether the N has resulted from the symbiotic N2 fi xation, from animal deposits or exclusively from the root uptake of deeply located N reserves. It is not clear either, which fac-tors actually are responsible for the increased herbaceous-layer productivity under and near the tree (Bernhard-Reversat 1982, Belsky et al. 1989, Belsky 1994, Deans et al. 1999). However, it has been commonly agreed that the greater fertility of canopy soils is linked to deep-rooted trees and shrubs, which recycle nutrients from the depth to surface soils (Bernhard-Reversat 1982, Belsky et al. 1989, Buresh and Tian 1998). Buresh and Tian (1998) summarised soil improvements by trees into three categories: i) increased supply of nutrients through increased inputs and reduced outputs (reduced leaching of nutrients); ii) increased availability of nutrients through enhanced nutrient cycling and conversion of nutrients to more labile forms; and iii) a more favourable environment for plant growth through improved soil chemi-cal and physical properties.
1.3.2. Signifi cance of N2 fi xation
Methodological problems in the quantifi cation of N2 fi xation for adult trees are major reasons for the uncertainty of the precise source of N accumu-lated beneath the woody legumes (Buresh and Tian 1998). Searching of nodules from adult trees is also a very tedious work (Bernhard-Reversat 1982, Högberg 1986). An approach to estimate N2 fi xation is the comparison of the natural 15N (δ15N) abundance in tissue of presumed N2-fi xing plants with that of presumed non-N2-fi xing plants growing on the same site, and with that of soil N. 15N is usually more abundant in soils than it is in the atmosphere. Consequently, the 15N abundance would be lower in N2-fi xing plants than in non-N2-fi xing plants (Shearer et al. 1983, Högberg 1986).
The use of this technique has indicated that in some areas the N was derived from N2 fi xation. In the Sonoran Desert, N2 fi xation of Prosopis was concluded to be very important at six out of seven sites, the relative contribu-tion of the N2 fi xation being 43-61%. At the site where Prosopis were not N2-fi xers, N2 fi xation was important for papilionoid legumes (70%; Shearer et al. 1983). Along an aridity gradient in Namibia, the contribution of N2 fi xa-tion to the total N concentration of the mimosoid trees (e.g. several Acacia species, Faidherbia albida, P. glandulosa) was estimated to be about 30%, but the variation within and between species was large (Schulze et al. 1991). The mulga is an arid or semiarid ecosystem in Australia, in which A. aneura
8
(mulga) and other acacias with similar properties comprise the dominant arboreal element. The 15N abundance analysis showed that in the regular mulga, acacias do not fi x N2, apparently due to high amounts of soil mineral N. Instead, in an exceptional, heavily leached sand plain mulga acacias and two papilionoid legumes were nodulated and fi xed N2 (Pate et al. 1998).
Prosopis spp. (P. alba, P. tamarugo) were able to grow in the rainless region of the Atachma Desert, Chile. Lack of surface moisture inhibited leaf decomposition and prevented N cycling. However, the trees were able to extract water and nutrients from ground water through the deep roots, and fi xed N2 with nodules located in moist soil layers (Ehleringer et al. 1992). Herbaceous legumes grown under tree canopies can also fi x N2 (Bernhard-Reversat 1982, Pate et al. 1998).
1.4. THE FAMILY LEGUMINOSAE (OR FABACEAE)
Of all plants used by man, only the grasses (Gramineae) are more important than the legumes (Allen and Allen 1981). Leguminous plants are used as forage for the cattle, and their cultivation alternated with cereals or other plants improves soil fertility. Plant sources contribute about 70% of the human protein need, and leguminous seeds account for 19% of this amount (Peoples and Herridge 1990). With approximately 750 genera and nearly 20 000 species, Leguminosae is the third largest family of fl owering plants (Allen and Allen 1981). Its species are found throughout the world but the greatest diversity is found in the tropics and subtropics. Whereas herbaceous legumes are common in the temperate regions, most legumes in the tropics are trees and bushes. All legumes bear pods, the characteristic by which they most easily can be recognised (NAS 1979).
Based on mainly fl oral differences (Allen and Allen 1981), taxonomists have traditionally divided the Leguminoseae family into three subfamilies. The subfamily Caesalpinioideae contains approximately 2 500 - 2 800 spe-cies, which are mainly trees of the tropical savannahs and forests of Africa, South America and Southeast Asia. Approximately 3 000 species of the sub-family Mimosoideae are mostly small trees and shrubs of tropical and sub-tropical regions of Africa, North and South America, Asia and Australia. The subfamily Papilionoideae contains 12 000-13 500 species. They are mainly herbs, distributed world-wide (NAS 1979, Doyle 1994, Sprent 2001). Nodu-lation is common among the subfamilies Mimosoideae and Papilionoideae but within the Caesalpinioideae subfamily, only few species or genera can nodulate (Sprent 2001).
9
1.5. ACACIA AND PROSOPIS TREES (MIMOSOIDEAE)
Due to the ecological and economical importance and morphological and biological similarities, Acacia and Prosopis trees are very often studied together (Bukhari 1997a). The name Acacia is derived from Greek and means ”point” or ”barb” in reference to thorns. Only Australian acacias are thorn-less. The Latin name Prosopis is an ancient Greek name for burdock (Allen and Allen 1981). Like the Latin name reveals, Prosopis trees also have thorns, though thornless cultivars occur. The name mesquite is used for several Prosopis species in North America, whereas the name algarrobo is used commonly in South America (NAS 1979).
1.5.1. Geographical distribution
The genus Acacia is pantropical having a wide distribution in the tropics. However, in Australia, in which their main development has presumably occurred (Brenan 1965), acacias have invaded temperate regions (Norris 1956). Out of the over 1200 Acacia species (Chappill and Maslin 1995) about 850 are endemic to Australia, seven to South America and 135 to Africa, with some species also occurring in Asia (Maslin and Stirton 1997). Several Acacia species grow in the arid and semiarid regions of the tropics.
The genus Prosopis, a tree characteristic for the American continent, is not particular tropical (Brenan 1965), with some species growing in warm deserts of Texas, New Mexico, Arizona and California (Fisher 1977). The 40 species are distributed in arid and semiarid areas of North and South Amer-ica, and three species are found in Northern Africa and Eastern Asia. Most of the species (31 out of 44) are indigenous to South America. Only one spe-cies, P. africana, is indigenous to Africa (Burkart 1976). Like several tropical or subtropical trees, many Acacia and Prosopis species have spread under the infl uence of man from one continent to an other.
1.5.2. Biological and ecological properties
Information about the Acacia and Prosopis species used in this thesis is gath-ered in Table 1-3. In general, Acacia and Prosopis trees resemble each other in appearance. They are small or medium-sized trees (5-15 m) or shrubs. Only few species, e.g. A. sieberiana (Maydell 1990), can reach up to 25 m height. Leaves are typically small and bipinnate. In Australian acacias they are reduced after the seedling stage to a leaf-like petiole termed a phyl-lodium. In dry areas, the trees are drought-deciduous, but otherwise they tend to be evergreen (Roshetko 2001). Except Australian acacias, Acacia and Prosopis trees bear thorns, which may disturb their cultivation if farmers are not used to them. The fl owers are fragrant, showy globular heads, spikes
10
Table 1-3. Geographical distribution, biological and ecological properties, and utilisation of Acacia and Prosopis species used in this work. Annual Temperature Tree species Geographic origin rainfall (mm) range °C Height
Tribe Acaciae Acacia A. angustissima Central America 900-3000 5 - 30°C Shrub - (Mill.) Kunze - in tropical and 7 m - 6 varieties subtropical areas A. holosericea G. Northern Australia 200-1500 Mild frost - Shrub - Don - hot, subhumid zones 39°C 8 m A. mellifera Arabian Peninsula, Shrub (Vahl.) Benth. Egypt-South Africa - 9 m A. nilotica (L.) Del. Africa, Arabian 250-1500 -1 - +50°C Shrub - 9 subspecies Peninsula-India-Burma - 20 m A. oerfota (Forssk.) Africa Schweinf. A. senegal (L.) Willd. Africa, Arabia, Iran 100-950 -4 - +48°C 2-6 m - 4 varieties India, Pakistan
A. seyal Del. Semiarid zones 250-1000 5 - 55°C 9-17 m - 2 varieties of Africa
A. sieberiana DC. Semiarid zones 400-800 < 25 m of Africa
A. tortilis (Forsk.) Arid and semiarid 40-1200 0 - 50°C 4-20 m Hayne ”umbrella thorn” zones of Africa, Arabia, - 4 subspecies Near and Middle East
Tribe Mimoseae Prosopis P. africana (Guill. & Senegal-Ethiopia, 4-20 m Perr.) Taub. Egypt-Sudan-Lake Victoria
P. chilensis (Molina) Peru, Bolivia, Chile, 350-400 -4 - +45°C < 12 m Stunz Argentina
P. cineraria India, Pakistan, Iran 75-880 -4 - +51°C 5-18 m (L. Druce) Afghanistan, Arabia
P. julifl ora (Sweet) Central America, 150-750 -4 - +51°C Shrub DC Mexico, introduced in - 15 m - several varieties Africa and Asia
P. pallida (Willd.) Dry coastlines of Peru 250-600 < 18 m Kunth. and Ecuador
Other trees growing in arid and semiarid regions
Faidherbia albida (Del.) A. Dry Africa, Near East 300 -1800 -6 - +44°C 15-31 m Chev. syn. A. albida Del. ”miracle tree” of the Sahel
P. glandulosa Torrey Mexico to southern 200 < frost - 38°C 3-7 m - 2 subspecies U.S.A. ”honey mesquite” References: von Maydell 1990, Hocking 1993, Roshetko 2001
11
Uses Special features
Potential for fodder Thornless, earlier medicinal and agroforestry plant of Maya Indians Windbreaks, erosion control, potential Introduced recently in for fuel and fodder (dried phyllodes) Africa and South Asia Fodder, fuel, fences, medicine Baskets made from the roots Fodder, fuel, tannin, gum, Tolerates alkaline andtimber, medicine saline soils, evergreen Synonym A. nubica Benth. Gum arabic, fuel, fodder, honey, Drought resistant, usedfood, soil stabiliser, medicines in agroforestry in Sudan
Fodder, fuel, gum talha, Suitable tree for silvo-medicines pastoral systems
Fodder (pods), fences, tools and One of the biggest umbrella-implements, honey, windbreaks shaped acacias in Africa
Fodder, fuel, fences, timber, Well-known Acacia species, honey, windbreaks, medicine, particular drought resistant, food, dune stabiliser in India a good silvo-pastoral tree
Fuel, fodder, medicines, The only native Prosopis in arts, craft Africa, ”ironwood”
Fuel, timber, fodder (pods), If ground water, growth food, honey, windbreaks, shade with 250 mm rainfall
Fodder, agroforestry, timber, fuel, Drought resistant, deep rooting, tolerates dune stabiliser, honey, medicines moderate saline and alkaline soils
Fodder (pods), food, fuel, timber, Deep rooting, 35 m; honey, windbreaks, shades, tolerates alkaline and salinelive fence, dune stabiliser soils, easily weedy, evergreen
Agroforestry, fodder, tools and Deep rooting, retains leavesimplements, medicines, seeds through the dry season,are eaten in times of famine drought tolerant, fast growing
Food and beverage in Mexico, Moderately salt and frost, tolerant,fodder (pods), timber, fences drought resistant if the roots reachhoney, medicines the ground water
12
or short racemes, the colour varying from cream-white (Acacia) or greenish white (Prosopis) to bright yellow (Acacia, Prosopis; Allen and Allen 1981, Hocking 1993). Several species are good honey plants (Maydell 1990, Roshetko 2001). The length of pods depends on the species (3-30 cm; Hocking 1993, Roshetko 2001). Usually, the mature pods fall a short distance from the tree. Both wild and domestic animals, particularly goats and sheep, are effective long-distance dispersers of seeds (Ahmed 1986).
The temperatures on the growth sites of Acacia and Prosopis can vary within a wide range. Both tree genera tolerate day temperatures of 38-51°C but are usually sensitive to frost (e.g. A. nilotica, P. pallida), or tolerate only mild frost. Especially the seedlings are sensitive to low temperatures (Hock-ing 1993, Roshetko 2001). Acacia and Prosopis trees are not very fi re-resist-ant (Hocking 1993). However, the seeds of A. sieberiana and A. gerradii are released from dormancy by heat shock generated by fi re (Khurana and Singh 2001).
Many Acacia and Prosopis species are very drought resistant. However, their survival in low rainfall conditions (e.g. 40-50 mm y-1 for A. tortilis and 75 mm y-1 for P. cineraria; Hocking 1993, Roshetko 2001) or during long rainless periods is possible due to their long tap roots reaching the ground-water or moisture soil layers. The tap roots of several Prosopis species, A. senegal and A. tortilis can penetrate 10-35 m of soil (NAS 1979, Hocking 1993, Deans et al. 1999, Roshetko 2001). The deepest plant roots (P. velu-tina) were recovered in Arizona from an open-pit copper mine at a depth of 53 m (NAS 1979).
Several Acacia and Prosopis species are capable of growing in saline soils. P. articulata, P. pallida and P. tamarugo could grow and fi x nitrogen in 0.3 M NaCl (Felker et al. 1981). Highly salt tolerant Australian acacias (A. auriculiformis, A. stenophylla, A. maconochieana) could survive even at 1.7-1.8 M NaCl and showed shoot growth at 1 M NaCl (Aswathappa et al. 1987).
1.5.3. Capacity for N2 fi xation
In spite of the fact that several Acacia and Prosopis species form N2-fi xing nodules (Allen and Allen 1981, Sprent 2001), information about their N2-fi xing capacity is restricted due to methodological problems (Dakora and Keya 1997). Available data (Table 1-4) suggest that Acacia and Prosopis trees of arid and semiarid regions are not very effi cient N2-fi xers (6-34 kg N ha-1 y-1) compared to woody legumes originating from more humid climate (100 - 800 kg N ha-1 yr-1; Peoples and Herridge 1990, Dakora and Keya 1997). Nevertheless, low N2-fi xing capacity might be connected to their low growth rate in dry regions.
13
1.5.4. Nodulation in the fi eld
If mineral N is present in the soil, for example after fi re, woody legumes prefer soil N to N2 fi xation, and the nodulation is poor (Monk et al. 1981, Pate et al. 1998). The legumes develop nodules only if the soil is poor in N, because development, maintenance, and function of nodules require extra energy and other resources from the plant. Young Acacia and Prosopis trees and seedlings are usually nodulated if the soil is defi cient in N (Jenkins et al. 1988b), but it has been diffi cult to recover nodules from adult trees (Högberg 1986, Virginia et al. 1986). This has led to the question, whether these trees actually fi x N2 in the fi eld. Johnson and Mayeux (1990) gave three explana-tions for the failure to fi nd nodules under Prosopis canopies: i) nodules were formed but decomposed too rapidly to be detected. ii) The density of nodules on the root system was too low in order to detect them by the coring method (ø 4.7-10 cm). iii) The nodules were fed on by insects.
A reason for the apparent lack of nodules might be that in areas with wet and dry seasons, nodulation and N2 fi xation are seasonal, nodule activity being at its highest during the wet season (Monk et al. 1981, Langkamp et al. 1982, Hansen and Pate 1987). In Australia, Acacia stands depended on mineral N early and late in the growing season, whilst in the middle of the growing season 68-75% of the plant N came from nodule activity (Monk et al. 1981). It is generally believed that the tree nodules are perennial (Allen and Allen 1981). However, survival of Australian Acacia nodules through the dry period into the second rainy season was not common (Monk et al. 1981).
Table 1-4. Estimation of the proportion and amount of plant N derived from N2
fi xation for Acacia spp., Prosopis spp. and Faidherbia albida grown in arid and semiarid regions of the tropics (ARA, acetylene reduction assay; 15N, 15N-isotopic technique).
N derived Annual from N2 Species Location kg N ha-1 fi xation Method Reference
A. holosericea Australia 12 ± 4 ARA Langkamp et al. 1979 Australia 6 19% ARA Langkamp et al. 1982 Senegal 10-20 30% 15N Cornet et al. 1985
A. pennatula Mexico 34 ARA Peoples & Herridge 1990
F. albida Austria 20% 15N, in Sanginga et al. 1990 greenhouse
P. glandulosa Sonoran desert, 25-30 50% N content Rundel et al. 1982 U.S.A. with Kjeldahl, fi eld study
P. julifl ora Senegal 20 15N, pots Diagne and Baker 1994
14
Another reason for the absence of nodule observations might be that the deep-rooted woody legumes bear nodules in deep soil layers. Root nodules of P. glandulosa were found in depths of 3-7 m in Sonoran Desert (Jenkins et al. 1988a) and in depths of 0.01-2.5 m in Texas (Johnson and Mayeux 1990). The nodules of P. alba and P. tamarugo grown in Atacama Desert, Chile, were located in the moist soil layers, 0.5-1.5 m below the surface (Ehleringer et al. 1992). In addition, nodules were not regularly distributed but occurred in patches scattered both in the vertical and horizontal plane (Johnson and Mayeux 1990). It is not known whether the nodulation near the soil surface is inhibited by unfavourable conditions (e.g. low moisture content, high tem-perature) or due to the elevated N content.
1.5.5. Multipurpose Acacia and Prosopis trees
Mankind has utilised Acacia and Prosopis trees for thousands of years. In Biblical times acacias (probably A. seyal, A. tortilis) were used to build the Ark of the Covenant and the Table of Tabernacle (Allen and Allen 1981). Their wood was also used by ancient Egyptians for pharaoh’s’ coffi ns (NAS 1979). In historical times, mesquite (P. glandulosa) was a widespread and important resource of the native people in south-western North America. It was utilised for food, fuel, shelter, weapons, tools, fi bre, and dye. The mesocarp of the pods, being largely sucrose and prepared as fl our and further as a gruel, cakes and beverages, provided a major carbohydrate component in native diets (NAS 1979). P. cineraria, which is an important tree for desert dwellers in India, providing fuel and fodder, is still considered to be a religious tree (Hocking 1993).
Nowadays, Acacia and Prosopis trees are still of great importance for the inhabitants of the arid and semiarid regions. In Africa, the foliage and pods of acacias are important fodders both for domestic and wild animals. The digestible protein content is 8-15% for leaves and pods, respectively. They also are rich in minerals. The seeds are high in crude phosphorus, an ele-ment usually scarce in grasslands. In Australia, cattle and sheep browse harsh acacia phyllodes during the dry seasons, when other plants are scarce (Roshetko 2001). In the case of American Prosopis species, only pods can be used as fodder and food. Due to their sugar containing mesocarp, the pods are sweet, maximum glucose content being 27-35% (Maydell 1990, Roshetko 2001). In times of famine, powdered bark (e.g. P. cineraria) has been mixed with fl our. In some areas, dried seeds of A. senegal (Hocking 1993) or pods of A. nilotica (Maydell 1990) and P. cineraria (Roshetko 2001) are included in the human diet. However, pods and leaves of some acacia species contain cyanogenetic glycosides, which may yield free hydrocyanic acid and alkaloids toxic to livestock (Allen and Allen 1981).
In Africa, Acacia and Prosopis are important sources of fi rewood (calo-
15
rifi c value 3000-5000 kcal/kg) and charcoal (Hocking 1993). The wood is generally hard and dense, and therefore also used for house construction and manufacturing of furniture, tools and implements (Maydell 1990, Hock-ing 1993). Leaves, bark, gum, roots, pods and seeds are used medicinally against a wide variety of diseases, wounds and burns (Hocking 1993). Leaves, pods and bark may contain tannin. A. nilotica is an important source of tanning material in Sahel and India (Maydell 1990, Hocking 1993). Acacia and Prosopis trees are commonly used for windbreaks, stabilisation of sand dunes, shading, and ornamentals. Thorny species are used for fenc-ing (Maydell 1990, Hocking 1993, Roshetko 2001).
Several Acacia and Prosopis trees produce gum after they have been wounded. A. senegal is the source of gum arabic, which is a highly water-soluble gum of relatively low viscosity. It is safe as a food additive. Collec-tion of gum, tapping, consists of slitting the bark and collecting walnut-sized gum globules. Gum is formed at leaf-fall and is carried to roots even when the tree has not been tapped. An average annual yield is 250 g of gum per tree. A. seyal produces water-soluble gum thala, which is traditionally used for non-food applications (textile and paper manufacture, explosives; Allen and Allen 1981, Roshetko 2001).
1.5.6. Problems
Properties that help Acacia and Prosopis trees survive, propagate and spread under harsh growth conditions may cause problems in more favourable envi-ronments. In South-Western United States at the turn of the 20th century, after the big herds of cattle had grazed for 30-50 years, the population of Prosopis plants (P. velutina, P. glandulosa) was so increased that they were considered as an agricultural pest. At the same time, their growth form changed from single or few-stemmed tall trees to shrubby thickets (Fisher 1977). A similar phenomenon has occurred in Argentina, Paraguay (Fisher 1977), and South Africa, in which Prosopis spp. were introduced at the turn of 20th century (Zimmermann 1991).
Below the soil surface Prosopis plants have a zone of buds which remain dormant as long as the primary bud or tree grows and develops normally. If the growth of the seedling or tree trunk is destroyed by freezing, drought, fi re, cutting or trampling, underground buds will initiate new growth, and multi-stemmed plants will form (Fisher 1977). Prosopis trees produce large quan-tities of seeds (Zimmermann 1991) which facilitate their effective spread, if there are no limiting factors. Elimination of natural periodic prairie fi res, reduction of natural grass cover, increased dissemination of seeds by large herbivores, and reduction or absence of seed-feeding predators contributed to the increased Prosopis densities in USA and South Africa (Fisher 1977, Zimmermann 1991). Drought resistant acacias (e.g. A. holosericea, A. nilot-
16
ica, A. tortilis) can also become weedy when introduced out of its native range, e.g. in more humid zones, or if their growth is not restricted (NAS 1979, Zimmermann 1991, Hocking 1993).
1.5.7. Systematics of Acacia spp.
The genus Acacia was fi rst systematically described in 1745. The major clas-sifi cation of acacias occurred 1875, when Bentham divided the subfamily Mimosoideae into fi ve tribes, Ingeae, Mimoseae, Mimozygantheae, Parkieae and Acacieae. Within the tribe Acaciae, the genus Acacia contained six series (Table 1-5) Chappill and Maslin 1995, Robinson and Harris 2000). The fi rst major realignment to this classifi cation occurred 1972, when Vassal (1972) viewed Acacia as a single genus consisting of three subgenera, Acacia, Aculeiferum and Phyllodineae (Table 1-5). This is perhaps the most commonly referred classifi cation of Acacia spp. so far, although numerous studies on their phylogeny and taxonomy have followed. However, no bigger rearrangements have been done, because the taxonomy of acacias as well as other mimosoid legumes is in a state of fl ux. The systematics of Acacia spp. is not simple, and their classifi cation (and nomenclature) will change in the future (Chappill and Maslin 1995, Robinson and Harris 2000).
There are inconsistent opinions, how the three subgenera, Acacia, Aculeiferum and Phyllodineae, are related to each other. According to the cladistic analysis by Chappil and Maslin (1995), based on morphological and physiological properties, the subgenera Aculeiferum and Phyllodineae were closely related, and subgenus Acacia differed clearly from them. Con-
Table 1-5. Outlines of the major classifi cations and distributions of Acacia spp. according to Robinson and Harrison (2000). Distribution Bentham (1875) Vassal (1972) Robinson and Harrison (2000) of Acacia spp.
Subfamily Mimosoideae
Tribe Mimoseae Tribe Mimoseae Tribe Mimozygantheae Tribe Parkiae Tribe Acaciae Tribe Acaciae Genus Acacia Genus Acacia series Gummiferae subgenus Acacia subgenus Acacia America, Africa, Asia and Australia series Vulgares subgenus Aculeiferum subgenus Aculeiferum America, Africa, series Filicinae Asia and Australia tribe Ingae series Botrycephalae subgenus Phyllodineae subgenus Phyllodineae Australia, New series Phyllodineae Guinea, Hawaii series Puchellae Mascarena Islands Faidherbia albida Faidherbia albida Africa Tribe Ingeae
17
tradictory, Bukhari et al. (1998, 1999) suggested according to the analysis of seed storage protein size variations and chloroplast phylogeny that the sub-genus Aculeiferum was ancestral to the subgenera Acacia and Phyllodineae. Based on a plastid DNA phylogeny, Robinson and Harris (2000) recently proposed, that the subgenera Acacia and Aculeiferum are sister taxa whilst the subgenus Phyllodineae is sister to tribe Ingeae (containing e.g. Calliandra spp.). Moreover, the tribe Acacieae may contain the genus Acacia only with two subgenera, Acacia and Aculeiferum (Table 1-5). The differences in inter-pretation of the analyses might be due to that some plant characters included in the tests have arisen more than once during the evolution, whereas others have been lost (Chappill and Maslin 1995, Robinson and Harris 2000).
Faidherbia albida (synonym A. albida), is a popular acacia in agroforestry because it is capable of maintaining leaves during the dry season. Its taxo-nomic status has been particularly equivocal. Vassal (1972) moved Acacia albida to the monotypic genus Faidherbia within the tribe Acacieae (Table 1-5). Several subsequent studies supported this division while others did not (Chappill and Maslin 1995, Bukhari et al. 1998, 1999). In the recent work of Robinson and Harris (2000) F. albida appeared to be basal to the tribe Ingeae and to the subgenus Phyllodineae (Table 1-5).
1.5.8. Systematics of Prosopis spp.
Burkart (1976) has recognised fi ve distinctive groups of species (sections) in the genus Prosopis (Mimosoideae). According to his view, Prosopis is an old genus, which diverged early into several principal lineages, some containing very similar species and others forming clearly distinct groups (sections). The systematics of Prosopis spp. is otherwise confusing because hybridisations are common (Allen and Allen 1981, Zimmermann 1991). As a consequence, several Prosopis species have been confused in the literature and/or in prac-tice: P. julifl ora and P. glandulosa (NAS 1979), P. chilensis and P. alba , P. chilensis and P. julifl ora (NAS 1979, Roshetko 2001). Many studies indi-cate that Acacia and Prosopis trees might be closely related (Chappill 1995, Bukhari 1997a, Bukhari et al. 1998). For example, A. oerfota (former A. nubica) was found to be distantly related to the subgenus Acacia (Bukhari et al. 1999), and fi tting better in Prosopis than in Acacia (Bukhari 1997b).
1.6. RHIZOBIA NODULATING ACACIA AND PROSOPIS TREES
Root- and stem-nodule bacteria belong almost exclusively to the α-subclass of Proteobacteria. The major rhizobia are currently divided into about 30 species within six genera: Allorhizobium, Azorhizobium, Bradyrhizo-bium, Mesorhizobium, Rhizobium and Sinorhizobium (Wang and Martinez-Romero 2000). However, recently new types of rhizobia have been found
18
from tropical legumes. Bacteria belonging to the genus Methylobacterium, exhibiting both nodulation ability and methylotrophic properties (grew facul-tatively e.g. on methanol), were isolated from Crotalaria legumes (Papilio-noideae; herbs and shrubs; Sy et al. 2001). Devosia sp. was isolated from stem-associated nodules of the aquatic legume Neptunia natans (Mimosoi-deae; Rivas et al. 2002). Moreover, nodulation of tropical legumes by Burkholderia sp. (Moulin et al. 2001) and Ralstonia sp. (isolated from Mimosa spp.; Chen et al. 2001), members of the β-subclass of Proteobacte-ria, indicated that wide-range of bacteria can form symbioses with legumes.
Traditionally, rhizobia have been divided into fast- and slow-growers (i.e. rhizobia and bradyrhizobia) but even this general idea may change. The term meso means intermediate growth habit between fast and slow-growers and thus, the name Mesorhizobium denotes both phylogenetic position and growth rate (Chen et al. 1995, Jarvis et al. 1997). Studies with Kenyan rhizobia showed that strains belonging to Bradyrhizobium spp. exhibited very slow to intermediate growth, those belonging to Mesorhizobium spp. showed intermediate to fast growth and strains belonging to Rhizobium spp. and Sinorhizobium spp. grew fast or very fast (Odee et al. 2002).
Compared to the taxonomy of well-known rhizobia nodulating crop leg-umes (one plant species usually is nodulated by one or few rhizobial spe-cies), the taxonomy of rhizobia nodulating woody legumes is generally confusing, because there is no strict host specifi city between the tree legume and the microsymbiont. A wide range of rhizobia, including both fast- and slow-growers, can induce nodules on them. However, the spectrum of rhizo-bia capable of forming effective nodules on their hosts, is more restricted (Zhang et al. 1991, de Lajudie et al. 1994, Odee et al. 1997, Nick et al. 1999a, b, Lafay and Burdon 2001).
1.6.1. The most common species and genera among tree rhizobia
Present data suggest that in dry Africa, acacias (subgenera Acacia and Aculeiferum) and the introduced Prosopis spp. are commonly nodulated by fast or moderately fast-growing rhizobia, and especially, by those belonging to the genera Sinorhizobium and Mesorhizobium (references in Table 1-6). Slow-growing bradyrhizobia have been isolated only occasionally from root nodules of African acacias (Dreyfus and Dommergues 1981, Odee et al. 1997). However, Bradyrhizobium species were capable of forming effective nodules on A. seyal (Odee et al. 2002). F. albida, which is distantly related to other acacias (Vassal 1972, Robinson and Harris 2000) seems almost solely to prefer Bradyrhizobium spp. (Table 1-6).
Australian acacias belonging to the subgenus Phyllodineae (Vassal 1972, Robinson and Harris 2000) and occurring in more temperate climate (Norris 1956) prefer slow-growing Bradyrhizobium species (Dreyfus and
19
Dommergues 1981, Barnet et al. 1985, Table 1-6). In a recent study, 96.6% of the 118 bacteria isolated from the nodules of 13 different Acacia spe-cies belonged to Bradyrhizobium sp. (Lafay and Burdon 2001). The situa-tion might be similar in other legumes since most of 735 rhizobial strains (94.3%) isolated from shrubby legumes of temperate south-eastern Australia were identifi ed as Bradyrhizobium sp. (Lafay and Burdon 1998). However,
Table 1-6. Rhizobial species isolated from the nodules of Acacia and Prosopis trees growing in arid and semiarid regions of the tropics.
Species Host plants References
AfricaBradyrhizobium sp. F. albida (Se, Ke), A. nubica (Ke) Dupuy et al. 1994; Odee et al. 2002 B. elkanii F. albida (Se, Ke) Dupuy et al. 1994; Odee et al. 2002 B. japonicum F. albida (Se, Ke) Dupuy et al. 1994; McInroy et al. 1999Mesorhizobium sp. A. polycantha; A. xanthophloea, Haukka et al. 1998; Odee et al. 2002 A. tortilis (Ke) huakuii A. arenaria (Ke) Haukka et al. 1998, McInroy et al. 1999 loti A. senegal, A. tortilis subsp. heteracantha McInroy et al. 1999 heteracantha (Ke) plurifarium Acacia sp., A. senegal, A. seyal, A. tortilis de Lajudie et al. 1998 subps. raddiana, P. julifl ora (Se, Su)Rhizobium sp. A. senegal (Su) Nick et al. 1999a R. tropici A. karroo (Ke) McInroy et al. 1999Sinorhizobium sp. A. cyanophylla, A. gummifera, A. horrida Khbaya et al. 1998; Nick et al. 1999b A. tortilis subsp. raddiana (Mo); A. senegal (Su), P. chilensis (Ke) S. arboris A. senegal. P. chilensis (Su, Ke) Nick et al. 1999b S. kostiense A. senegal. P. chilensis (Su) Nick et al. 1999b S. saheli A. senegal. P. chilensis (Su) Nick et al. 1999b S. terangae bv. acaciae A. horrida, A. mollissima, A. tortilis subsp. de Lajudie et al. 1994; Lortet et al. raddiana, A. senegal (Se); A. senegal (Su), 1996; Nick et al. 1999b P. chilensis ( Ke)
AsiaMesorhizobium loti A. catechu, A. nilotica (North India) Kumar et al. 1999
AustraliaBradyrhizobium sp. A. obliquinervia Lafay and Burdon, 1998 B. elkanii Several Acacia spp. Lafay and Burdon, 2001 B. japonicum A. saligna; several Acacia species Marsudi et al. 1999; Lafay and Burdon, 2002 spp. (lupinus) A. saligna; Marsudi et al. 1999 Rhizobiumleguminosarumbv. phaseoli A. saligna Marsudi et al. 1999 tropici A. saligna; several Acacia species Marsudi et al. 1999; Lafay and Burdon 2001
Latin AmericaMesorhizobium chacoense P. alba Velazquez et al. 2001Sinorhizobium sp. Prosopis sp. (Mexico) Haukka et al. 1998 P. julifl ora (Brazil) Haukka et al. 1998; Moreira et al. 1998
Abbrevisions: A., Acacia; P., Prosopis; Ke, Kenya; Mo, Morocco; Se, Senegal; Su, Sudan
20
also fast-growing rhizobia, showing similarity with R. leguminosarum bv. phaseoli and R. tropici have been identifi ed from root nodules of Austral-ian acacias (Table 1-6).
The taxonomy of rhizobia nodulating Prosopis trees on the American continent is still poorly known. The fi rst studies indicated that some dry growth sites of Prosopis spp. were occupied only by fast-growing rhizobia (Jenkins et al. 1988b, Milnitsky et al. 1997), whereas the other sites were occupied by fast and slow-growers (Jenkins et al. 1987, 1989, Waldon et al. 1989). Two strains belonging to Sinorhizobium sp. have been identifi ed from Prosopis nodules in Brazil and Mexico (Table 1-6). Hence, a new Mesorhizo-bium species, M. chacoense has been characterised from the nodules of P. alba trees grown in central Argentina (Velazquez et al. 2001).
1.6.2. Occurrence of rhizobia in deep soils
As discussed in Chapter 1.5.2, several Acacia and Prosopis species occu-pying arid and semiarid regions develop deep root systems, located at a depth of 4-40 m. Apparently, deep-rooted trees develop two distinct lateral root systems, one near the surface, the other exploiting deep-water resources (Jenkins et al. 1987). Recovery of nodules in deep soils indicates that these soils contain rhizobia. Indeed, under F. albida in Sahelia, 103 rhizobia g-1 of soil were found as deep as 34 m below ground (Dupuy and Dreyfus 1992). In the warm deserts of USA, rhizobia capable of inducing N2-fi xing nodules on Prosopis spp. were observed at a depth of 4-13 m (Jenkins et al. 1987, 1988a, 1989). Generally, population densities of Prosopis rhizobia were substan-tially greater (103-107 cells g-1 of soil) in deep soil layers than were at dry soil surfaces (< 10 cells g-1 of soil; Virginia et al. 1986, Jenkins et al. 1988a). The situation was similar with bradyrhizobia under F. albida grown either in dry or moist sites (Dupuy and Dreyfus 1992).
Interestingly, rhizobial population at different depths under Prosopis trees generally consisted of both fast-growers and slow-growers. Although the con-centration of salts, P and N decreased with increasing depths (Virginia et al. 1986, Jenkins et al. 1988a, Ehleringer et al. 1992), no single factor (e.g. soil moisture, nutrient levels) explained cell concentrations or the distribution of fast- and slow-growers across the desert ecosystems (Jenkins et al. 1988a, 1989). According to the physiological tests, slow-growers from phreatic soils and those from surface soils formed two distinct groups (Waldon et al. 1989). Genetic studies indicated that the fast and slow-growers fell into soil zone-correlated groups of divergent but related isolates (Thomas et al. 1994). The recently characterised Argentinean species, M. chacoense, which effectively nodulated at least P. alba, P. chilensis and P. fl exuosa, was described as slow-growing. Colonies (1-3 mm in diameter) appeared on yeast-mannitol (YEM) agar after incubation for seven days at 28ºC; generation time 10-24 h in YEM
21
broth (Velazquez et al. 2001). Thus, perhaps the slow-growers found under other Prosopis trees represented Mesorhizobium sp.
1.7. DEVELOPMENT OF SYMBIOSES BETWEEN RHIZOBIA AND LEGUMINOUS PLANTS
The development of the symbiotic relationship between rhizobia and legumes is a highly interactive process that include i) molecular communication between the free-living organisms, ii) an infection phase when rhizobia enter root or stem nodules and iii) a fi nal symbiotic phase when rhizobia occupying nodules fi x N2. The dialogue between a legume and rhizobia starts when the plant releases organic compounds that attract rhizobia near their roots. The plant also produces fenolic compounds, fl avonoids, which induce the expres-sion of nodulation (nod) genes of rhizobia. The nod genes encode nodulation factors (lipochitin oligosaccharides) which in return induce early events of the infection process in the host plant, such as changes in the root epidermis, deformation and curling of root hairs (if it has hairs), and cell divisions in the root cortex (= initiation of nodule primordium). During formation and function of the nodule, the host plant expresses nodule-specifi c or nodule-enhanced genes, nodulin genes (Fisher and Long 1992, Kijne 1992, Phillips 1999).
Depending on the species, rhizobia can nodulate several different plant species (broad host range) or only one or few plant species (narrow host range). Some bacteria have even adapted to a particular plant cultivar. Both rhizobial Nod factors and plant fl avonoids are involved in the host specifi city. Each rhizobium/plant species produces several, various Nod factors/fl avonoids, and differences in their sets and amounts infl uence host specifi city. Other host recognition systems are also required to promote proper development of N2-fi xing nodules (Brewin 1998, Hadri and Bisseling 1998, Hadri et al. 1998).
In general, the plants try to protect themselves against attacks of patho-genic bacteria and other microbes, for example, by generation of H2O2 (called oxidative burst). During a proper Rhizobium-legume association, this hyper-sensitive reaction is controlled by decomposing H2O2. (Bueno et al. 2001). The sugars present on cell surfaces (i.e. exopolysaccharides, EPS; capsular polysaccharides, CPS; lipopolysaccharides, LPS) are essential for rhizobia to avoid host defence responses during the invasive phase (Brewin 1998, Hadri and Bisseling 1998, Albus et al. 2001). Albus et al. (2001) proposed recently that LPSs released from the bacterial surface function as specifi c signal to the plant, and promote symbiotic interaction and suppress defence responses.
1.7.1. Colonisation and attachment of rhizobia on the plant root
Rhizobia are polarly or subpolarly fl agellated (Bradyrhizobium) or peritri-chously fl agellated (Rhizobium, Sinorhizobium) and thus, they can move.
22
Non-motile and non-fl agellated cells are common in laboratory grown cultures (Yost and Hynes 2000). The motility is especially important when rhizobia compete for nodule occupancy (De Troch and Vanderleyden 1996).
In the rhizosphere, rhizobia migrate toward plant roots by chemotactic response. Rhizobia are nonspecifi cally chemotactic to a wide range of attract-ants (e.g. dicarboxylic acids, amino acids, sugars, aromatic and hydroaro-matic acids). The fl avonoids attract rhizobia more specifi cally (De Troch and Vanderleyden 1996, Yost and Hynes 2000). Attachment of rhizobia on the roots and root hairs is a complex process, including surface factors of bacteria (e.g. rhicadhesin protein, Ca2+ ions, cellulose fi brils, CPS, LPS, EPS , cyclic 1-2 β-glucans) and the plant (lectins and other adhesins (De Troch and Vander-leyden 1996, Matthysse and Kijne 1998). According to the lectin recognition hypothesis, plant lectins mediate specifi city in the Rhizobium-legume sym-biosis. Although this hypothesis was presented already in the early 1970s
Table 1-7. Major features of the three infection modes. Subfamilies of the plants according to Allen and Allen (1981). M=Mimosoideae, P=Papilionoideae.
Penetration Induced cell Penetration Spread inside into the root divison in the cortex nodule
1A Through root Inner cortex Infection Infection hairs and via threads threads infection threads 1B Outer cortex Between cells Short infection threads, host cell divisions
2A Crack entry Middle and Between cells Infection inner cortex threads
2B ? Between cells Short infection threads
2C Inner cortex Between cells Though altered cell walls
3 Between cells Inner cortex Between cells Infection thread of intact like structures epidermises
23
(Hirsch 1999), it was shown only recently that a novel root lectin, present on the root surface of the legume Dolichos bifl orus, is capable of binding rhizo-bial Nod factors (Etzler et al. 1999). Closely related lectin homologs were also found from other legumes (Roberts et al. 1999).
1.7.2. Infection and nodulation
The early stages of the symbiosis, when rhizobia enter the plant root, pene-trate in the root cortex and spread inside nodules, vary between legume species (Table 1-7). In general, rhizobia can enter the root i) through root hairs via a plant derived tunnel, named as an infection thread, ii) by crack entry, or iii) between cells of intact epidermises. In the root cortex, rhizobia penetrate towards the nodule primordium through transcellular infection threads or by colonisation intercellular spaces. Inside the nodule, bacteria
Noduletype Plant species References
Indeterminate Medicago sativa, Pisum sativa Kijne 1992, Brewin, 1998 Hadri Trifolium. spp., Vicia hirsuta; et al. 1998; Lipsanen & Lindström Galega spp.; (P) temperate 1988, Räsänen et al. 1991 herbsDeterminate Glycine max; Lotus corniculatus; Turgeon & Bauer 1982, 1985, Kijne Lotus japonicus; Phaseolus 1992, Hadri et al. 1998; Vance et al. vulgaris; Vigna radiata; (P) 1982; Schauser et al. 1999; tropical, subtropical 1994; Tate et al. 1994 Cermola et al. and temperate herbs 2000; Newcomb & McIntyre 1981
Indeterminate Mimosa scabrella (M) de Faria et al. 1988, Sprent & tropical tree de Faria 1988
24
may spread via infection threads, by entering through altered cell walls and/or together with the division of host cells (Figure 1-2, Table 1-7).
There are two main nodule types: elongated indeterminate nodules and speherical determinate nodules (Figure 1-3). The indeterminate type is characterised by a persistent apical meristem, while the determinate type lacks such a meristem. Indeterminate nodules are elongated and cylindrical because new cells are constantly being added to the distal end of the nodule. The spherical form of determinate nodules mainly results from cell enlarge-ment, while cell division ceases early (Hirsch 1992). Typically, indeterminate nodules are induced in the inner cortex whereas determinate nodules origi-nate in the outer cortex (Table 1-7).
It is generally assumed that the nodule morphogenesis, the mode of infec-tion and the uptake of rhizobia in nodule cells are under the control of the host plant (Dart 1977, Kijne 1992). According to Boogerd and van Rossum (1997), this view is supported by the following evidences: i) nodulating and non-nodulating phenotypes are conserved among taxonomically related legume species. ii) The nodule type bears botanical taxonomic relevance. iii) One and the same strain may infect different host legumes by different pathways.
It appears that infection through root hairs is common for temperate legu-minous herbs, whereas plants infected by crack entry grow mainly in the trop-ics (Table 1-7). Indeterminate nodules are common on woody legumes in the Mimosoideae and on temperate herbaceous legumes, whereas determinate nodules are found mostly among tropical leguminous herbs (Table 1-7).
1.7.3. A closer look at the infection process
This section is based on the data about the well-known hair infection route (Figure 1-2), but similar features presumably occur also in other infection modes. Roots and root hairs are only transiently susceptible to rhizobia (Bhuvaneswari et al. 1981). The most frequently nodulated region is located above the zone of rapid root elongation and below the smallest emergent hairs present at the time of inoculation (Turgeon and Bauer 1985). Usually, not all hairs are deformed but deformed hairs have a patchy distribution on the roots (Vandenbosch et al. 1985). Infections also occur in localised groups (Dart 1977). Although numerous hairs deform, only few of them become infected, and only a small percentage of the infection sites result nodules. For example, 10 M. sativa seedlings were estimated to have 80 000 hairs. From them, only 52 infection threads were found, and 27 nodules were formed (Wood and Newcomb 1989). In the case of Trifolium spp. and Vicia hirsuta, 1.4 - 38% of infected hairs resulted in nodules (Dart 1977). The host plant tends to restrict the number of infections and nodulations by eliciting hypersensitive reactions, such as oxidative burst (Vasse et al. 1993, Santos et al. 2001).
25
Figure 1-2. Diagrammatic presentations of the different infection modes among legumes. Drawings are based on the references mentioned in Table 7-1. Figures are not drawn to scale.
26
Root hairs occur only inconsistently. Rhizobia penetrate the mucigel m, and primary layers of radial walls, pw, of epidermal cells, and enter between host cells possibly by digesting the middle lamella, ml. The primordium is induced in the inner cortex (a). Rhizobia penetrate deeper within a matrix (b). In the nodule, multiplication of bacteria results in cell walls being pushed inwards, allowing intracellular penetration of bacteria (c). Inside the cell, bacteria pultiply, and remain enclosed by cell wall material. Structures resembling infection threads cross cell boundaries, infecting adjacent cells (d). pw = primary cell walls, sw = secondary cell walls.
27
28
In many cases, rhizobia can penetrate the hair cell or root epidermis, when bacteria are compressed between two adjacent epidermal cells of the hair or/and the root (Kijne 1992). Obviously, rhizobia are capable of enter-ing the hair through a completely eroded hole, which they have created by localised enzymatic hydrolysis of the host cell walls (Mateos et al. 2001). The plant-derived infection thread consists of similar compounds as the plant cell wall (cellulose, xyloglucan, methyl esterifi ed pectins, non-esterifi ed pectins) but it is more resistant to cell wall-degrading enzymes. The infection thread matrix contains at least proline-rich proteins and glycoproteins (Verma and Hong 1996, Brewin 1998).
1.7.4. The structure of nodules
In fully developed nodules two major tissue types can be recognised: the peripheral tissue and the central tissue (Figure 1-3). In general, the periph-eral tissue consists of large vacuolated cortical cells (outer cortex), a layer of small sclerenchymatic cells (nodule endodermis), and several layers of small, densely packed, vacuolated cells (parenchyma or inner cortex). The vascu-lar bundles and the vascular endodermis are completely surrounded by the parenchyma (Figure 1-3). The cortical cells outside the nodule endodermis might be modifi ed in a variety of ways, e.g. by corky cells with thickened walls or by lenticels, which play role in gaseous diffusion (Robertson and Farnden 1980). In the determinate nodule, the endodermis is located around the infected central tissue (Tate et al. 1994; Figure 1-4), whereas in the case of indeterminate nodule with apical meristem the endodermis stop short before the apex (Dart 1977).
The central tissue of the indeterminate nodules contains large infected cells and smaller, vacuolated, uninfected cells (Brewin 1991, Hirsch 1992). In M.
29
sativa, the central tissue consists of several well-defi ned histological regions (Figure 1-3), in which bacteroids are differentiated gradually by fi ve steps. From the apical part to the basal region, the following zones can be distin-guished: the bacteria-free meristematic zone I, the infection zone II, in which bacteria are released from infection threads, the narrow interzone, II-III, the N2-fi xing zone III and the senescent zone IV (Robertson and Farnden 1980, Vasse et al. 1990), in which host cells autolysis leads to the senescence of all bacteroids (Paau et al. 1980, Vance et al. 1980). Timmers et al. (2000) recently described an additional, saphrophytic zone V, locating proximal to the senescent zone IV. In this new zone, bacteria were released passively from remaining infection threads and invaded plant cells, which were already com-pletely senesced. These intracellular rhizobia did not resemble morphologi-cally bacteroids but were rod shaped. They could not fi x N2 and were not subjected to the host cell autolytic processes. Thus, they behaved like sapro-phytic bacteria (Timmers et al. 2000). From fi ve different types of bacteroids, only one type, locating in distal zone III, can fi x N2 (Vasse et al. 1990).
The zones above described have not been found in determinate nodules (Dart 1977; Figure 1-4). Therefore, it has been supposed that the cells in the central tissue are proximately in a similar developmental stage (Hadri et al. 1998). However, according to a recent study invasion of nodule cells is not synchronous, and not all cells of the central tissue differentiate simultane-ously (Cermola et al. 2000). The uninvaded cell aggregates are distributed among infected cells so that they are in contact with all invaded cells (Tate et al. 1994). The structure of the aeschynomenoid-type determinate nodule, such as that of Arachis hypogaea, groundnut, differed from other determinate nodules. The endodermis-like cell layer was absent in the nodule cortex, and the infected central tissue lacked interspersed uninfected cells (Boogerd and van Rossum 1997).
30
1.7.5. The symbiosome and the symbiosomal membrane
Inside nodules, rhizobia are taken into the host cytoplasm by endocytosis (Robertson and Farnden 1980), Brewin 1998). When the thread reaches a target cell, synthesis of the infection thread wall stops and the bacterium being released is surrounded only by the host plasma membrane (= infection droplets; Verma and Hong 1996, Brewin 1998; Figure 1-2). In some woody legumes, such as in Andira spp. (Papilionoideae; Faria et al. 1986) and in all genera examined in the subfamily Caesalpinioideae (Faria et al. 1987) rhizobia are not released but develop the N2-fi xing capacity within tubular threads.
After bacteria have been released into the nodule cytoplasm, the plasma membrane transforms to the symbiosomal (peribacteroid) membrane. The entity, consisting of the bacteroid(s) and symbiosomal membrane is called the symbiosome. Extension of the membrane, due to the subsequent growth and division of bacteroids, is achieved by inclusions of vesicles derived from the Golgi or endoplasmic reticulum (Roth and Stacey 1989, Verma and Hong 1996, Brewin 1998). The symbiosomal membrane provides a physi-cal barrier between the host cytoplasm and the bacterial cell and controls the exchange of substrates and signal molecules between the two partners. For example, it allows import of inorganic minerals and reduced carbon to bacteroids and export of fi xed N and heme for leghemoglobin synthesis to the plant cytoplasm (Verma and Hong 1996).
After bacteroids have ceased division, they develop the capacity of N2 fi xation. In indeterminate nodules, this phase is accompanied by enlarge-ment and morphological differentiation of bacteroids. This phenomenon has not been detected in determinate nodules and fi xation threads (Vasse et al. 1990, Hadri et al. 1998). During the differentiation, bacteroids have to adapt to new physiological conditions (i.e. low oxygen content, low pH; Brewin 1998). Each symbiosome contains various numbers (1-20) of bacteroids, depending on the host species and age of the nodule (Mellor and Werner 1990). It seems that in indeterminate nodules of temperate legumes, bacter-oids are singly enclosed by the symbiosomal membrane, whereas in deter-minate nodules each symbiosome contains groups of bacteroids (Dart 1977). The fusion of small symbiosomes to form a large one partly explains the existence of multi-bacteroid symbiosomes (Cermola et al. 2000).
1.7.6. N2 fi xation
Nitrogenase is the enzyme that catalyses the reduction of N2 into NH3 in all N2-fi xing organisms. The rhizobial nitrogenase consists of two components, dinitrogenase (FeMo protein) and dinitrogenase reductase (Fe) protein. The nitrogenase catalyzes the following reaction:
N2 fi xation is an extremely energy-consuming reaction. Therefore, symbi-otic N2 fi xers that benefi t from plant photosynthesis, mainly contribute bio-logically fi xed N2 on the earth (Kaminski et al. 1998). In isolated bacteroids, 95% of the fi xed N2 was reported to be excreted into the medium as ammo-nia, the remainder being incorporated into bacteroids (Robertson and Farn-den 1980). Fixed N2 is exported from bacteroid side across the symbiosome membrane to the plant cytoplasm as ammonium (Tyerman et al. 1995) and as amino acids (Rosendahl et al. 2001). From the nodules the fi xed N is exported to the plant as amides (e.g asparagine) or ureides (Boogerd and van Rossum 1997). Nitrogenase catalyses the reduction of H2, which is an inhibi-tor molecule for N2 fi xation. The bacteroids of some rhizobial strains are capable of removing H2 by oxidation, with the help of uptake hydrogenase (Mellor and Werner 1990, Bergersen 1997).
Nitrogenase is inactivated rapidly by excess O2 (Bergersen 1997). There-fore, several physiological and biochemical components regulate O2 supply to nodules, such as a thin diffusion barrier in the inner nodule cortex, the presence of O2-binding leghaemoglobins in the host cytosol, and respira-tion by large numbers of bacteroids and plant mitochondria (Bergersen 1997, Kaminski et al. 1998). Leghemoglobin is the predominant plant protein of the nodule (up to 30%), giving usually a red colour characteristic for the effective nodule (Kaminski et al. 1998).
Symbiotic N2 fi xation has a high demand for carbon sources. In addition to nodule growth and bacterial proliferation, carbon compounds are essen-tial for the generation of ATP and reducing power needed for nitrogenase. Carbon skeletons are needed for the assimilation of fi xed ammonia (Kaminski et al. 1998). The photosynthates enter the nodule as sucrose via the phloem. In indeterminate nodules, bacteroids take carbon mainly as dicarboxylic acids (mainly succinate), apparently after conversion of sucrose through a fermenta-tion pathway (Mellor and Werner 1990, Kaminski et al. 1998).
Free-living rhizobia have observed to fi x N2 under special conditions. Several Bradyrhizobium strains and some Rhizobium strains were able to fi x N2 (but did not grow) when the cells were cultivated under microaerophilic conditions on defi ned medium containing approriate N2 source (Kaminski et al. 1998).
1.7.7. Altruistic behaviour of N2-fi xing bacteroids
In indeterminate nodules, the senescence of both nodules and bacteroids has been considered as the end point of the symbiotic interaction. How then can N2 fi xation be evolutionary important for rhizobia, if they die when the
32
nodules degrade? According to a theory, rhizobia have an altruistic behav-iour; they sacrifi ce their lives in order to help the reproduction of others (Jimenez and Casadesus 1989), for example by enhancing production of root exudates into the soil (Olivieri and Frank 1994). Because one or few cells can initiate a nodule (Dart 1977), the rhizobial population within the nodule can be considered as a genetically homogenous clone (Brewin 1998). It has been shown that vegetative rhizobia, capable of saprophytic growth in soil and nodule debris, were present in infection threads of the wild type nodules and in ineffi cient nodules (Paau et al. 1980, Vance et al. 1980). Thus, although bacteroids die when the nodule degrades, bacteroids benefi t their undiffer-entiated relatives occupying the same nodule by producing more nodule debris, i.e. food, through N2 fi xation (Jimenez and Casadesus 1989, Olivieri and Frank 1994, Brewin 1998). Recently, Timmers et al. (2000) suggested that one function of “altruistic” bacteroids is to sustain nodule lifetime so that nodule zone V forms and becomes occupied by intracellular, saprophytic rhizobia. In all above-mentioned cases, the apparent suicide of N2-fi xing bacteria would result in the increase of the local rhizobial population when rhizobia are released from degraded nodules (Olivieri and Frank 1994, Timmers et al. 2000).
1.8. INHIBITION OF N2-FIXING SYMBIOSIS BY ABIOTIC FACTORS
The following two phenomena are familiar for the researchers working with N2-fi xing Acacia and Prosopis trees growing in arid and semiarid regions. First, seedlings grow better and fi x more N2 in optimal greenhouse condi-tions (but lacking mineral N) compared to seedlings of similar ages grown in natural ecosystems (Hansen and Pate 1987). Secondly, although no nodules are found on the root systems of the trees, soils sampled from their rhizo-sphere induce nodules on the same species in optimal greenhouse condi-tions (Barnet and Catt 1991). It is assumed that several abiotic factors in the soil, such as water stress, high soil temperature, salinity, nutrient defi ciencies, alkalinity and acidity may limit growth, nodulation and N2 fi xation of the legumes grown in the tropics (Reddel 1993, Dakora and Keya 1997, Hungria and Vargas 2000).
1.8.1. Abiotic constraints of arid and semiarid regions
Lack of water is the single most important factor that limits growth and symbiotic N2 fi xation of leguminous plants growing in arid and semiarid regions. In areas where evapotranspiration exceeds rainfall, saline soils may also become a problem due to the accumulation of salts into the topsoil. In addition, high soil temperatures may have detrimental effects. In the sections 1.8, 1.9 and 1.10 effects of these major constraints on Acacia and Prosopis
33
trees, their symbiotic partners and the development of symbiosis between them are discussed. In order to understand how the Acacia and Prosopis trees can survive in harsh environments, the general mechanisms associated with tolerance of plants to water, salt and heat stresses are also examined.
1.8.2. Movement of water in the plant and soil
In plants, water is continually lost from the leaf to the atmosphere when stomata are open to allow the uptake of CO2 for photosynthesis. The water lost by transpiration is replaced by water uptake from the soil through the root, stem and leave via the xylem (Turner 1986). Transport of water is driven by physical forces. Only the development and maintenance of structures needed for water transport require an energy input.
Plant physiologists have used the water potential (ψw) concept to defi ne transport of water across plant membranes, and as a measure of the plant water status. Several separable components can contribute to the plant water potential:
ψw = ψs + ψp + ψg (+ ψm)
The osmotic (solute) potential ψs represents the effect of dissolved solutes on the water potential. The term ψp is the hydrostatic pressure. When it refers to the positive hydrostatic pressure within cells, ψp is usually called turgor pressure. The gravity causes water to move downward unless the force of gravity is opposed by an equal and opposite force. Thus, the potential for water movement, ψg , depends on the height of plants. The matric potential ψm has been used to describe the reduction of free energy of water when it exists as a thin surface adsorbed onto the dry soil particles, seeds and cell walls. At the plant cellular level, signifi cant components of the water poten-tial are the osmotic potential (due to dissolved solutes) and the hydrostatic pressure (Taiz and Zeiger 1998).
The water potential of wet soils may be divided into two major compo-nents, the osmotic potential ψs and the hydrostatic pressure ψp. In general, ψs
is negligible because solute concentrations of the soil water are low. Instead, ψs is signifi cant in saline soils. The hydrostatic pressure ψp is close to zero for wet soils, but when the soil dries, it decreases (getting negative values). Dry soils contain so little free water that it is not practical to measure the soil ψs+ψp. In that case the soil water potential is defi ned by the soil matric poten-tial, caused by the capillarity and interaction of water with solid surfaces of the soil (Taiz and Zeiger 1998).
Water fl ow, especially in rapidly transpiring plants, is a passive process. Water moves toward regions of low water potential or free energy, from soil and roots to the leaves and the atmosphere. Accordingly, the water potential decreases consistently when water is moving from the soil to the atmosphere.
34
However, the components of the water potential can vary at different parts of the pathway. Diffusion, bulk fl ow and osmosis across membranes also help move water from the soil through the plant to the atmosphere (Turner 1986, Taiz and Zeiger 1998).
1.8.3. Drought stress and drought tolerance strategies in plants
In meteorological terminology drought is an absence of rainfall for a period long enough to cause depletion of soil moisture and damage to plants (Kramer 1980). The drought is permanent in arid regions whereas in areas with well-defi ned dry and rainy periods it is seasonal. Drought has also been defi ned as an environmental factor that produces water defi cit or water stress in plants (Kozlowski and Pallardy 1997). As the soil dries, its matric potential becomes more negative, decreasing the soil water poten-tial. The plants can continue to absorb water only as long as their water potential is more negative than that of the soil (Taiz and Zeiger 1998). Water defi cit initiates the development of a low plant water potential and falling of cell turgor below its maximum value (Kozlowski and Pallardy 1997). Consequently, reduced cell turgor causes closure of stomata, and reduction in cell enlargement, thereby affecting several structures and func-tions, e.g. decrease in leaf surface area, transpiration, photosynthesis and plant growth (Kramer 1980, Taiz and Zeiger 1998).
Drought resistance mechanisms of the plants can be divided into three types: i) drought-escaping plants complete their productive life cycle before the dry season begins; ii) dehydration postponement occurs by means of different morphological and physiological mechanisms that reduce tran-spiration or increase absorption of water, e.g. thick cuticle, leaf rolling, responsive stomata and deep root systems (Kramer 1980, Turner 1986); iii) dehydration tolerance, defi ned as the ability of cells to continue metabo-lism at low leaf water status, is considered to arise at the molecular level and to depend on the membrane stability and enzyme activity (Turner 1986, Turner et al. 2000).
Maintenance of turgor during a change in the water status is essential for maintenance of metabolic processes in plant cells. Turgor can be main-tained i) by maintaining water uptake, ii) by reducing water loss, and/or iii) by osmotic adjustment. Moreover, abscisic acid and other phytohormones are connected to the control of transpiration and water loss (Turner 1986, Turner et al. 2000).
1.8.4. Dehydration postponement in Acacia and Prosopis
Many Prosopis and some Acacia trees can maintain a constant water uptake by developing deep taproots by which they are capable of reaching moist
35
soil layers or the ground water (Nilsen et al. 1983; see section 1.5.2). One of the most general mechanisms to reduce water loss is a reduction in leaf area, due to either reduction in leaf area development and/or leaf senes-cence (Turner 1986). African acacias (subgenera Aculeiferum and Acacia) are generally deciduous trees, dropping their leaves during the dry season. Australian acacias (subgenus Heterophyllum) produce bipinnate leaves when they are seedlings, but later the leaves are replaced by variously shaped, highly lignifi ed phyllodes accompanied by thick cuticles. The advantage of the phyllodes is supposed to reside in their capacity to persist during elon-gated periods of drought (Ullmann 1989). Then, the phyllodes can response rapidly to the fi rst rains and start photosynthesis without using resources for the production of new foliage (Montagu and Woo 1999).
Stomatal closure is another way to reduce water loss through transpiration (Turner 1986). Acacia and Prospis trees have been observed to have both diur-nal and seasonal stomatal regulation (Nilsen et al. 1983, 1986, Ullmann 1989, Montagu and Woo 1999). Some acacias exposed to drought stress were capa-ble of developing a greater water use effi ciency, the ratio between dry matter produced and water consumption (Nativ et al. 1999). Mechanisms related to the osmotic adjustment, which plays an important role in the drought toler-ance are discussed more precisely in the section 1.8.7.
1.8.5. Salt stress and salt tolerance mechanisms of the plants
Salinity reduces plant growth and photosynthesis due to the complex effects of osmotic, ionic, and nutritional interactions, which are, however, still poorly understood (Shannon 1997). In A. cyanophylla, the salinity decreased the concentrations of shoot N, P and K (Hatimi 1999). Saline soils may impose specifi c ionic effects on plants because high concentrations of ions (Na+, Cl-, SO4
2-) accumulate in cells, inactivating enzymes and inhibiting protein synthesis and photosynthesis. High concentrations of dissolved salts in the rooting zone also generate a negative osmotic potential that lowers the soil water potential. The general water balance is then affected, because leaves need to develop lower water potential in order to maintain a “down-hill” gradient of water potential between the soil and the leaves. To prevent loss of turgor, most plants are capable of adjusting osmotically (Taiz and Zeiger 1998; see section 1.8.7). Other salt tolerance mechanisms may be based on accumulation or exclusion of ions.
Salt sensitive plants (glycophytes) try to restrict ion movement from roots to shoots whereas salt resistant plants (halophytes) tend to take up Na+ ions (Hasegawa et al. 2000). Many Acacia and Prosopis trees can grow in saline soils. Low foliar Na contents of a few salt tolerant species studied suggested that the trees were capable of excluding Na (Virginia and Jarrel 1983, Marcar et al. 1991).
36
1.8.6. Heat stress and heat tolerance mechanisms of the plants
Few higher plants can survive a steady temperature above 45oC. At high temperatures both photosynthesis and respiration are inhibited, but photo-synthesis is repressed before respiration. High temperature impairs the thermal stability of membranes, breaking the balance between the relative strengths of hydrophobic and hydrophilic interactions among proteins, lipids and the aqueous environment. As a result, membrane compositions and structures change, causing leakage of ions.
In environments with intense solar radiation and high temperatures, plants avoid heating of their leaves by decreasing their absorption of solar radiation (by leaf waxes, leaf rolling and vertical leaf orientation). The same mechanisms protect plants from water defi cit. Heat shock proteins are con-nected to the improved heat tolerance but their precise role is still unclear (Taiz and Zeiger 1998).
1.8.7. Osmotic adjustment and compatible solutes in plants
Osmotic adjustment is a process by which the water potential of the plant can be decreased without a decrease in accompanying turgor (Taiz and Zeiger 1998). Turner et al. (2000) have defi ned osmotic adjustment as an active accumulation of solutes by the plant in response to increasing water defi cits in the soil and/or plant, thereby maintaining turgor or reducing the rate of turgor loss, as water potentials decrease. Osmotic adjustment enables plants to survive and grow also in saline environments. The osmotic adjust-ment occurs then either through compartmentation of toxic ions away from cytoplasm into the vacuole and/or through accumulation of organic solutes, such as compatible solutes, in the cytosol (Hasegawa et al. 2000). With the aid of the osmotic adjustment the plants exposed to water defi cit or high salinity can maintain photosynthesis, delay leaf senescence and death, and improve root growth (Turner 1986, Turner et al. 2000).
Compatible solutes, accumulated during the osmotic adjustment, are highly soluble compounds that carry no net charge at physiological pH, and are non-toxic at high intracellular concentrations and at higher tempera-tures. Under unfavourable osmotic conditions, compatible solutes raise the osmotic pressure in the cytoplasm and stabilise proteins and membranes (Yancey et al. 1982, Csonka 1989, McNeil et al. 1999). The osmoprotective mechanisms of the compatible solutes are supposed to relate to the absence of perturbing effects on macromolecule-solvent interactions. Inorganic ions, such as NaCl, interact directly with proteins, favouring unfolding and leading to denaturation of proteins, whereas compatible solutes are excluded from protein surfaces (Rhodes and Hanson 1993, McNeil et al. 1999).
The plants can synthesise three types of compatible solutes: i) betaines
37
or quaternary ammonium compounds (e.g. glycine betaine) and tertiary sulfonium compounds (e.g. β-dimethylsulfonepropionate, DMSP in marine micro-algae), ii) polyols and sugars (e.g. trehalose) and iii) amino acids (e.g. proline) (McNeil et al. 1999). The plant can synthesise several various com-patible solutes, the composition of the set varying between species. Proline is the most common osmolyte among plants (Erskine et al. 1996). Glycine betaine accumulation occurs in diverse marine algae and in about 10 dis-tantly related fl owering plant families. The family Chenopodiaceae has sev-eral species that accumulate glycine betaine, e.g. sugar beet, Beta vulgaris and spinach (McNeil et al. 1999).
1.8.8. N2 fi xation under stress
Numerous studies have shown that water stress, salinity and high soil temper-ature reduce dramatically N2 fi xation and nitrogenase activity of nodulated leguminous herbs (reviewed by Zahran 1999). Similar phenomena have been observed in acacias and other woody legumes (Marcar et al. 1991, Purwantari et al. 1995, Zou et al. 1995). Although environmental stress factors disturb normal photosynthesis, nitrogenase activity may not be limited by the lack of photosynthetates (Delgado et al. 1993, Serraj et al. 1998). Nitrogenase activity can even be more sensitive to a decrease in soil water potential than stomatal closure (Pararajasingham and Knievel 1990).
During stress experiments, several changes in nodule metabolism and compounds were associated with the reduced N2 fi xation or nitrogenase activity, decreased hemoglobin content resulting reduction in O2 availability being the most common observation (Table 1-8). It is not clear, how stress conditions cause a decline in N2 fi xation. One of the common hypotheses is that the effects of environmental stresses are mediated through variations in nodular gas diffusion resistance (Sutherland and Sprent 1993). More pre-cisely, the rate of N2 fi xation may be limited by the supply of O2 (Walsh 1995). In water-stressed nodules, O2 supply to bacteroids may be restricted in two ways: i) packing of different layers of the cortical cells limits diffu-sion of O2 (Guerin et al. 1990), and ii) alkaline proteases degrade O2-binding leghemoglobin (Guerin et al. 1991). P enriched in the nodule cortex may also associate with part of the O2 barrier (Franson et al. 1991).
A decrease in N2 fi xation may involve variation in the balance between the export of nitrogenous solutes via the xylem and the import of photosynt-hetates via the phloem (Walsh 1995). According to the proposal of Walsh (1995), stress events modify the content of soluble sugars and nitrogenous solutes in the nodule cortex, resulting in changes in cortical cell turgor. Sub-sequently, this event affects for example the shape of the cortical cells and the intercellular water content, further changing permeability of the nodule cortex to gases. Hence, because nodule anatomy and physiology are hetero-
38
geneous among legume plants, the responses of N2 fi xation to environmental stresses may be a species-specifi c property (Walsh 1995).
1.8.9. Nodulation and infection under stress
Several studies both on herbaceous and woody legumes have shown that drought, salinity and heat stresses cause decrease in nodule numbers and nodule dry weight (e.g. (Singleton and Bohlool 1984, Zahran and Sprent 1986, Arayangkoon et al. 1990, Marcar et al. 1991, Purwantari et al. 1995, Hatimi 1999). At least in the case of salt stress, symbiotic properties of the legumes (N2 fi xation, nodule dry weight, nodule number) were more sensi-tive to salinity than plant growth (Marcar et al. 1991). If the stress conditions
Table 1-8. Effects of drought, salinity and high temperatures on nodule compounds and structures in herbaceous legumes.
Decreased nitrogenase Symptoms activity a) Stress
Nodule compoundsDecrease in leghemoglobin content + Drought + + Salinity + Decrease in water potential and content + Drought Decrease in lipids & proteins + Drought Decrease in organic acids ND d) Salinity Decrease in P conc. & P-use effi ciency + Drought Decrease in bacteroid respiration + Salinity Increase in proline content + Drought ND Salinity Increase in alkaline proteolysis + Drought Increase of nodule intracellular pH + Increase in soluble sugars + Increase in ureide concentration ND Increase in total carbohydrates ND ND Salinity Increase in cytosolic PEPC f) + Increase in bacteroid MDH g) +
Nodule structureDisturbances in bacteroid development ND Temperature
Nodule senescence ND Temperature Oxidative damage in lipids and proteins + Drought
a) Nitrogenase activity has been determined by acetylene reduction method, 15N dilution method or by measuring N content. b) Conditions with harmful effects are indicated in boldface. c) Whw = With holding water. d) ND = not determined. f) PEPC, nodule cytocolic phosphoenolpyruvate carboxylase. g) MDH, bacteroid malate dehydrogenase
39
are long or/and strong enough, formation of nodules will cease completely. It is generally believed that the infection process is the most sensitive phase during nodule development (Singleton and Bohlool 1984). Once nodules have formed, they are less affected by stress factors than the initial nodule formation process (Purwantari et al. 1995). Studies with herbaceous legumes having the root hair infection mode indicated that drought and salinity stresses cause changes in root hair morphology and decrease the numbers of markedly curling hairs (Table 1-9).
In mildly water-stressed Vicia faba nodules, the loss of turgor induced a reduction in the peribacteroid space, the vesicle membrane became sinous, and the cell cytoplasm appeared granular. Under more severe water stress, a total disappearance of peribacteroid spaces was associated with the rupture
Growth conditions b) c); growth medium Plant species References
-0.04 & -0.17 MPa; soil Glycine max Ruiz-Lozano et al. 2001Whw 2-9d: -0.8,- -2.4 MPa; sand Vicia faba Guerin et al. 19910.05 M NaCl; Leonard jars Pisum sativum Delgado et al. 19930.05, 0.1 M NaCl; Leonard jars Pea, faba bean, bean Delgado et al. 1994Whw 0, 2, 4-8 d; 20-4%; soil Vigna unguiculata Pararajasingham & Knievel 1990 -0.04 & -0.17 MPa; soil Glycine max Ruiz-Lozano et al. 20010.1, 0.15 M NaCl, solution Medicago sativa Fougere et al. 1991Whw 12-13d; soil Glycine max Franson et al. 19910.05, 0.1 M NaCl; Leonard jars Pea, faba bean Delgado et al. 1994 -0.04 & -0.17 MPa; soil Glycine max Ruiz-Lozano et al. 20010.1, 0.15 M NaCl, solution Medicago sativa Fougere et al. 1991Whw 2-9d: -0.8,- -2.4 MPa; sand Vicia faba Guerin et al. 1991 “ “ “ “ “ “ “ “-0.04 & -0.17 MPa; soil Glycine max Ruiz-Lozano et al. 2001Dehydration of the soil mixture Glycine max Serraj et al. 1998 “ “ “ “ “ “ “ “0.1, 0.15 M NaCl, solution Medicago sativa Fougere et al. 19910.05 M NaCl; Leonard jars Pisum sativum Delgado et al. 1993 “ “ “ “ “ “ “ “
22, 30oC; agar slopes Trifolium subterraneum Pankhurst & Gibson 1973
22, 30oC; agar slopes Trifolium subterraneum Pankhurst & Gibson 1973 -0.04 & -0.17 MPa; soil Glycine max Ruiz-Lozano et al. 2001
40
of vesicle membranes. The bacteroid content appeared to be very heteroge-neous (Guerin et al. 1991). Use of mineral N may improve tolerance to some stresses. For example, N2-fi xing acacias were more sensitive to salinity than N-fertilised plants (Marcar et al. 1991).
1.9. Abiotic stresses and free-living rhizobia
Desiccation of bacteriaRhizobia living in the rhizosphere of Acacia and Prosopis trees are obviously exposed to similar environmental stresses as their hosts. When the soil dries, the availability of water is assumed to be reduced at least due to increased solute concentrations of the soil water (Miller and Wood 1996). According to Potts (1994), the immediate environment of the cell, being either the atmos-phere or aqueous solution, distinguishes matric and osmotic systems. He has defi ned desiccation as the removal of a substantial fraction of the cell water through matric water stress. Desiccation can also be designated as air drying, because bacterial cells lose water when they are exposed to a gas phase with a water activity that is lower than that of the cell compartment. Desic-cation plays a decisive role in bacterial communities that are found in aero-phytic environments, on and in rocks and soils, in the phyllosphere, dusts and aerosols (Potts 1994). The responses of bacterial cells to desiccation can be: shrinkage of the bacterial cytoplasm and capsular layers, increase
Table 1-9. Effects of drought, salinity and high temperature on the infection process and nodulation in herbaceous legumes (PEG, polyethylene glycol).
Growth conditions a) and Symptoms Stress medium used
Root hairs Decrease in hair numbers Salinity 0.2, 0.4, 0.6% NaCl; Jensen tubes Decrease in rhizobial colonisation Salinity 1.0% NaCl; silica sand Decrease in deformation Salinity 1.0, 1.2, 2% NaCl; silica sand Decrease in mark curling Osmotic 0.1, 0.2 M PEG; sand Salinity 0.1 M NaCl, sand Short and swollen Drought Soil moisture 5.5-3.5% (-0.36 to -3.6 105 Pa); sand Short, swollen, not deformed Salinity 0.2, 0.4, 0.6% NaCl; Jensen tubes Shrinkage of hairs Salinity 1% (0.15 M) NaCl; silica sand
Infection process Decrease in infection thread numbers Drought Soil moisture 5.5-3.5% (-0.36 to -3.6 105 Pa; sand Osmotic 0.2 M PEG, sand Salinity 0.2, 0.4, 0.6% NaCl; Jensen tubes 0.1 M NaCl; sand Multiple branching & distortion Temperature 22, 30oC; agar slopes
a) Conditions with harmful effects are indicated in boldface.
41
in intracellular salt levels, crowding of macromolecules, damage to external layers (pili, membranes), changes in ribosome structure, and decrease in growth. Reactive oxygen species can also damage proteins and DNA, leading to accumulation of mutations (Potts 1994).
Death of rhizobial cells during desiccation was suggested to associate with changes in membrane permeability (Bushby and Marshall 1977). It has been hypothesised that during dehydration, the removal of water hydrogen bond to the phospholipid headgroups of the membrane decreases the spa-cing between adjacent lipids. This leads to increased van der Waals interac-tions between the lipid side chains, resulting in an increase in the membrane phase transition temperature. Then, the membrane is converted from the liquid crystalline into the gel phase already at room temperature. Subsequent rehydration results in a further phase transition of the membrane back to the liquid crystalline phase. As consequence, the membrane barrier is disrupted, leading to leakage of membranes (Potts 1994, Welsh 2000).
Bacteria under osmotic stressSeveral reviews have focused on the responses of bacteria to variations in the solute content of their aqueous environment (e.g. Csonka 1989, Galin-ski 1995, Miller and Wood 1996, Welsh 2000). Hyperosmotic conditions cause rapid effl ux of water and loss of turgor, and the cells may plasmo-lyse. Upon hypo-osmotic shock, water fl ows into the cell and increases
Plantspecies References
Medicago sativa Lakshmi-Kumari et al. 1974Glycine max Tu 1981Glycine max Tu 1981Vicia faba Zahran & Sprent 1986Vicia faba Zahran & Sprent 1987
Trifolium subterranum Worral & Roughley 1976Medicago sativa Lakshmi-Kumari et al. 1974Glycine max Tu 1981
the cytoplasmic volume and/or cell turgor (Csonka 1989, Poolman and Glaasker 1998). Most of the osmotic stress studies have been performed with NaCl that, however, infl uences bacterial cells trough both ionic and osmotic phenomena. In soils, the salinity and moderate drought stress cause osmotic stress. The osmolarity of the rhizosphere, extending a few millimeters from the surface of the plant root or rhizoplane, has been assumed to exceed that of bulk soil water, because the root tissue excludes salts, and both the plant roots and bacteria secrete organic nutrients and mucilage (Miller and Wood 1996).
High growth temperatureAll organisms respond to a sudden increase in growth temperature by induc-ing the synthesis of a number of heat shock proteins (Hsps). Hsps consist of chaperons (such as GroEL, DnaK, DnaJ), small heat shock proteins (sHsps) and proteases. Chaperons are involved in the proper folding of dena-turated proteins, and proteases are required for the degradation of irrevers-ibly damaged proteins. Many chaperons and proteases are also important during normal growth but sHsps presumably function only under heat stress. According to the present model sHsps bind to denaturated proteins accumulated under stress and maintain them in a folding-competent state (Munchbach et al. 1999, Nocker et al. 2001). When bacteria generally have 1-2 Hsps, rhizobia were found to contain multiple Hsps. For example, a Bradyrhizobium strain had even 12 sHsps (Munchbach et al. 1999). Reason for this phenomenon is unclear. It is also not known why a heat-sensitive bean-nodulating Rhizobium strain had several Hsps but a heat-tolerant one contained only two sHsps (Michiels et al. 1994).
1.9.1. Osmoadaptation in rhizobia
In the same way as plants, bacteria exposed to the salts can maintain osmotic equilibrium across the membrane by allowing infl ux of salt or solutes (mainly Halobacteriaceae) or by exclusion of salts via the production of compatible solutes or other organic osmolytes (other bacteria). However, if organic osmolytes are present in surrounding medium, bacteria prefer uptake over synthesis de novo. Exogenous osmolytes that improve cell growth under adverse osmotic conditions are referred to osmoprotectans (Galinski 1995). Apparently, bacteria also adjust to moderate desiccation by synthesising compatible solutes (Potts 1994).
Endogenous osmolytesThe responses of bacteria to growth-inhibiting osmolarity have been studied extensively in the family Enterobacteriaceae and in the case of Rhizo-biaceae, in S. meliloti. Under elevated salinity S. meliloti can accumulate K+
43
-ions and an organic anion, glutamate, and synthetise the following compat-ible solutes: N-acetylglutaminylglutamine amide (NAGGN), trehalose, and glycine betaine, if the medium contains its precursor, choline (Table 1-10). Recent data indicate that the set of endogenous osmolytes produced by rhizobia can vary at least at the species level (Table 1-10). For example, NAGGN has been found only in S. meliloti (Table 1-10). The types of the accumulating osmolytes also depend on the stress level and on the growth phase of the cell culture. In S. meliloti, glutamate accumulated at low salt concentrations, but at higher levels, glutamate and NAGGN were observed. All three osmolytes, glutamate, NAGGN, and trehalose accumulated only at extremely high NaCl concentrations (Smith et al. 1994). When glycine betaine was exogenously supplied in the growth medium, it accumulated as the major osmolyte during the lag and early exponential phases, whereas glutamate and NAGGN prevailed at the late exponential phase (Talibart et al. 1997). Trehalose was the major osmolyte of the stationary phase (Smith et al. 1994, Talibart et al. 1997). In conclusion, osmoregulation is very complex. A number of different environmental factors, such as the level of osmotic stress, growth phase of the culture, carbon source and osmolytes of the growth
Table 1-10. Accumulation of ions, endogenous osmolytes and compatible solutes in Rhizobiaceae under osmotic stress (mostly in NaCl).
Compound Bacterial species References
Iones K+ B. japonicum, Rhizobium sp. Miller and Wood 1996 S. fredii, S. melilotiEndogenous osmolytes glutamate Rhizobium sp. (Prosopis) Hua et al. 1982 S. meliloti Botsford & Lewis 1990, Smith et al. 1994, Talibart et al. 1994, 1997 A. tumefaciens, S. fredii, Smith et al. 1994 Rhizobium sp. (Acacia)Endogenous compatible solutes glycine betaine - by direct uptake A. tumefaciens Smith et al. 1990 S. meliloti Talibart et al. 1997 - via synthesis from S. meliloti Smith et al 1988 its precursor choline mannosucrose A. tumefaciens Smith et al. 1990 N-acetylglutaminylglutamine S. meliloti Smith and Smith 1989 amide (NAGGN) Smith et al. 1994 Talibart et al. 1994, 1997 trehalose S. meliloti Smith and Smith 1989 Breedweld et al. 1990, 1993 Rhizobium sp. (Acacia) Smith and Smith 1989 R. leguminosarum bv. trifolii Breedweld et al. 1991, 1993 Smith et al. 1994 Talibart et al. 1994 Rhizobium sp. (peanut) Ghittoni and Bueno 1995
44
medium control the combination of naturally occurring endogenous osmo-lytes in rhizobial cells (Smith et al. 1990, 1994).
Exogenous osmolytes (osmoprotectants)As indicated above, bacteria prefer uptake of compatible solutes over synthe-sis de novo. The transport systems for external osmolytes (osmoprotectants) are relatively unspecifi c, and bacteria accept components of plant and animal origin (Galinski 1995). Present data indicate that compounds, which can act as osmoprotectants, vary between different rhizobial species (Table 1-11).
In contrast to Esherichia coli, S. meliloti and many other rhizobial spe-cies can use glycine betaine as a C and/or N source under low osmotic stress but as an osmoprotectant in high osmolarity. Only the slow-growing B. japonicum, which is considered as an osmosensitive species, is incapable to transfer glycine betaine and its precursor choline (Boncompagni et al. 1999). Under low osmotic stress glycine betaine is degraded via successive demet-hylation to glycine, while at elevated osmolarity the catabolism is blocked and glycine betaine is accumulated in cells (Sauvage et al. 1983, Bernard et al. 1986, Smith et al. 1988, Boncompagni et al. 1999).
Table 1-11. The effect of exogenously supplied osmolytes (osmoprotectants) on the growth of rhizobia under osmotic stress. The tests were mostly performed in NaCl (DMSP = b-dimethylsulfonipropionate).
Compound Growth stimulation No growth stimulation References
Accumulatorscholine S. meliloti, A. tumefaciens, A. rhizogenes, B. japonicum Boncompagni M. huakuii, R. galegae, R. etli, all biovars of et al. 1999 R. tropici, S. fredii, R. leguminosarum
glycine betaine S. meliloti, A. tumefaciens, A. rhizogenes, B. japonicum Sauvage et al. 1983, M. huakuii, Rhizobium sp. R. etli, Rhizobium sp. Bernard et al. 1986, (Hedysarum), R. galegae (Sesbania), all biovars of Boncompagni R. tropici, S. fredii, M loti R. leguminosarum et al. 1999DMSP S. meliloti Pichereau et al. 1998
DL-pipecolic acid S. meliloti Gouffi et al. 2000
Chemical mediatorsectoine S. meliloti, B. japonicum, Rhizobium sp. (Hedysarum) Talibart et al. 1994 Rhizobium bv. trifolii and bv. viciaeDisaccharides sucrose S. meliloti, B. japonicum, M. huakuii, Gouffi et al. 1998, R. leguminosarum R. leguminosarum bv. viciae, Gouffi et al. 1999 bv. phaseoli and trifolii S. fredii
trehalose B. japonicum Elsheikh and Wood S. meliloti B. japonicum, M. huakuii, 1990 R. leguminosarum R. leguminosarum bv. viciae Gouffi et al. 1999 bv. phaseoli and trifolii S. fredii
45
Intracellular trehalose was also observed to be metabolised by S. meliloti and R. leguminosarum bv. trifolii at low osmolarity (Breedveld et al. 1993). Under osmotic stress, trehalose is not accumulated in the cytosol of rhizo-bia but after degradation during the early exponential phase it is indirectly contributed to enhance levels of two endogenous osmolytes, glutamate and NAGGN (Gouffi et al. 1999). Trehalose is usually broken down into two glu-cose units (Galinski 1995).
It is supposed that glycine betaine is released into ecosystems by pri-mary microbial producers upon dilution stress (rain fall, fl ooding), by decay-ing plant and animals, or by mammals in the form of excretion fl uids (e.g. urine Galinski and Truper 1994, Sleator and Hill 2001). The precursor of glycine betaine, choline, is a common constituent of the eukaryotic mem-branes (phosphatidylcholine) and should therefore be widespread in the soil (Boncompagni et al. 1999). Plant polysaccharides, cellulose and starch, after degradation by bacteria and fungi are presumably natural sources of disac-charide osmoprotectants (Gouffi et al. 1999). Gouffi et al. (1999) suggested that the plant-derived disaccharides are highly osmoprotective for plant-benefi cial bacteria, such as for rhizobia, but are not osmoprotective for other soil bacteria, e.g. for B. subtilis and P. aeruginosa.
Glycine betaine and bacteroidsGlycine betaine may protect rhizobia occupying salt-stressed nodules. Isolated S. meliloti bacteroids were able to transport glycine betaine, and the uptake activity was increased after addition of NaCl (Fougere and Le Rudulier 1990b). The bacteroids also transported choline, which was further converted to glycine betaine when the bacteroids were subjected to salt stress. Generally, bacteroids/nodules degraded glycine betaine at low salinity but the catabolism was blocked by increasing salinity, allowing accumula-tion of glycine betaine to the bacteroids/nodules (Fougere and Le Rudulier 1990a, Pocard et al. 1991). The high level of glycine betaine presumably maintained better water status in the nodule and protected N2 fi xation against salt stress (Pocard et al. 1991). Increased levels of other osmolytes, proline and pinitol (a carbohydrate) have been detected in the cytosol and bacteroids of salt-stressed M. sativa nodules (Fougere et al. 1991) but their function has remained unknown.
1.9.2. Role of trehalose during desiccation
Many bacteria, yeast, invertebrates and some plants are able to survive long periods in a desiccated state, where even more than 99% of the water has been lost. A common feature for all these organisms is the presence of high concentrations (< 20%) of disaccharides, trehalose (microbes, inverte-brates) and/or sucrose (plants) in their dried tissues (Welsh 2000). It has been
46
hypothesised that compatible solutes are important components to maintain viability under moderate water defi cit, whereas the disaccharides, trehalose and sucrose, offer protection under extreme water defi cit (Potts 1994).
Trehalose has been proposed to stabilise dried biological membranes and preserve the integrity of proteins and enzymes (Potts 1994). Obviously, direct hydrogen bonding of trehalose replaces water and maintains the normal packing of the phospholipid headgroups. Thus, the membrane is maintained in the liquid crystalline phase and does not undergo damaging transition to the gel phase or leakage of its contents upon rehydration (Welsh 2000). According to another hypothesis, the property of dried sugar solution to form amorphous glasses is the critical factor. Biological molecules entrapped within glass would be stable at all temperatures below the glass-liquid tran-sition temperature. The chemically inert nature of trehalose may also prevent deleterious chemical reactions between dried molecules and the glass matrix (Welsh 2000). In addition to osmotic stress (as compatible solute) and desic-cation, trehalose also protects bacterial cells from heat and low O2 levels (Potts 1994, Welsh 2000).
1.10. INHIBITION OF N2-FIXING SYMBIOSES BY BIOTIC FACTORS
Not only the abiotic constraints of arid and semiarid regions can inhibit formation of the N2-fi xing symbiosis between rhizobia and leguminous plants but the factors associated with the bacterial population in the soil may cause similar results. The main reasons for the lack of nodules or formation of inef-fective ones on Acacia and Prosopis roots may be: i) the soil is lacking suit-able rhizobia or there are too few of them; ii) indigenous rhizobia induce ineffective nodules; iii) the nodules are induced by Agrobacterium sp.
The total absence of rhizobia is probably very rare in soils because rhizo-bia seem to be capable of persisting in harsh soil conditions even without the host plant (Barnet et al. 1985, Odee et al. 2002). Instead, the lack of com-patible rhizobia for a given host plant is probably a common phenomenon. Rather low numbers of compatible rhizobia are needed for the nodulation, at least in favourable conditions. Inoculation of woody legumes signifi cantly increased shoot N, when the soils contained as few as 50 cells g-1 soil (Turk et al. 1993). However, almost one-half of the tropical soil samples con-tained less than 100 rhizobia g-1 of soil (Singleton and Bohlool 1984). In Kenyan soils collected from different ecological zones, the size of the rhizo-bial population was very varied, ranging from 3.6 to 2.3 x105 cells g-1 soil (Odee et al. 1995). Thus, low amounts of compatible rhizobia present in soil may cause impaired nodulation.
Several experiments with acacias have shown that although nodules were formed on the roots, not all were capable of fi xing N2 (Lawrie 1983, Barnet et al. 1985, Odee et al. 2002). In Australia, only 36% of the strains isolated
47
from Acacia spp. signifi cantly increased plant growth (Barnet and Catt 1991). Also a later study with rhizobia isolated from Australian acacias showed a very wide variation in effectiveness, some host-rhizobia combinations being close to parasitic (Burdon et al. 1999).
Bacteria belonging to the genus Agrobacterium are closely related to rhi-zobia but they induce tumours on plant roots or shoots. Agrobacteria are common soil and rhizophere organisms. Strains resembling or belonging to Agrobacterium sp. have often been identifi ed from the root nodules of Acacia and Prosopis trees grown in arid and semiarid Africa (Khbaya et al. 1998, Nick 1998, de Lajudie et al. 1999, Odee et al. 2002). Nevertheless, when reinoculated in laboratory conditions, they did not induce nodules (Nick 1998, Odee et al. 2002). Only after massive inoculation some strains induced small number of ineffective nodules on different Acacia species (de Lajudie et al. 1999). The explanation for the presence of agrobacteria in the nodule is still unknown, but it has been speculated that either agrobacteria enter nodules together with effective rhizobia, or that agrobacteria carrying some symbiotic genes can infect legume roots (de Lajudie et al. 1999). Con-sequently, agrobacteria represent one group of bacteria that can induce non-fi xing nodules on Acacia and Prosopis roots.
48
2. AIMS OF THE STUDY
N2-fi xing, drought tolerant and multipurpose Acacia and Prosopis species are appropriate trees for reforestation of degraded areas in arid and semi-arid regions of the tropics and subtropics. Inoculation of tree seedlings with tested, effective rhizobia already at the nursery stage may be crucial in order to exploit their N2-fi xing capacity after transplantation to the fi eld. The collection of Sudanese tree rhizobia, gathered in the Sudan 1987-1988 with a Finnish Sudanese Forestry Program (Karsisto and Lindström 1992), is well characterised (Zhang et al. 1991, 1992, Zahran et al. 1994). De Lajudie et al. (1998) and Nick et al. (1999b) described three new species among them (S. arboris, S. kostiense and M. plurifarium), and Haukka et al. (1998) studied their genetic diversity and phylogeny. However, very little was known about the symbiosis between the rhizobia and Acacia/Prosopis trees. Consequently, it was not known which factors in arid and semiarid soils on one hand favour and on the other hand impair or prevent the development of effec-tive symbiosis. This knowledge is important when developing effective inoc-ulants. Therefore, this thesis has following objectives (corresponding papers indicated in parentheses):
1. Investigation of symbiotic properties of fi ve putative inoculant strains, belonging to Sinorhizobium sp. The strains were inoculated on nine Acacia and fi ve Prosopis species, characteristic for semiarid and arid regions in Africa and Latin America. The symbiotic properties studied included host specifi city, infection mode, nodule type, and the effective- ness of the symbiotic association (I).
2. Characterisation of effects of thermal, salt and osmotic stresses on the growth (II, III) and activity (IV) of free-living Sinorhizobium cells.
3. Investigation of infection and nodulation of inoculated A. senegal see- dlings exposed to heat (II) and drought stresses (III), and evaluation of the potential factors explaining failures in the nodulation process.
4. Evaluation of the roles of two compatible solutes, glycine betaine and trehalose, for the Acacia senegal–Sinorhizobium symbiosis exposed to drought stress (III).
5. Examination of potential adaptation mechanisms connected to several peculiar responses of S. arboris populations or cells during the stress experiments (II, III, IV).
49
3. MATERIALS AND METHODS
3.1. BIOLOGICAL MATERIAL
Bacterial strains used in this study are listed in Table 3-1. Tree species used are listed in Table 3-2 according to their taxonomic position.
3.2. PARAMETERS AND STRESS CONDITIONS STUDIED
Methods used and parameters examined are described in detail in the origi-nal publications and summarised in Table 3-3. Stress experiments performed are gathered in Table 3-4.
Table 3-1. Bacterial strains and plasmids used in this study.
Geographical Used in Species Host plant origin Reference paper
Sinorhizobium sp. HAMBI 1480 A. senegal Sudan, Kosti Zhang et al. 1991, I Nick et al. 1999abS. arboris HAMBI 1552 P. chilensis Sudan, Kosti “ I, II, III, IVS. kostiense HAMBI 1483 P. chilensis Sudan, Kosti “ HAMBI 1489 A. senegal Sudan, Kosti “ I HAMBI 1493 P. chilensis Sudan, Khartum “ I S. saheli HAMBI 1496 A. senegal Sudan, El Fau “ I, IIIS. terangae HAMBI 1550 A. senegal Sudan, El Obeid “ I HAMBI 1551 A. senegal Sudan, El Fau “ IS. terangae bv. acaciae ORS 1058 A. mollisima Senegal de Lajudie et al. 1994, I Lortet et al. 1996
Modifi ed strains Properties Parental strainS. arboris HAMBI 1552 R smr HAMBI 1552 IV IV HAMBI 2180 containes gusA21 HAMBI 1552 II II, III HAMBI 2190 containes luc HAMBI 1552 IV IV
Plasmids Description pCAM121 Paph-gusA-TrpA ter translational Wilson et al. 1995 II fusion with adjacent unic SpeI site in mini Tn5 Sm/Sp pAM102 lacI, luc,ori R6K, oriT RP4, Ptac, Möller and Jansson, IV NotI, bla, Kmr 1998 pRK2013 repcolE1, Kmr, Nmr Ditta et al. 1980 IV
50
Table 3-2. Generic names, original distribution and geographic origin of the seeds of the tree species used in this work.
Geographic Original origin of geographic Tree species seeds distribution a)
Tribe Acaciae Subgenus Acacia Acacia nilotica (L.) Del. Sudan Africa, Arabian Peninsula, India Acacia oerfota (Forssk.) Schweinf. Sudan Africa (synonym A. nubica Benth.) Acacia seyal Del. Sudan Africa Acacia sieberiana DC. Sudan Africa Acacia tortilis subsp. raddiana Senegal Africa, Near and Middle East (Savi) Brenan. Subgenus Aculeiferum Acacia angustissima (Mill.) Kunze Chile Central America Acacia mellifera (Vahl.) Benth. Sudan Acacia senegal (L.) Willd. Kenya Africa, Middle East, India, Pakistan Senegal Sudan Subgenus Phyllodineae Acacia holosericea G. Don Senegal Northern AustraliaTribe Mimoseae Prosopis africana (Guill. & Perr.) Taub. Senegal Africa Prosopis chilensis (Molina) Stunz Chile South America Sudan Prosopis cineraria (L. Druce) Senegal India, Pakistan, Middle East Prosopis julifl ora (Sweet) DC. Senegal South and Central America, Mexico Prosopis pallida (Willd.) Kunth. Peru South America
Subfamily PAPILIONOIDEAE Sesbania rostrata Bremek & Oberm Senegal a) Allen and Allen 1982, Roshetko 2001.
3.3. STATISTICAL ANALYSIS
In papers I and II, analysis of variance was performed with the Excel program package version 2000 for Windows 98 (Microsoft Corporation) and with the SAS System program package version v.6.12 for Windows (SAS Institute Inc.). Tukey’s test was applied to compare means at P < 0.05 with the SAS system. In paper III, comparisons between treatments were carried out with the program SPSS 10.0 for Windows by using either one-way ANOVA or a non-parametric Kruskal-Wallis test. To compare means at P < 0.05, Tukey’s test was applied after ANOVA and t’-test was applied after the Kruskal-Wallis test. Principal component analysis (PCA) was performed with the Matlab program package version 42.c1 for Windows (The Mathworks, Inc) equipped with the Data analysis Toolbox (ProfMath Oy, Helsinki, Finland; paper I).
51
Table 3-3. Parameters and methods used in this study, and the respective papers where they appear and references.
Used Parameter Method in paper References
Cell morphology Epifl uorescence microscopy IV after staining with acridine orange
Culturable cells Plate counts (cfu) II III IV
Esterase active cells Epifl uorescence microscopy IV Tsuji et al. 1995, IV after staining with SFDA
Glycine betaine HPLC III Rajakylä and Paloposkicontent of the plants 1983
Growth of bacteria Plate counts II III IV Optical cell density (A 600 nm), III IV Culturing on Bioscreen III
Growth of plants Determination of shoot length. I II III number of nodules and dry I II III weights of shoots and roots I II III
Infection process Light microscopy after staining plant roots with methylene blue I III Vasse and Truchet 1984 and GUS substrate X-glc-A II III Wilson et al. 1995
Luciferase activity Light output measured IV Möller and Jansson with the luminometer 1998, IV
Maximum temperature Temperature-gradient II Andersson et al. 1995,for bacterial growth incubator IIand GUS production
Nitrogen fi xation Acetylene reduction method I II III Lindström 1984a, b Räsänen et al. 1991
Nodule structure Semi-thin nodule sections I Sprent et aI 1989, I
Number of culturable CFU count II IIIbacteria in the rhizosphere
Permabilised (dead) Epifl uorescence microscopy IV Roth et al. 1997, IVcells after staining with SYTOX green
Soil moisture Moisture fraction III
Total cell number Epifl uorescence microscopy IV Bloem 1995, IV after staining with acridine orange
Transformation Plate mating and II Wilson 1996of marker genes triparental mating IV Maniatis et al. 1982
52
Table 3-4. Stress experiments performed in this study (GB, glycine betaine; PEG, polyethylene glycol). MarkerStress Reseach subject Conditions gene Paper
Heat stress Growth of bacteria Root temperature of gusA II and plants, 28, 36, 38 , 40, and 42oC symbiotic interactions
Heat stress Growth, activity and Culturing of cells at 28, 37, luc IV physiological status and 40oC of the cell cultures
Drought Growth of plants Well and less watered gusA IIIstress and bacteria, seedlings without or with symbiotic interactions glycine betaine and trehalose Salt and Growth of liquid Culturing of cells in IIIosmotic cell cultures NaCl ( 0.1, 0.5, and 1.0 M)stress PEG 6000 (9, 17, and 24%) with or without osmoprotectants GB (0.01, 0.1, and 1.0 M) trehalose (0.01, 0.1, and 0.9 M)
53
4. RESULTS AND DISCUSSION
4.1. SYMBIOTIC PROPERTIES OF AFRICAN SINORHIZOBIA
4.1.1. Host specifi city
Few studies have so far been reported on the host specifi city of the rhizo-bia nodulating Acacia and Prosopis trees. Extensive cross nodulation tests performed between 15 Australian Acacia species and about 40 rhizobial strains isolated from different Australian acacias indicated that isolates that performed well on one Acacia species often performed well on others as well (Thrall et al. 2000). A similar phenomenon could be detected in this work. The fi ve African Sinorhizobium strains tested, namely S. arboris strain 1552T, S. kostiense strains 1489 T and 1493, S. saheli strain 1496 and S. terangae bv. acaciae ORS 1058 (I: Table 1), were capable of effectively nodulating a wide range of African acacias (A. mellifera, A. nilotica, A. oerfota (synonym A. nubica), A. nilotica, A. senegal, A. seyal, A. sieberi-ana, A. tortilis subsp. raddiana) and several Latin American Prosopis spp. (I: Table 2).
Although a wide range of rhizobia are capable of inducing N2-fi xing nodules on Acacia and Prosopis trees, there seems to be a certain degree of host specifi city between rhizobia and tree species (Khbaya et al. 1998, Zerhari et al. 2000). Thrall et al. (2000) found that in Australian acacias, species with more limited distribution or tighter ecological requirements have a greater degree of specifi city than widespread species. A similar spe-cifi city seems to occur within rhizobial species. In African S. terangae, two biovars, sesbaniae and acaciae have been distinguished according to which plants the strains induce effective nodules on (Lortet et al. 1996). Haukka et al. (1998) suggested that an equal division could be done in S. saheli, because S. saheli strains isolated from Sesbania spp. induced effec-tive nodules on several Sesbania spp. but ineffective ones on Acacia spp. (de Lajudie et al. 1994), whereas the S. saheli strain isolated from Acacia sp. and used in this thesis work behaved in an opposite way (I: Tables 1, 2). Interestingly, the nodA sequences were clearly different depending, on whether S. saheli strains were isolated from Acacia spp. or Sesbania canna-bina (Haukka et al. 1998). Unlike acacias, trees and bushes of the papili-onoid genus Sesbania grow in moist and waterlogged soils (Allen and Allen 1981, Roshetko 2001). Thus, taxonomically similar Sinorhizobium species appear to have adapted to different environments and adjusted to nodulate different host plants, probably, through lateral transfer of symbiotic genes (Haukka et al. 1998).
The fact that how Acacia and Prosopis trees belonging to different genera share same rhizobia might be explained by these trees being closely
54
related (Chappill 1995, Bukhari 1997, Bukhari et al. 1998). Comparison of symbiotic genes (nodA, nifH) revealed that Latin American sinorhizobia, which were isolated from Prosopis nodules, differed from African sinorhi-zobia (Haukka et al. 1998). Thus, it would be interesting to know if Latin American sinorhizobia are also capable of effectively nodulating African acacias.
It was surprising that the African sinorhizobia induced only ineffective nodules on African P. africana although they were capable of forming N2-fi xing nodules on Latin American Prosopis spp. (P. chilensis, P. julifl ora and P. pallida) and on Afro-Asian P. cineraria (I: Table 2). P. africana may grow in slight moister environments, with annual rainfall over 600 mm (Keay 1989, Bellefontaine et al. 2000), but otherwise it should share simi-lar growth sites with several Acacia species studied here. P. africana was found to be ancestral to other Prosopis trees (Bukhari et al. 1998) forming a separate series (Burkart 1976). The growth habit of P. africana seedlings differed from that of other Prosopis species; P. africana had a sturdy but smooth stem whereas the stem of other Prosopis (and Acacia) seedlings became woody early (I). Perhaps P. africana has diverged from other Pro-sopis species a so long time ago that they do not share similar rhizobia any more.
As discussed in the section 1.5.7, African and American acacias belon-ging to the subgenera Acacia and Aculeiferum are more closely related than they are to the Australian acacias belonging to the subgenus Phyllo-dineae (Robinson and Harris 2000). This might explain why the African sinorhizobia induced only small, ineffective nodules on A. holosericea (I: Table 2). Probably A. holosericea, like other Australian acacias, prefers slow-growing bradyrhizobia (Dreyfus and Dommergues 1981, Marsudi et al. 1999, Lafay and Burdon 2001). Interestingly, the nodulation pattern of the broad host range Rhizobium sp. strain NGR 234, originating in Papua New Guinea, distinguished Australian acacias from other Acacia species. The strain effectively nodulated Australian acacias but not African and South American ones, and neither Prosopis spp. (Pueppke and Broughton 1999).
4.1.2. Variation in the effectiveness of the symbiotic association
Except for A. holosericea and P. africana, inoculation of Acacia and Prosopis with effective Sinorhizobium strains improved plant yield compared to that of uninoculated seedlings (I: Fig. 5). However, no strain appeared to be supe-rior to the others (I: Fig. 6B), and there were no differences between any rhizobia-Acacia or rhizobia-Prosopis combinations (I: Fig. 5).
The effectiveness of symbiotic associations (determined mostly as growth performance) between native rhizobia and acacias can vary a lot (Barnet
55
et al. 1985, Barnet and Catt 1991, Burdon et al. 1999, Thrall et al. 2000). Parallel results were obtained in this work when A. senegal seedlings were inoculated with seven Sudanese strains. Based on the ability to improve plant yield, the strains could be classifi ed as effective, poorly effective and ineffec-tive (I: Fig. 7).
It is a generally accepted theory that rhizobia have adjusted to nodulate various legumes through lateral transfer of symbiotic genes (Haukka et al. 1998). Consenquently, rhizobia may also lose symbiotic genes, especially, if they are located on symbiotic plasmids, which are easily movable genomic elements. Indeed, many rhizobial strains have been observed to lose their N2 fi xation or nodulation ability during storage (Sutherland et al. 2000, Odee et al. 2002a). In our laboratory, Sudanese sinorhizobia were observed to carry several large plasmids, one of those containing symbiotic genes (Haukka et al. 1998). According to earlier studies, Sinorhizobium sp. strain 1480 formed effective nodules on A. senegal and P. chilensis plants (Zhang et al. 1991). Later, no plasmids were detected (Haukka et al. 1998), and in this work strain 1480 induced small, white and ineffective nodules on A. senegal (I: Table 1; Fig. 7). In laboratory conditions, the heat treatment can trigger elimination of plasmids from rhizobial cells (Baldani and Weaver 1992), but it is still unk-nown to which extent loss of symbiotic plasmids occurs among indigenous or inoculant rhizobia in the soil.
Both herbaceous and woody legumes have exhibited genetic variability in the properties related to growth and N2 fi xation (Sanginga et al. 1990, Vance 1998). In this work, the Sudanese provenance of A. senegal showed better growth and symbiotic performance than Kenyan and Senegalese A. senegal (I: Fig. 6B). Thus, by taking into consideration symbiotic and N2-fi xing properties when breeding or selecting woody legumes, it would be possible to enhance biological N2 fi xation.
4.2. Development of the symbiosis between sinorhizobia and Acacia and Prosopis trees
As discussed in Section 1.7.2., depending on the legume species, rhizobia can enter plant roots either via wounds (crack entry), intact epidermis, or by the classic root hair pathway. Both the crack infection and root hair infec-tion may give rise to either determinate or indeterminate nodules (Table 1-4). Except for the extensive work of Corby (Corby 1988), who classifi ed legumes according to the nodule morphology, there are only few reports in which infection or nodulation on Acacia/Prosopis trees have been examined (P. glandulosa; Baird et al. 1985). This work showed that, except in A. seyal, the infection mode between sinorhizobia and Acacia or Prosopis species studied resembled that of classical root hair infection, and that the nodules formed shared features with indeterminate nodules (I).
56
4.2.1. Infection through the root hairs
The infection process between S. arboris and Acacia/Prosopis seedlings had several common features with the classical root hair infection: presence of root hairs, deformation of hairs, and formation of infection threads (I: Table 2, Fig. 1 and 2). Infection threads were also found in P. glandulosa hairs (Baird et al. 1985).
As in herbaceous legumes (Dart 1977, Vandenbosch et al. 1985, Räsänen et al. 1991), inoculated Acacia/Prosopis hairs were shorter and broader than uninoculated ones (I: Fig. Ia, b), markedly curled hairs occurred in patches (I) and infection occurred in localised groups (L. Räsänen, unpublished). The latter phenomenon was also observed on P. glandulosa (Baird et al. 1985). P. chilensis, P. julifl ora and P. pallida had long deformed hairs with knob or shepherd’s crook-like structures (I), similar to the characteristics of alfalfa hairs (Räsänen et al. 1991).
There were some features which were characteristic of the early infec-tion process of Acacia/Prosopis trees. In herbaceous papilionoid legumes, the hairs, which mainly are long, form a continuous cover in the elonga-tion region of the roots, and hairs of different sizes are deformed (e.g. Räsä-nen et al. 1991). In Acacia spp., the short hairs grew on lateral roots in patches, and there were many bare roots. Prosopis species had a denser cover of longer hairs on the lateral roots (I). Deformed hairs of Acacia and P. chilensis were usually dwarfed, swollen and bent against the root surface (I: Fig.1b-d). In longer hairs of all species, infection pockets, i.e. intercellular spaces for rhizobia, were detected near the site of rhizobial penetration (I: Fig. 1f). Later, when bacteria had proliferated, the pockets were enlarged, appearing as sac-like structures, from which the infection threads originated (I: Fig. 2a-c, f). Studies with the papilionoid S. rostrata indicated that the formation of infection pockets, which is a prerequisite for the nodule initiation, was accompanied by local plant cell death. Pro-bably, Nod factors induced an oxidative burst generating reactive oxygen species and local ethylene production, subsequently causing death of plant cells (D’Haeze et al. 2002).
In the case of A. seyal only single hairs developed at lateral root junc-tions and no proper infection threads were found on them (I). Thus, it is possible that this species is infected for example by crack entry (Fig. 1-2).
In optimal laboratory conditions the infection process between sinorhi-zobia and Acacia/Prosopis seedlings proceeded almost as fast as in herba-ceous legumes. Although the lateral roots were still short or developing and hairs were uncommon four days after inoculation, deformed hairs were common three days later. The fi rst infection threads appeared seven days and fi rst nodules 11 days after inoculation, respectively.
57
4.2.2. The structure of effective indeterminate nodules
In Acacia and Prosopis, the nodules developed singly or grouped mostly on lateral roots and occasionally, on taproots, or on short and thin rootlets, which arose from the taproot. Young nodules were spherical, but later they became elongated and often branched, being able to grow from several points (I: Figures 3C, 4C). The nodules showed typical indeterminate arrange-ments of tissues: an outer cortex, nodule endodermis, several vascular bundles, and a central tissue composed of bacteroid containing cells and uninfected cells (I: Figures 3, 4A, B). The cortex layers of A. nilotica and A. sieberiana nodules were especially thick (I: Figures 3A, B). Except P. afri-cana nodules, those of other Prosopis species, A. nilotica and A. sieberiana contained dark stained depositions, presumably tannins in the nodule cortex (I: Fig 3C). Both the rigid cell layers and tannin inclusions have been reported to be features of perennial nodules (Fred et al. 1939, Allen and Allen 1981).
Though Acacia and Prosopis nodules were basically indeterminate, and in common with the descriptions of Baird et al. (1985) and Corby (1988), the following evidence suggests that the nodules diverge to some degree from the classical indeterminate nodule (Vasse et al. 1990). P. glandulosa, a common tree in warm deserts of North America, usually forms elongate, indeterminate nodules. The morphology of nodules showed, however, high variability when the plants were inoculated with cowpea (Vigna unguiculata) rhizobia (Virgi-nia et al. 1984, Baird et al. 1985) or with soil collected from various depths under Prosopis canopies (Johnson and Mayeux 1990). P. glandulosa seed-lings produced two types of N2-fi xing nodules: typical indeterminate nodu-les and spherical ones that resembled the determinate type lacking a clear meristemic/invasion zone (Virginia et al. 1984). It was assumed that the varia-bility in nodule morphology represented different stages of maturity (John-son and Mayeux 1990). Also in my thesis work, the apical meristem was not always easily detectable (I: Fig. 3A). Like in Australian A. saligna (Marsudi et al. 1999), a low proportion of bacteroid fi lled cells were characteristic for the central tissue of Acacia and Prosopis nodules studied here (Fig. 3B, 3C).
4.2.3. Ineffective nodules of A. holosericea and P. africana
The internal structure of ineffective nodules may vary a lot, being like effective ones or resembling more a tumour than a nodule. Obviously, the more the ineffective strain is related to those rhizobia that induce N2-fi xing nodules on the given plant, the more nodules formed resemble characteristic ones. In P. glandulosa, ineffective nodules contained also bacteroids (Virginia et al. 1984, Baird et al. 1985). In this thesis work, the fi ve African sinorhizobial strains induced hair deformation and ineffective nodules on the roots of Australian A. holosericea and African P. africana (I: Table 2). This indicated that some recog-
58
nition occurred between bacteria and plants. Nevertheless, the strains tested only distantly resembled compatible rhizobia of A. holosericea or P. africana, because tumour-like structures with undifferentiated cell tissue formed on A. holosericea roots (I: Fig. 4D) and nodule-like structures, having some organisa-tion of cell layers, developed on P. africana roots (I: Fig. 4E).
Acacia and Prosopis trees seem on one hand to be capable of forming N2-fi xing nodules with a wide range of rhizobia and on the other hand they are not able to exclude nodulation by incompatible rhizobia. Vegetative rhi-zobia, being capable of saprophytic growth in soil and nodule debris, has been found both from effective and ineffective nodules (Paau et al. 1980, Vance et al. 1980, Timmers et al. 2000). Sprent suggested (2001) that at the community level both legumes and rhizobia gain advantage from promis-cuity. Ineffective (as well as effective) Acacia nodules may be a way of main-taining rhizobial populations under stress conditions. Rhizobia ineffective with Acacia sp. may be effective on other legumes growing in the same area. Subsequently, in the wet season for example annual herbs benefi t from rhi-zobia released from the ineffective nodules.
4.3. TOLERANCE OF SINORHIZOBIA TO ABIOTIC STRESSES
4.3.1. Properties of two potential inoculant strains
The two major strains, S. arboris 1552 and S. saheli 1496, used in this work differed positively from other strains due to their competent symbiotic proper-ties already during the fi rst nursery tests performed in the Sudan (Karsisto and Lindström 1992). Later studies together with this work indicated that they also possess other favourable properties (Table 4-1). Strain 1552 can be considered as salt and osmotolerant, whereas strain 1496 is sensitive to salt but tolerated higher growth temperatures (Table 4-1). According to the nodulation tests, both strains were competitive against an ineffective strain, but strain 1496 was a better competitor when matched with strain 1552 (Zhang et al. 1992).
Trehalose appeared to be a good carbon source for the strains 1552 and 1496 (III). When the growth of the strains was tested on minimal medium with six sugars (e.g. glucose, mannitol, trehalose), trehalose also caused the most rapid growth (L. Räsänen, unpublished).
4.3.2. Tolerance of heat
Tropical rhizobia isolated either from herbaceous or woody legumes can grow at around 40˚C (e.g. Hafeez et al. 1991, Baldani and Weaver 1992, Surange et al. 1997, Hashem et al. 1998, Zerhari et al. 2000). Similar obser-vations were done with the Sudanese tree rhizobia (Zhang et al. 1991, Zahran et al. 1994). Extremely thermotolerant Pakistan and Indian strains,
59
such as a strain isolated from the tree Albizia lebbek, could grow at around 50°C (Hafeez et al. 1991, Surange et al. 1997). With reference to data above, S. arboris strain 1552 and S. saheli strain 1496 are well adapted to tropical soils, with Tmax around 41°C and 41.5-44°C, respectively (Table 4-1). However, the study with the heat-tolerant rhizobial strain indicated that it did not possess better abilities to fi x N2 at high temperature. Probably, the advantage of the thermotolerance is connected to the fact that the strain has capacity to survive the periods of thermal stresses and to recover afterwards (Michiels et al. 1994).
When strain 1552 was grown as liquid cultures at 40ºC, growth was instable (IV: Fig. 1A), indicating that it suffered from thermal stress. Interes-tingly, strain 1552 probably experienced some stress already at 37ºC because the luc-marked derivative of strain 1552 showed large variation in optical density, i.e. EPS production. In addition, luciferase activity decreased rapidly at the end of the incubation, revealing that energy reserves were exhausted (IV: Tables 1 and 2).
4.3.3. Tolerance of salt
Arid and semiarid soils often suffer from elevated salinity due to high evapo-ration. Thus, it was not surprising that like S. arboris strain 1552 (Table 4-1), many rhizobia isolated from Acacia and Prosopis trees were moderately salt tolerant, capable of growing in 0.3-0.5 M (2-3%) NaCl (Craig et al. 1991, Zhang et al. 1991, Zahran et al. 1994). Several very salt tolerant strains (0.6-0.86 M or 3.5-5% NaCl) were isolated from the nodules of Indian, Pakistani and Moroccan acacias acacias (Surange et al. 1997, Kumar et al.
Table 4-1. Properties of two major strains, Sinorhizobium arboris HAMBI 1552
and S. saheli HAMBI 1496, used in this work.
Strain Parameter 1496 1552 References
Tmax; YEM agar 44.2oC 41.5oC Zahran et al. 1994Tmax; TY agar 41.2-41.7oC 40.8oC L. Räsänen, II
NaCl tolerance in YEM broth 1.5% (0.26 M) 3% (0.5 M) Zhang et al. 1992Growth in 0.5 M NaCl; YEM agar - ++ Zahran et al. 1994Growth in 0.1 M NaCl; BD broth +++ +++ IIIGrowth in 0.5 M NaCl; BD broth + ++ III
Tolerance of sucrose; YEM agar 0.34 M 1.0 M Zahran et al. 1994
Growth in 9% PEG 6000 ++ ++ IIIGrowth in 24% PEG 6000 not determined + III
Colony morphology small, compact large, mucuous II, III, IV
- no growth; + weak growth; ++ moderate growth; +++ good growth
60
1999) Zerhari et al. 2000). About a third of the strains isolated from different legumes growing in salt-affected soils in Morocco were very (0.86 M NaCl) or extremely salt tolerant (1.7-2.4 M, 10-14% NaCl) (Abdelmoumen et al. 1999). The growth of the salt sensitive S. saheli 1496 strain was stimulated by 0.1 M NaCl (III: Fig. 2). Perhaps this phenomenon was due to that NaCl infl uences trough ionic and osmotic factors and/or rhizobia occupying arid and semiarid soils are generally adapted to grow with some salts.
There is a couple of studies, in which the salt sensitive strain was more affected than the salt tolerant one or failed in nodulation of salt tolerant Acacia seedlings grown under salt stress (Zou et al. 1995, Kumar et al. 1999). However, the salt tolerance in the laboratory does not necessary indicate salt tolerance in the fi eld.
Subsequently, it is not known whether the salt tolerance is a favourable pro-perty only for free-living bacteria occupying saline soils or does it play some role also during the infection process and symbiosis when the host legume grows in saline soil. In the symbiotic state rhizobia live in a cell where the solute concentration is high, approximately 0.3-0.4 M (Sprent 1984). When free-living rhizobia differentiate into bacteroids, LPSs located in the outer sur-face undergo structural modifi cations that are somehow involved in the chan-ges in environmental conditions (Kannenberg et al. 1998). Possibly, bacteroids in any case adapt to high osmolarity through LPS modifi cations, and therefore the salt tolerance of a strain may not be relevant during the symbiosis.
The model organisms Escherichia coli, Salmonella typhimurium and S. meliloti reacted to salt stress mainly by increasing levels of K+ and glutamate (Galinski 1995, Miller and Wood 1996). Interestingly, rhizobial strains that were isolated from the nodules of Acacia sp. growing in salt-affected soils behaved like halobacteria, by accumulating NaCl but not KCl when grown in NaCl. The relative proportion of glutamate also remained fairly constant irrespective of the salt treatment (Craig et al. 1991). Thus, rhizobia may also have other kinds of salt tolerance mechanisms in addition to the presently known ones.
Salt tolerance of rhizobia may correlate with drought tolerance (Athar 1998). This information can help when looking for drought tolerant strains. Recently, eight gene loci were found to be involved in salt tolerance of R. tropici. Most of these genes were also required for adaptation to hyperosmo-tic media, and all genes were important for the bacteroid development or function (Nogales et al. 2002).
4.3.4. Roles of glycine betaine and trehalose for sinorhizobia
Compatible solutes as a carbon source. Contrary to E. coli and S. typhimu-rium, rhizobia have a capacity to utilise glycine betaine as a carbon and/or nitrogen source at low osmolarity (Miller and Wood 1996). S. arboris strain
61
1552 and S. kostiense strain 1496 could use 0.01 and 0.05 M glycine betaine as a sole carbon source but the growth was delayed compared to that with 0.006 M glucose or 0.01 M trehalose (III: Fig. 1A, 2A. This suggested that glycine betaine is not a primary carbon source and/ or it is degraded through a secondary pathway.
Compatible solutes as osmoprotectants. Although the potential to tran-sport glycine betaine is widespread among rhizobia (Sauvage et al. 1983, Bernard et al. 1986, Smith et al. 1988, Boncompagni et al. 1999), not all rhi-zobia are able to use glycine betaine as an osmoprotectant (Boncompagni et al. 1999). In the case of the tropical Sinorhizobium strains used in this work, 0.01, 0.05 and 0.1M glycine betaine and trehalose functioned as osmopro-tectants under osmotic stress induced by 9 and 17% PEG but only trehalose improved cell growth under salt stress (0.5 M NaCl; III: Fig. 1, 2) According to Gouffi et al. (1999) exogenous trehalose serves under osmotic stress both as a carbon source and as an osmoprotectant. This theory may explain why the stressed Sinorhizobium cultures produced better a growth response with trehalose than with glycine betaine (III: Fig. 1, 2).
4.4. THE A. SENEGAL-S. ARBORIS SYMBIOSIS UNDER HEAT AND DROUGHT STRESS
4.4.1. Rhizobial population in the soil
Dry and hot seasons restrict N2 fi xation and nodulation of leguminous trees and bushes (Barnet et al. 1985, Hansen and Pate 1987). In arid and semiarid regions of the tropics, the soil temperatures near the surface can be very high. In Australian sandy soils, the temperature near the soil surface was 59°C at the air temperature 39°C. However, the soil temperature decreased rapidly with depth, being moderate 35°C, at 15 cm (Fig. 4-1).
There is no information available, how rhizobia tolerate high temperatu-res in natural desert soils. It appears, however, that rhizobia are more resis-tant to high temperatures in soil than in laboratory medium (AbdelGadir
Fig. 4-1. Soil temperatures in bare sandy soils of Western Australia according to Chatel and Parker (1973). The maximum air tempera-ture was 39°C. At 5.1 cm the temper-ature was above 40°C for seven hours and at 10.2 cm above 35°C for eight hours.
62
and Alexander 1997). A similar phenomenon was observed in this work. Although the Tmax for S. arboris strain 1552 was around 41°C in laboratory conditions (Table 1-4), the numbers of culturable cells in the A. senegal rhi-zosphere were similar (107 cfu g-1) irrespective of the daily maximum soil temperature (30°C or 42°C; II).
The study performed with Indian desert soils suggested that not the high soil temperature but the low organic matter and poor soil moisture were the major factors that reduced the numbers of different micro-organisms (Rao and Venkateswarlu 1983). Indeed, in drought-affected A. senegal soils the numbers of culturable rhizobia were signifi cantly reduced, approximately from 107 to 106 cfu g-1 (III: Table 1). In conclusion, one can assume that during the dry season, the water defi cit together with the high soil temperature will considerably decrease rhizobial numbers or cause a lack of rhizobia in the surface soils (0-10 cm).
4.4.2. Adaptation of A. senegal seedlings to heat and drought
Adult A. senegal trees tolerate high daily temperatures, the mean maximum temperatures being around 48ºC (Table 1-3). A. senegal seedlings also appeared to be resistant to high temperatures, by surviving several weeks at the maximum root temperature of 42ºC (II). Nevertheless, seedlings responded to the thermal stress by stopping shoot growth and nodulation (II: Fig. 2, 3). Unlike uninoculated, non-stressed plants, thermally stressed seedlings lacking nodules did not suffer from a shortage of N, although they were grown in nitrogen-free media (II). Due to the big size, A. senegal seeds are capable of germinating under drought stress. Their seeds are also a large source of energy and nutrients (Khurana and Singh 2001). These properties might explain how thermally stressed A. senegal seedlings with limited nutrient sources could survive a prolonged period on the seed reserve (II).
Adult A. senegal trees are very drought resistant, having several dehyd-ration postponement strategies (e.g. deep rooting system, reduction in leaf area, drought-deciduous growth habit; section 1.8.4). Formation of gum arabic is also supposed to be a natural response of A. senegal to store strongly hydrophilic carbohydrates under drought stress (Roshetko 2001). A. nilotica and A. tortilis seedlings were able to increase the root:shoot ratio under drought stress, resulting in enlargement of the absorptive root biomass (Michelsen and Rosendahl 1990, Otieno et al. 2001). When A. senegal were grown under favourable watering conditions, root hairs usually occurred in patches and there were many bare roots (I). However, under water defi cit both inoculated and uninoculated seedlings developed more hairs on their roots than unstressed ones (III), obviously, in order to enhance effi cient water uptake.
63
4.4.3. Infection and nodulation under stress
Use of the GUS-marked S. arboris strain 2180, which expressed ß-glucuroni-dase (GUS) both in free-living cells and during the early stages of symbiosis, turned out to be a sensitive and rapid tool for following and comparison of infection and nodulation between stressed and non-stressed roots. Contrary to the traditional stain methylene blue, only areas colonised or occupied by GUS-marked strain were stained blue. This was important in the case of A. senegal roots, because the hairs were short and unevenly distributed, and infection threads occurred in very short hairs.
The unfavourably high root temperature and drought stress caused A. senegal hairs to be partly abnormally deformed. Short and swollen bottle-like hairs were typical under heat stress (II: Fig. 4b-d), and very short or dwarfed, swollen hairs were characteristic of water-stressed hairs (III: Fig. 8d). Under both stresses, infection threads were rare, and they seemed to be either disintegrated (II: Fig. 4e) or they had weak glucuronidase activity (III Fig. 8e). Similar observations were made in Trifolium subterraneum nodules, where the high root temperature disrupted growth of infection threads and disturbed transformation of bacteria into bacteroids. Thermal stress may also accelerate nodule senescence (Pankhurst and Gibson 1973).
In heat stress experiments, nodulation of A. senegal seedlings was impai-red at 38°C, reduced to 50% at 40°C and completely inhibited at 42°C (II: Fig. 2).
The water stress retarded or stopped the normal nodule development so that the numbers of nodule initials were higher on water-stressed plants than on well-watered plants. For the numbers of true nodules the situation of was reversed (III: Fig. 7). The nodules formed showed signs of premature senes-cence. After incubation with GUS substrate buffer, nodules displayed weak glucuronidase activity (III: Fig. 8e), indicating low cell numbers and/or low activity of rhizobia occupying nodules.
In T. subterraneum, water defi cit was also observed to prevent the deve-lopment of nodules (Worral and Roughley 1976).
In many tropical crop legumes, a root temperature of around 40°C (5-8 h per day) inhibited development of new nodules (Arayangkoon et al. 1990, Kishinevsky et al. 1992, Hungria and Franco 1993). Of the tropical woody legumes, Sesbania sesban still nodulated at 40°C but Calliandra calothyrsus did not (Purwantari et al. 1995). In the case of A. senegal seedlings, nodula-tion was completely inhibited at 42°C (II). However, the inhibition of nodu-lation was reversible. After A. senegal seedlings were transferred from 42°C to 30°C, N2-fi xing nodules developed on those parts of lateral roots that were developed after completion of thermal stress (II).
When Acacia and Prosopis seedlings were grown in optimal growth chamber or greenhouse conditions, big nitrogen-fi xing nodules were gene-
64
rally formed just below the soil surface (0-5 cm), on the hairs located at the root base (L. Räsänen, unpublished results). Data of Chatel and Parker (1973) suggest that the temperature in the surface soil (0-10 cm) can be so high, ran-ging between 60 and 39°C (Fig. 4-1) that the nodulation of the upper roots of the tree seedlings could be prevented in the fi eld.
In spite of the seedlings being exposed to the high root temperature of 42°C or severe drought stress, it appeared that the stressed A. senegal seedlings had the most important elements that were needed for proper infection. The plants had lateral roots and root hairs (II: Fig. 4; III: Fig. 8), although their development was delayed at high root temperature (II). The hairs were colonised by rhizobia, and they were partly deformed (II: Fig. 4b; III), indicating that S. arboris cells were capable of producing Nod factors. Both the thermally stressed and water-stressed soils contained suffi cient numbers of rhizobia for nodulation, 105 -107 cfu g-1 (II, III). Also according to Williams and de Mallorca (1984), the reduc-tion in nodule numbers on soybean roots under water stress was not associated with a reduced number of rhizobia in the rhizosphere, nor due to an effect on root growth or hair formation. What could then explain why nodulation was inhibited? Generally, plants are more sensitive to environmental stresses than rhizobia, and it is usually the plant that restricts nodulation. The plant hormone abscisic acid (ABA) is one of the major signals operating during drought and saline stress (Bray 1997). Foliar application of ABA to unstressed soybean roots was found to inhibit nodulation. In addition, under leaf moisture stress the endogenous content of this plant hormone was increased in the roots. Thus, ABA may contribute to the extensive inhibition of nodulation observed under water stress (Williams and Sicardi de Mallorca 1984).
4.5. ROLE OF COMPATIBLE SOLUTES ON THE A. SENEGAL-SINORHIZOBIUM SYMBIOSIS
4.5.1. Endogenous glycine betaine in A. senegal
Glycine betaine is a compatible solute, which enables some fl owering plants to grow in salt-affected soils. It appears that the capacity to synthesise glycine betaine is widespread, being characteristic for angiosperms but expressed strongly by some and weakly by others (Rhodes and Hanson 1993). Accord-ing to Rhodes and Hanson (1993), under non-stressed condition glycine betaine levels are for accumulating and nonaccumulated species 5-100 µmol g-1 dry wt and <1 µmol g-1 dry wt, respectively. Under natural or experimental saline or dry conditions accumulators contain glycine betaine 40-400 µmol g-1 dry wt. According to this defi nition, acacias do not accu-mulate glycine betaine because the glycine betaine content was less than 1 µmol g-1 dry wt for non-stressed plants (Rhodes and Hanson 1993; III) and 8.5 µmol g-1 for drought-stressed Australian acacias (Erskine et al. 1996).
65
Proline and polyols (pinitol, mucoinositol) are the most common compatible solutes detected in acacias so far (Erskine et al. 1996).
Many crop plants are not able to synthesise glycine betaine but they can translocate foliar-applied glycine betaine from leaves to other organs. In addition, plants usually are incapable of degrading glycine betaine (Mäkelä et al. 1996, McNeil et al. 1999). Therefore, foliar application of glycine betaine, which is a major by-product when sugar is manufactured from sugar beet (Beta vulgaris), has been used to improve drought tolerance and sub-sequently, to increase crop production of drought-sensitive plants (Agboma et al. 1997, Diaz-Zorita et al. 2001). Addition of glycine betaine to the soil has functioned similarly to the foliar application (Blunden et al. 1997, Xing and Rajashekar 1999). The suitability of foliar application was also tested on Sudanese A. senegal seedlings (III). Then, the foliage could absorb and translocate 0.01 M glycine betaine into roots but the leaves were sensitive to 0.01 M glycine betaine, tolerating concentrations of 0.001 M.
4.5.2. Effects of exogenous compatible solutes on A. senegal seedlings
Moisture stress is considered as the major factor that causes mortality among tree seedlings during the dry period (Bukhari 1998, Khurana and Singh 2001). Seedlings that develop some physiological and/or morphological adaptive characteristics to water stress can account for greater success in restoration efforts (Khurana and Singh 2001). To understand mechanisms that presumably function when A. senegal seedlings adapt to the drought stress, roles of two compatible solutes, glycine betaine and trehalose, were evaluated (III).
0.0003 M glycine betaine in the soil mix helped A. senegal seedlings, so that 60% of the plants were still alive 42 days after inoculation whereas the drought-stressed plants grown without any increments were wilted 30 days after inoculation (III: Fig. 6). The daily transpiration rates of the glycine betaine-treated soybeans were relatively lower than those of plants not trea-ted (Agboma et al. 1997). Application of glycine betaine allowed also beans (Phaseolus vulgaris) to maintain a better water status during water stress so that they developed wilting symptoms much later compared to stressed plants without glycine betaine (Xing and Rajashekar 1999). In this work, drought-affected soils containing glycine betaine tended to keep more water than non-treated soils (III: Table 1). Taken together these data suggest that application of glycine betaine enabled A. senegal seedlings to control water uptake and transpiration more effi ciently.
0.0003 M trehalose slightly increased the numbers of A. senegal plants survived (III: Fig. 6). However, plants do not accumulate trehalose but an another disaccharide, sucrose (Goddijn and van Dun 1999). Thus, whether the slightly improved survival of the seedlings was for example, due to the
66
capability of trehalose to preserve biological structures against desiccation (Potts 1994, Goddijn and van Dun 1999, Welsh 2000), or based on its ability to absorb water, remained unknown.
The advantageous effect of compatible solutes may be also due to the enhanced EPS production or other activity of rhizobia. Apparently, rhizobial EPS absorbs water, because EPS layers are formed by the accumulation of various types of polymeric substances of high viscosity around bacterial cell walls (Potts 1994). Inoculation of plant roots with EPS-producing Rhizobium sp. improved soil structure, probably, by increasing soil adhesion to roots and/or by enhancing the stability of soil aggregates around roots (Alami et al. 2000). Thus, an explanation for the survival of A. senegal seedlings with gly-cine betaine under drought stress (III) could be that the increased rhizobial population with enhanced EPS production and/or other activities colonises roots and rhizospheres, subsequently protecting roots from drought.
4.5.3. Effects of exogenous compatible solutes on rhizobia
Both glycine betaine and trehalose functioned as osmoprotectants for the cells of S. arboris strain 2180, when they occupied A. senegal soil exposed to drought stress. Although the drought stress signifi cantly reduced the numbers of culturable cells from 107 - 106 cfu g-1, the presence of small amounts of glycine betaine or trehalose (0.0001 and 0.0003 M) in the soil maintained the cfu counts at the same level as detected in non-stressed soils (III: Table 1). Based on previous data, (Sauvage et al. 1983, Bernard et al. 1986, Smith et al. 1988, Boncompagni et al. 1999), the roles of glycine betaine for sinorhizobia might have varied during the course of the plant experiment. Under favourable conditions and at the beginning of the drought experiments, the cells utilised glycine betaine as an additional carbon and/or nitrogen source but when the cells started to suffer from water defi cit, glycine betaine was used as an osmo-protectant. Benefi cial effect of trehalose might also contribute to its ability to protect cells from desiccation (Bushby and Marshall 1977, Potts 1994).
4.6. RESPONSES OF S. ARBORIS POPULATIONS TO CHANGING ENVIRONMENTAL CONDITIONS
Bacteria were earlier considered to exist as simple, solitary cells. This opinion has changed and bacterial populations are viewed as colonial or multicellu-lar organisms that use, for example, cell-to-cell communication to facilitate their adaptation to environmental fl uctuations (Whitehead et al. 2001). The cell-to-cell communication at high population density (commonly known as quorum sensing and mostly through N-acyl homoserine lactones in Gram-negative bacteria) prevents single bacteria from spending resources for a certain function before the bacterial population is big enough. According
67
to recent data (e.g. Swift et al. 2001) rhizobia possess four quorum sensing systems, which control several phenotypes: restriction of nodule number, root nodulation, conjugation of plasmids, growth inhibition and stationary phase adaptation.
To ensure survival in changing environments, bacteria have regulatory elements that allow them to respond rapidly to stress conditions. The sigma factors RpoS (ss) and RpoE (sss) belongs to these elements. RpoS plays an important role when bacteria transform from the exponential growth phase to the stationary phase, and during starvation process and under osmotic stress (Kolter et al. 1993, Munro et al. 1995). In Pseudomonas fl uorescens, the sigma factor RpoE controled EPS production and tolerance of towards desiccation and osmotic stress (Schnider-Keel et al. 2001).
Multiphasic alterations in the expression of bacterial phenotypes, the phase variations, continuously generate cells with varied phenotypes leading to a mixed population. This type of diversity ensures that a certain percentage of cells in the population will express the phenotype needed for survival in new environmental conditions. Phase variations are generally reversible and random events occurring at high frequency (>10-5 per generations; Hender-son et al. 1999). When Sinorhizobium cells were exposed to heat, drought, salt and osmotic stresses in my thesis work, several extraordinary features appeared.
4.6.1. Alterations in EPS production
Rhizobia, as other Gram-negative bacteria, produce variety of surface polysaccharides, including EPS, LPS, CPS and periplasmic ß-glucans. EPS and LPS are important when rhizobia infect root hairs and live as bacteroids inside nodules. Periplasmic ß-glucans play an important role during osmotic adaptation (Kijne 1992, Brewin 1998).
S. arboris strain 1552 as well as its gene-modifi ed derivatives, formed slimy colonies on agar, indicating that they were EPS producers (II, IV). If the medium contained excess glucose or trehalose, it was apparently used for both EPS production and cell growth (III). Generally, when strain 1552 was grown in liquid medium, the optical cell density determined with a spectrophotometer increased in a proportion similar to the cell numbers (IV, Fig. 1A). However, in thermally stressed S. arboris cultures, the fi nal optical density did not correlate with total cell counts or the numbers of culturable cells (IV; Fig. 1B, Tables 1, 2). The reason for this phenomenon is unknown.
Strain 1552 also showed another peculiarity when grown under heat stress. A small portion of the colonies changed their morphology from mucous to dry when they were resuscitated from lyophilised cultures (L. Räsänen unpblished), frozen cultures (IV) and thermally stressed A. senegal rhizospheres (II) and liquid cultures (IV). The experiments performed with
68
the dry morphotype showed that the new colony morphology was a stable property (IV: Fig. 2C, Tables 1, 2), and that the dry morphotype had retained its nodulation ability (II). The change in EPS production seemed not to be involved in improved heat tolerance (IV; Fig. 2). Instead, the cell morpho-logy between the mucous and dry form differed to some degree when the two morphotypes were grown as liquid cultures at 40ºC (IV: Fig. 4b-e, 5b, c). This suggested that additional changes in cell morphology and physiology were also associated with the altered EPS production.
In Gram-negative bacteria, phase variation particularly involves alterations in expressions of cell surface structures (e.g. fi mbria, fl agella, outer membrane, proteins, EPS, LPS), which often give rise to changes in observable phenotypes (such as colonial morphology; Henderson et al. 1999). Thus, in the case of S. arboris, the temperature shift, such as cold shock, might have induced a change of the colony morphology through the phase variation mechanisms. A halotolerant S. meliloti strain showed a decrease in mucoidy and produced 40% less EPSs when the medium was supplemented with NaCl (Lloret et al. 1998). In S. meliloti, EPS consists of EPS I (succinoglucan) and EPS II (galac-toglucan). Later studies revealed that the changes in colony morphology in salt were linked to EPS II. Regulation of EPS II appeared to be complex, varying between strains. Possibly, regulation of EPS I and EPS II genes depends on the environmental conditions (Lloret et al. 2002).
In R. galegae the tolerance of low pH was associated with the abundant EPS production and long O-chain LPS (Räsänen and Lindström 1997). EPS tends to absorb water (Potts 1994) and EPS is suggested to protect bacteria from desiccation by altering their microenvironment (Roberson and Fires-tone 1992). The protective role of EPS against desiccation could explain why the colony morphology of S. arboris 1552 cells remained unchanged or became even more mucous when the bacteria were obtained from drought-affected A. senegal soils (L. Räsänen, unpublished).
The complete sequence of the megaplasmid pSymB revealed that S. meli-loti carries many more putative polysaccharide synthesis genes than was previously known. From the 11 gene clusters (containing presumably 188 genes, 12% of the genes on pSymB), the existence of nine clusters was unknown. The large number of surface polysaccharide genes was supposed to refl ect the very different conditions (e.g. desiccation and starvation) and environments (e.g., soil, rhizosphere, and legume nodule) to which S. meli-loti has had to adapt (Finan et al. 2001).
4.6.2. Changes in cell activity and culturability
Many studies have indicated that though bacteria rapidly lost culturability after inoculation to soil, defi ned as visible colony formation on plates, the cells were still metabolically active (Binnerup et al. 1993, Winding et al.
69
1994, Heijnen et al. 1995). This reproducible loss of culturability but main-tenance of metabolic activity, studied especially in Vibrio sp., led to the description of bacterial cells in this state as ”Viable But NonCulturable” (VBNC; McDougald et al. 1998). Thus, natural bacterial population may consist of i) culturable cells, which can be quantifi ed as colony-forming units (cfu) on solid media, ii) VBNC cells, and iii) dead cells. However, the exact defi nitions are still under discussion (Kell et al. 1998).
During the routine growth tests with S. arboris strain 1552 in liquid medium, it appeared that the strain grew oddly at 40ºC compared to at 28 and 37ºC (IV: Fig.1). Later studies revealed that in thermally stressed S. arboris cultures the number of culturable cells (i.e. cfu counts) could decrease considerably immediately after inoculation (IV: Fig.1B), but after prolonged incubation, the cultures transiently regained culturability. To fi nd out whether this phenomenon was associated to the VBNC state, three fl uo-rescent stains refl ecting different physiological cell states were applied: i) acridine orange – total cell numbers (IV; Fig. 4a-c, 5a, b), ii) Sytox Green - dead or dying cells with permeable cell membrane (IV; Fig. 4f), and iii) SFDA (fl uorochromo 5- (and)-sulfofl uorescein diacetate) - esterase active cells (IV: Fig. 4d, e, 5c). Esterases were assumed to indicate basal enzyme activity, being independent of the growth status and expressed by bacteria during their entire lifetime. Thus, SFDA stained cells represented metabo-lically active cells. Moreover, strain 1552 was tagged with the luc gene coding for the fi refl y (Photinus pyralis) luciferase (IV). The light output reac-tion (560 nm) of the fi refl y luciferase requires ATP, in addition to the pre-sence of the D-luciferin substrate, O2, and Mg2+ (Bronstein et al. 1994). Therefore, S. arboris cells tagged with the luc gene served as a sensitive indicator of cellular energy reserves and growth status.
In heat-stressed S. arboris cultures, there were major changes in luciferase activity and in the numbers of culturable cells but only minor changes in the numbers of esterase active cells (IV: Fig. 2A, 3, Table 1). This suggested that although the cells had low energy reserves and were no longer able to be cultured on agar, they still maintained some basal enzyme activity. Thus, these cells could be classifi ed as VBNC cells by broad defi nition. Copper in the growth medium, and starvation together with salinity have induced VBNC state in rhizobia (Alexander et al. 1999, Lippi et al. 2000). The present work (IV) indicated that a similar phenomenon may occur when rhizobia are cultured under heat stress.
McDougald et al. (1998) proposed that there are subpopulations within VBNC cultures, refl ecting stages of VBNC formation. In the initial stage, cells lose culturability but maintain the potential for resuscitation. In later stages, VBNC cells will gradually degrade and lose the potential for resuscitation (McDougald et al. 1998). Pseudomonas fl uorescens tagged with the green fl uorescent protein (gfp) gene was observed to remain fl uorescent following
70
starvation and entry into the VBNC state. Based on the GFP fl uorescence and the fl uorescent dye propidium iodide (stains cells with permeable cell mem-branes), three cell categories were distinguished: i) dead cells, ii) damaged/dying cells, and iii) viable cells (Lowder et al. 2000). Similarly, thermally stressed S. arboris cell populations could be divided into three categories: i) dead cells, stained with Sytox Green, ii) viable cells with energy reserves (cells forming colonies on agar media and showing high luciferase activity), and iii) VBNC cells, showing esterase activity but negligible luciferase acti-vity and not able to form colonies on agar.
The extent to which VBNC cells containing resting forms can be resusci-tated is not yet known. However, the sudden changes in luciferase activity, both positive and negative (IV Fig. 2C, Tables 1 and 2), suggested that some signalling occurred within S. arboris populations.
4.6.3. Moderately and highly stressed cultures
The studies with S. arboris strain 1552 tagged with the luc gene (= strain 2190), revealed that the heat-stressed cultures did not grow in a logistic manner. Based on cfu counts and luciferase activity, different growth phases of the cell cycle could be distinguished (lag, exponential, stationary, and decline phases), but the number and length of each phase varied considerably between repli-cate cultures (IV: Fig. 2B, C). Thermally stressed cultures could be separated in moderately stressed and highly stressed cultures. In the former case, the change in cfu counts and luciferase activity were small and gradual (IV: Fig. 2B, Tables 1 and 2) whereas in the latter case the two parameters changed drastically (IV: Fig. 2C, Tables 1 and 2), and the cultures had a long lag phase and/or a stationary phase was short or lacking (IV: Fig. 2C).
The rapid adaptation of E. coli strains to high growth temperature was explained by equivalent useful mutations occurring at the same rate both at 37°C and 42°C, but being lost in favourable growth conditions because of a smaller selective advantage (Bennett et al. 1990). Multiphasic phase variation, which ensures that a certain percentage of cells express proper-ties needed for survival in new conditions (Henderson et al. 1999), toget-her with other regulator mechanisms could explain why replicate S. arboris cultures responded differently to thermal stress. In the beginning the cultu-res contained some cells that could tolerate heat or were capable of beco-ming heat tolerant. Other biological response systems could strengthen or impair the development of heat tolerance. If the cells were heat-adapted, they became only moderately stressed during heat treatment. On the other hand, the cfu counts and luciferase activity of non-adapted cells were dec-reased considerably during the lag phase and the culture was regarded as highly stressed. This phenomenon might explain why there was no defi nite maximum growth temperature for liquid rhizobial cultures, above which
71
growth suddenly would cease (IV). Instead, the higher the growth tempera-ture was, the fewer replicated cultures recovered, or they failed to recover.
4.6.4. Changes in cell morphology
In indeterminate nodules of many temperature legumes, differentiation of vegetative rhizobial cells into N2-fi xing bacteroids involves an increase in size and pleomorphism (Fred et al. 1939, Vasse et al. 1990, Sprent 2001). Some rhizobial species are known to be capable of changing cell morphol-ogy in particular conditions, such as in media containing some stress factor, for example, at low or high pH, under osmotic stress, N defi ciency and low O2 concentration, and with the presence of bacteriophages or succinate (Fred et al. 1939, Pankhurst and Craig 1978, Urban and Dazzo 1982).
In this work, the cell morphology changed remarkable when S. arboris cells were exposed to heat stress. At 28 and 37 C S. arboris cells were typical of rhizobia; small, spherical rods (IV, Fig. 4a), but after inoculation at 40 C, they increased markedly in size and showed pleomorphism (IV: Fig 4b-e). Surprisingly, the changes in cell morphology were related to the growth phase. The cells sampled from slowly growing S. arboris cultures (i.e. lag and early exponential phase) were approximately 2-4 times larger than nor-mally, swollen and irregularly shaped (IV: Fig. 4b). In the exponential phase the cell morphology had changed again, and the cells were large, elonga-ted and slightly curved or wavy (IV: Fig. 4c, d, e). Especially elongated cells demonstrated basal esterase activity (IV: Fig. 4d). Obviously, swollen cells present in the lag phase were able to transform to elongated, culturable ones, after the culture started to activate or grow. This study together with previous studies suggests that the change in rhizobial cell morphology can be utilised as an indicator of cell stress.
There is evidence that rhizobia are able to change cell morphology in the soil. According to early study of Fred et al. (Fred et al. 1939), three forms of rhizobia were always recovered from inoculated sterile soils: i) unbanded rods, ii) banded or vacuolated rods, and iii) cocci. Proportions of each cell form varied during the incubation, the rods predominating at the start and the cocci after longer incubation. In a more recent study cells from several rhizobial species underwent changes in the cell wall morphology after they were transformed from broth cultures into peat. Obviously, these changes were associated to enhanced survival of cells in peat (Feng et al. 2002).
4.7. IS IT POSSIBLE TO EXPLOIT SYMBIOTIC N2 FIXATION BY INOCULATION OF ACACIA AND PROSOPIS SEEDLINGS?
Many studies have shown that inoculation of woody legume seedlings with effective rhizobia has improved plant growth in greenhouse and nursery
72
conditions in the absence of fertiliser N (e.g. Odee et al. 2002b, I). However, there are only few studies, in which effects of inoculation of tree seedlings has been followed in the fi eld (Lal and Khanna 1996, Requena et al. 2001). Although these studies suggested that, inoculation is worthwhile in soils defi -cient in N, extensive fi eld studies are still needed before the signifi cance of inoculation can be estimated. Perhaps, the major questions are: which role, at which growth state and under which environmental conditions N2 fi xa-tion plays a role for tree seedlings. Furthermore, is the symbiotic N2 fi xation more important for young seedlings than for adult trees? Many of these ques-tions can be solved only by using, in addition to traditional methods, modern isotopic and/or molecular techniques and by studying both partners, host plant and its symbiont, at the same time. However, because several factors infl uece concurrently the organisms living in the fi eld, laboratory and green-house experiments are also needed for more detailed studies, as well as to make it easier to decide where the future studies should be directed in the fi eld.
4.7.1. Evaluation of the role of inoculation in the fi eld
Tropical tree seedlings are likely to be planted to the fi eld either by direct seeding or after pre-germination in nurseries. Both techniques have both good and weak points. The former method is economic but mortality of seedlings can be high. The latter technique requires more work and rapidly growing taproots of Acacia and Prosopis spp. may become disturbed if they are not transplanted in time. However, presumably the establishment of inoculant rhizobia on the root systems may be more pronounced in nurseries than in the fi eld.
In arid and semiarid regions, seeding or transplanting of seedlings occur during the wet season. Then introduced rhizobia, applied together with seeds or seedlings, will penetrate deeper to the soil with growing roots and rain-water. Liquid inoculant added around the root collar immediately after tran-splanting turned out to be suitable inoculation procedure for Calliandra calothyrsys seedlings grown in local soils containing indigenous rhizobia (Odee et al. 2002b).
Under favourable conditions, Acacia and Prosopis seedlings can nodu-late rather fast, approximately within 10 days after sowing and inoculation of germinated seedlings (size 5-10 mm; II). How fast nodulation of pre-germi-nated seeds occurs under fi eld conditions is not known.
In general, Acacia and Prosopis seedlings grown under favourable condi-tions tended to have most N2-fi xing nodules on roots located near the soil surface (L. Räsänen, unpublished). Apparently, these nodules will decom-pose in the dry season due to drought and thermal stress (Monk et al. 1981, II, III). It is unknown, at which extent young tree seedlings, aged between 0
73
and 5 years, have N2-fi xing nodules in deeper soils, whether these nodules are perennial or at which extent new nodules are formed during next wet season. Furthermore, it is not known how long time the introduced rhizobia will persist in arid and semiarid soils and at which extent it depends on the properties of the strain. However, it can be possible that inoculant rhizo-bia will colonise soils for rather long time and survive trough unfavourable growth periods by entering VBNC state.
Although there are numerous open questions to be research, the follo-wing aspects support application of inoculation on Acacia and Prosopis see-dlings:
Assuring formation of N2-fi xing nodules. Development both of effective and ineffective nodules consumes plant resources but in the latter case the host plant does not get any benefi t (N). Inoculation could be especially important when planting in to the fi eld introduced tree species that have dif-ferent rhizobial preferences compared to those of local ones, or when the seedlings are grown in nurseries in local soils that may contain ineffective indigenous rhizobia.
Improvement of N2 fi xation under stress. Numerous studies on crop legu-mes have demonstrated that by application high amounts of rhizobia per legume seed it is possible to improve nodulation and N2 fi xation under stress conditions (Lupwayi et al. 2000 and references therein). Large rhizo-bial population may protect plant roots due to their metabolic activity and capacity to produce EPS (III).
Increased root growth. There are evidences that nodulated N2-fi xing legumes are capable of taking up more soil N compared to their non-nodu-lated counterparts, possibly because the fi xed N supports increased root growth (Unkovich and Pate 2000). Improved root growth would be impor-tant for Acacia and Prosopis seedlings, because their survival through dry periods is often depending on that whether the roots reach deep water sour-ces or not.
4.7.2. Inoculation as a procedure
Inoculation is benefi cial for tree seedlings only, if inoculants of good quality is available. I.e. it should be pure (no contaminants) and contain high amount of viable, effective and tested rhizobia (107 - 109 cells g-1; Lupwayi et al. 2000). In addition, the inoculant strain should be stable in its symbiotic capacities, adequately competent against indigenous rhizobia and moderate tolerant of environmental stresses. The widely accepted minimum quantita-tive per-seeds standards are 103, 104 and 105 per small, medium and big seeds (Lupwayi et al. 2000). In some cases, most effective strain-tree prov-enance combinations could be proposed (Galiana et al. 1994), but testing of different combinations is tedious. Presumably, the inoculant consisting of a
74
few effective rhizobial strains is more reliable in changeable soil conditions than a single-strain inoculant (Sutherland et al. 2000).
Properties of the host tree play a big role. There are differences between tree provenances with respect to growth, effi ciency of the symbiosis and resistance to various stresses (Aswathappa et al. 1987, Craig et al. 1991, Kumar et al. 1999). The last-mentioned property is important because the host tree determines growth ranges (Kumar et al. 1999).
Defi ciency of P is quite common in African soils (Dakora and Keya 1997), and low availability of P has limited symbiotic performance of Acacia see-dlings in Australia (Hansen and Pate 1987). Colonisation of plant roots by arbuscular mycorrhizal fungi (AM fungi) helps plants to take up P that is immobile in the soil and may improve plant drought tolerance (Michelsen and Rosendahl 1990, Ruiz-Lozano et al. 2001). Several studies have indica-ted that dual inoculation of Acacia seedlings with rhizobia and AM fungi produced better results in plant growth and nutrition, and in soil fertility and quality than inoculation with rhizobia or AM-fungi alone (Colonna et al. 1991, Hatimi 1999, Requena et al. 2001). Dual inoculation has also inc-reased the concentration of a compatible solute, proline in the leaf tissue of Australian A. cyanophylla (Hatimi 1999).
75
5. SUMMARY
In this thesis I fi rst studied symbiotic properties and stress tolerance of fi ve putative inoculant strains belonging to the fast-growing Sinorhizobium species, namely S. arboris, S. kostiense, S. saheli and S. terangae bv. acaciae. The strains were previously isolated from the nodules of Acacia or Prosopis trees growing in arid and semiarid in the Sudan and Senegal.
Host specifi city. The above fi ve strains induced N2-fi xing nodules on all African Acacia tested (A. mellifera, A. nilotica, A. oerfota (synonym A. nubica), A. nilotica, A. senegal, A. seyal, A. sieberiana, A. tortilis subsp. rad-diana) and on Latin American A. angustissima. In addition, the strains were able to induce effective nodules on Afro-Asian P. cineraria and on several Latin American Prosopis spp., both on native P. chilensis and P. pallida, and on those that were introduced into Africa from Latin America (P. julifolora, P. chilensis). The nodulation pattern of S. saheli confi rmed the earlier view that as in S. terangae, two biovars, bv. acaciae and bv. sesbanie, could be dis-tinguished in S. saheli. The African Sinorhizobium strains induced ineffective nodules on African P. africana and Australian A. holosericea.
Infection mode and nodule type. In general, all species had root hairs on their roots but were particularly sparse in Acacia spp. The infection mode resembled that of classical root hair infection. The infection threads, along which the rhizobia penetrate the hair, were mostly formed on short hairs in acacias and Afro-Asian P. cineraria and on longer ones in Latin Ame-rican Prosopis spp. Elongation and ramifi cation of the nodules indicated that Acacia and Prosopis had indeterminate nodules although a persistent apical meristem, characteristic feature of the indeterminate nodule, was not easily detectable.
Stress tolerance of the two major effective Sinorhizobium strains. Two major strains used in this study had competent and stable symbiotic proper-ties. They also showed to be adapted to survive in arid and semiarid soils. S. arboris strain 1552 was salt and osmotic tolerant strain whereas strain 1496 was sensitive to NaCl but tolerated higher growth temperatures.
Infection and nodulation under heat and drought stress. The unfavou-rably high root temperature and drought stress caused A. senegal hairs to be partly abnormally deformed. Short and swollen bottle-like hairs were typical under heat stress and very short or dwarfed, swollen hairs were characteristic of water-stressed hairs. Under both stresses, infection threads were rare, and they seemed to be either disintegrated or they had weak glucuronidase acti-vity. Nodulation of A. senegal seedlings was impaired at 38°C, reduced to 50% at 40°C and completely inhibited at 42°C. The drought stress retarded or stopped the normal nodule development so that the numbers of nodule initials were higher on drought-stressed plants than on non-stressed ones. For the numbers of true nodules the situation was reversed. The nodules formed
76
showed signs of premature senescence and displayed weak glucuronidase activity, indicating low cell numbers and/or low activity of rhizobia occu-pying nodules.
The role of compatible solutes for plants and bacteria. Regarding endo-genous glycine betaine, A. senegal was a non-accumulator plant. 60% of the plants grown in drought-affected soils with 0.0003 M glycine betaine were still alive 42 days after inoculation whereas the drought-stressed plants grown without any increments were wilted 30 days after inoculation. The treated soils tended to have more water left compared to that in untreated soils. Possibly, glycine betaine enabled A. senegal seedlings to control water uptake and transpiration more effectively. 0.01 or 0.05 M glycine betaine and trehalose functioned as osmoprotectants for S. arboris strain 1552 and S. saheli strain 1496 grown in liquid media with 9 and 17% PEG 6000 but in the case of salt stress (NaCl) only trehalose had a favourable effect. In drought-affected A. senegal soils the number of culturable rhizobia dec-reased signifi cantly but the addition of 0.0001 or 0.0003 M glycine betaine and 0.0003 M trehalose maintained cfu counts of at the same level as in non-stressed soils (107 cfu g-1).
The heat induced peculiar changes in the numbers of culturable cells and cell morphology when S. arboris 1552 cells were grown in batch cul-tures. Several complementary techniques (luciferase activity of S. arboris strain marked with the luc gene; plate counts; optical cell density; two fl uorescent stains in order to microscopically identify metabolically active and dead cells) were applied. It appeared that although the numbers of cultural cells and cellular energy reserves decreased considerably during heat stress, a majority of the cell population maintained basal enzymatic activity. In other words, under adverse conditions, rhizobial cells did not only die but also entered into a state, in which they were viable but non-culturable. Change in cell morphology, which was observed to be partly related to the growth phase of the cell culture, can be utilised as an indi-cator of cell stress.
Heat experiments, both in batch cultures and in soil conditions, caused changes in colony morphology in S. arboris strain 1552. Normally, this strain produces mucous colonies but after the heat treatments, small por-tion of the colonies grew as small and compact ones. This dry colony mor-phology was a stable property for the bacterium, and it had retained its symbiotic properties. Similar characteristic was observed, when strain 1552 was retrieved from lyophilised or frozen cultures. Change in colony mor-phology might be associated with the phase variation mechanisms, which in Gram-negative bacteria involves alteration in expression of surface struc-tures, such as EPS and LPS. The above mechanism ensures that a certain percentage of cell population will express phenotype needed for a survival in new conditions.
77
6. TIIVISTELMÄ
Tropiikissa palkokasvit, kuten Acacia ja Prosopis -sukujen edustajat, ovat pääasiassa puita ja pensaita. Nykyisin tunnetaan noin 1200 Acacia -lajia, joista 850 lajia tavataan Australiassa ja 135 lajia Afrikassa. Sen sijaan Pro-sopis -sukuun kuuluvat lajit kasvavat lähinnä Amerikan mantereella (40/44). Monet Acacia ja Prosopis –lajit ovat sopeutuneet kasvamaan tropiikin kui-villa ja puolikuivilla alueilla. Jotkut lajit voivat kasvattaa pääjuurta jopa 10-30 metrin syvyyteen vedensaannin turvaamiseksi. Akaasioita ja prosopik-sia on perinteisesti hyödynnetty monipuolisesti mm. rehuna, polttopuuna, puuhiilen ja lääkkeiden raaka-aineena, viherlannoitteena, rakennusaineena ja ihmisravintona. A. senegal –puusta saadaan arabikumia, jota käytetään mm. elintarvikkeiden ja lääkkeiden valmistuksessa. Kestävät ja monikäyttöi-set akaasiat ja prosopikset sopivatkin erityisen hyvin karujen, eroosion vai-vaamien alueiden metsittämiseen.
Ruohovartisten palkokasvien tavoin (esim. apila ja herne) suurin osa Aca-cia- ja Prosopis -lajeista kykenee symbioosissa juurinystyräbakteerien (rit-sobit) kanssa sitomaan ilmakehän typpeä (N2) kasveille käyttökelpoiseen muotoon. Monien viljeltävien, ruohovartisten palkokasvien ja niitä nystyröi-vien bakteerilajien väliset symbioosit tunnetaan varsin hyvin. Sen sijaan palkokasvipuiden ja niiden ritsobeiden välisestä symbioosista ja sen merki-tyksestä taimille ja aikuisille puille tiedetään erittäin vähän.
Tässä työssä pyrittiin selvittämään sudanilaisten A. senegal – ja P. chilen-sis puiden nystyröistä eristettyjen, eri Sinorhizobium-sukuja edustavien kan-tojen symbioottisia ominaisuuksia, ts. mitä muita Acacia ja Prosopis -lajeja kannat nystyröivät ja miten ritsobi pääsee puun taimen juureen. Tämän lisäksi tarkasteltiin muodostuneiden nystyröiden rakennetta. Työssä tutkittiin myös korkean lämpötilan ja kuivuuden vaikutuksia ritsobien ja A. senegal- taimien kasvuun ja symbioosin kehittymiseen. Lisäksi kartoitettiin niitä tekijöitä, jotka mahdollisesti parantaisivat ritsobien ja A. senegal -taimien osmoottisen ja kuivuusstressin sietoa.
Sudanilaisten ritsobien isäntäspesifi syys oli laaja, eli ne muodostivat typpeä sitovia nystyröitä sekä useisiin afrikkalaisiin Acacia-lajeihin että Latinalaisesta Amerikasta kotoisiin oleviin Prosopis-lajeihin. Ritsobikannat indusoivat tehottomia eli typen sidontaan kykenemättömiä, nystyröitä muis-tuttavia rakenteita australialaiseen A. holosericea -lajiin ja ainoaan koto-peräiseen afrikkalaiseen Prosopis-lajiin (P. africana). Ritsobit pääsevät Acacia ja Prosopis -puun juureen juurikarvaan muodostuneen käytävän eli infektio-langan kautta. Muodostuneiden nystyröiden kärjessä on jatkuvasti kasvava meristeemisolukko, mikä merkitsee sitä, että puiden nystyrät voisivat olla monivuotisia.
Yleensä ritsobit sietivät korkean lämpötilan ja kuivuuden aiheuttamaa stressiä paremmin kuin A. senegal –taimet. Työn kuluessa ilmeni, että
78
kasvatettaessa ritsobeja stressiolosuhteissa, joko ravintoliemessä tai/ja A. senegal-taimien juuristossa niiden solu -ja pesäkemorphologia sekä kas-vukäyrät poikkesivat suotuisissa oloissa kasvaneiden ritsobien vastaavista ominaisuuksta. Viljeltäessä kuumuusstressistä kärsineitä ritsobeja maljoilla osa pesäkkeistä oli muuttunut pintarakenteeltaan kannalle ominaisen limai-sen muodon sijasta pieniksi, pinnaltaan kuiviksi pesäkkeiksi. Tämä ilmiö liittyy Gram-negatiivisten ritsobeiden pintarakenteisiin kuuluvien pintasoke-reiden (eksopolysakkaridit eli EPS) tuotannon muuttumiseen. Stressatut ritso-bit kykenivät menemään tilaan, jossa ne olivat eläviä, mutta eivät viljeltäviä. Tämä muoto (VBNC, Viable-But-Not-Culturable) saattaisi olla jonkinlainen säilymismuoto. Ilmeisesti samankin kannan ritsobipopulaatiossa muodostuu jatkuvasti ominaisuuksiltaan erilaisia osapopulaatiota, mikä varmistaa sen, että ympäristön muuttuessa äkillisesti ainakin osa bakteereista jää henkiin. Glysiinibetaiini ja trehaloosi ovat osmolyyttisiä yhdisteitä, joita monet bak-teerit voivat syntetisoida tai ottaa ympäristöstä suojautuakseen suolan ja kuivuuden aiheuttamalta osmoottiselta stressiltä. Kummankin aineen lisäys kasvualustaan tai A. senegal –taimien juuristoon lisäsi ritsobien kasvua osmoottisen ja kuivuusstressin aikana.
Sekä A. senegal -taimien kasvualustan korkea lämpötila (38-40°C) että kuivuus vähensivät nystyröiden määrää juurissa. Korkeammassa lämpöti-lassa (42°C) tai kuivuuden jatkuessa nystyröiden kehittyminen pysähtyi kokonaan. Optimioloissa nuoret Acacia ja Prosopis -taimet pyrkivät kas-vattamaan suurimmat typpeä sitovat nystyrät lähelle maan pintaa 1-5 cm syvyyteen. Sen sijaan luonnossa maan korkea lämpötila estänee juuri-nystyröiden muodostumista 10 cm syvyyteen saakka, ja kuivana kautena huomattavasti syvemmältäkin. Jotkut suolaa sietävät kasvit suojautuvat sen aiheuttamilta haitoilta keräämällä soluihin glysiinibetaiinia. Glysiinibetaii-nin lisäys A. senegal -tainten juuristoon alensi veden puutteesta kärsivien kasvien kuolleisuutta selvästi. Oletettavasti glysiinibetaiini tehosti taimien vedenkäyttöä. Trehaloosin lisäyksellä oli samantapainen, mutta lievempi vaikutus. Trehaloosin tiedetään suojaavan biologisia rakenteita kuivuudelta. Lisääntynyt ritsobipopulaatio sinänsä voi myös stimuloida juurten kasvua ja suojata juuristoa.
Palkokasvipuiden ymppäys eli ritsobien lisäys kasvualustaan taimitar-hoilla tai maahan siementen kylvön yhteydessä näyttää varmistavan sen, että puulajille sopivaa ritsobilajia on saatavilla heti istutuksen jälkeen. Jos taimet/siemenet istutetaan sadeajan alussa, jolloin maassa on riittävästi kosteutta, ensimmäiset typpeä sitovat nystyrät kehittyvät parin viikon kuluessa. Typen-sidonta saattaa olla ratkaisevaa juuren kasvun kannalta. Selvitäkseen kui-vasta kaudesta taimen on ehdittävä kasvattaa pääjuurta mahdollisimman syvälle maahan ennen kauden alkamista. Kenttätutkimuksia kuitenkin tarvi-taan, jotta ymmärrettäisiin paremmin missä vaiheessa ja minkälaisissa olo-suhteissa Acacia- ja Prosopis -taimien ymppäyksestä olisi hyötyä.
79
7. ACKNOWLEDGEMENTS
This work was carried out during the years 1994-2002 at Department of Applied Chemistry and Microbio-
logy, University of Helsinki. Work was supported by funds and grants of The Academy of Finland, EU contract No. TS3-CT93-0232, The Kemira Oyj Founda-tion, Nordic Network, University of Hel-sinki and The Finnish Cultural Foundation, which is gratefully acknowledged.
I’d like to express my warmest thanks to my supervisor, docent Kristina Lind-ström, for the possibility to work in her nice research group, N2-group, in an inspiring, international and relaxed atmosphere. I am especially grateful for her giving me the opportunity to work with the fascinating topic, to see the world and for correcting my English writing.
I would like to thank Elena Lapina-Balk for her nice company and help during the experimental works. Someti-mes the work was pleasant like working on beautiful summer days under apple trees, and sometimes we were as stres-sed as our research objects. I am grateful to Salla Saijets, who promptly entered into the diffi cult world of osmotic and drought stresses, and Eva Tas for her patie-nce when teaching me the use of different computer programs.
I want to thank all present and former members of the N2 group for their help in the lab , and for the good atmosphere and relaxing company during coffee and lunch hours: Leena S, Gera, Zewdu, Minna, Aneta, Leone, Jyrki, Kati, Petri, Jenny, Inka, Endalka, Kaisa W, Seppo, Aimo, Gise, Vesa, Kaisa H, Sirpa, Anne, Gilles, Lena, Hannamari and all the others not mentioned here.
I am grateful to Professor Mirja Sal-kinoja-Salonen for her kind support, and to all the staff of the Department of App-lied Chemistry and Microbiology for good working facilities and for the friendly attitude that I have always experienced.
I wish to thank co-authors, Prof. Janet Jansson for giving me the opportunity to visit Stockholm University and Annelie Elväng for introducing the world of luc-marked bacteria. The other co-authors, Prof. Janet Sprent (University of Dundee, Scotland) and Dr. Kari Jokinen (Kemira Oyj, Helsinki) are also deeply acknow-ledged for their expertise.
Prof. Ken Giller (Wageningen Univer-sity, the Netherlands) and Dr. David Odee (Kenya Research Forestry Institute) are acknowledged for reviewing this manusc-ript and for their valuable comments that signifi cantly improved the text.
I like to thank my friends, Anne, Hanna, Kirsi, Kirsti and Jukka for several nice moments with good food and drinks, and my godmother Aune for her kind advises how to keep well.
Finally, my family deserves all my thanks for their love and support. My father Eino and my mother Anni, who is not with us any more, have offered a place where I could forget all my scientifi c thoughts and worries. My brothers Pekka and Tuomo have always been helpful in transporting me or my things. Without the practical help of my brother-in-law and graphic designer Kari and my sister Eeva-Liisa fi nishing this book would have been a much more stressing experience. As journalists they had the courage to ask all the silly but important questions.
Helsinki, November 2002
80
REFERENCES
AbdelGadir, A.H. and M. Alexander. 1997. Procedures to enhance heat resistance of Rhizobium. Plant Soil. 188, 93-100.
Abdelmoumen, H., A. Filali-Maltouf, M. Neyra, A. Belabed and M. Missbah El Idrissi. 1999. Effect of high salt con-centrations on the growth of rhizobia and response to added osmotica. J. Appl. Microbiol. 86, 889-898.
Agboma, P.C., M.G.K. Jones, P. Peltonen-Sainio, H. Rita and E. Pehu. 1997. Exogenous glycinebetaine enhances grain yield of maize, sorghum and wheat grown under two supplementary watering regimes. J. Agr. Crop Sci. 178, 29-37.
Ahmed, A.E.H. 1986. Some aspects of dry land afforestation in the Sudan with special reference to Acacia tortilis (Forsk.) Hayne, A. senegal Willd. and Prosopis chilensis (Molina) Stunz. For. Ecol. Manage. 16, 209-221.
Alami, Y., W. Achouak, C. Marol and T. Heulin. 2000. Rhizosphere soil aggregation and plant growth promotion of sunfl owers by an exopolysaccharide-producing Rhizobium sp. strain isolated from sunfl ower roots. Appl. Envi-ron. Microbiol. 66, 3393-3398.
Albus, U., R. Baier, O. Holst, A. Puhler and K. Niehaus. 2001. Suppression of an elicator-induced oxidative burst reac-tion in Medicago sativa cell cultures by Sinorhizobium meliloti lipopolysaccharides. New Phytol. 151, 597-606.
Alexander, E., D. Pham and T.R. Steck. 1999. The viable-but-nonculturable condition is induced by copper in Agro-bacterium tumefaciens and Rhizobium leguminosarum. Appl. Environ. Microbiol. 65, 3754-3756.
Allen, O.N. and E.K. Allen. 1981. The Leguminosae. The University of Wisconsin Press, Madison, WI. 812 pages.
Andersson, M., M. Laukkanen, E.-L. Nurmiaho-Lassila, F.A. Rainey, S.I. Niemelä and M. Salkinoja-Salonen. 1995. Bacillus thermosphaericus sp. nov. a new thermophilic ureolytic Bacillus isolated from air. Syst. Appl. Microbiol. 18, 203-220.
Arayangkoon, T., H.H. Schomberg and R.W. Weaver. 1990. Nodulation and N2 fi xation at high root temperature. Plant Soil. 126, 209-213.
Aswathappa, N., N.E. Marcar and L.A.J. Thomson. 1987. Salt tolerance of Australian tropical and subtropical acacias, p. 70-73. In Turnbull, J.W., (ed.), Australian acacias in developing countries: proceedings of an international workshop held at Forestry Training Centre, Gympie, Qld., Australia, 4-7 August 1986. Brown Prior Anderson Pty Ltd, Victoria.
Athar, M. 1998. Drought tolerance by lentil rhizobia (Rhizobium leguminosarum) from arid and semiarid areas of Pakistan. Letters in Applied Microbiology. 26, 38-42.
Baird, L.M., R.A. Virginia and B.D. Webster. 1985. Development of root nodules in a woody legume, Prosopis, glan-dulosa Torr. Bot. Gaz. 146, 39-43.
Baldani, J.I. and R.W. Weaver. 1992. Survival of clover rhizobia and their plasmid-cured derivatives in soil under heat and drought stress. Soil. Biol. Biochem. 24, 737-742.
Barnet, Y.M. and P.C. Catt. 1991. Distribution and characteristics of root-nodule bacteria isolated from Australian Acacia spp. Plant Soil. 135, 109-120.
Barnet, Y.M., P.C. Catt and D.H. Hearne. 1985. Biological nitrogen fi xation and root-nodule bacteria (Rhizobium sp. and Bradyrhizobium sp.) in two rehabilitating sand dune areas planted with Acacia spp. Aust. J. Bot. 33, 595-610.
Barth, R.C. and J.O. Klemmedson. 1978. Shrub-induced spatial pattern of dry matter, nitrogen, and organic carbon. Soil Sci. Soc. Am. J. 42, 804-809.
Bellefontaine, R., A. Gaston and Y. Petrucci. 2000. Management of natural forests of dry tropical zones. FAO Con-servation Guide 32. FAO, Rome.
Belsky, A.J. 1994. Infl uences of trees on savanna productivity: tests of shade, nutrients, and tree-grass competition. Ecology. 75, 922-932.
Belsky, A.J., R.G. Amundson, J.M. Duxbury, S.J. Riha, A.R. Ali and S.M. Mwonga. 1989. The effects of trees on their physical, chemical and biological environments in a semi-arid savanna in Kenya. J. Appl. Ecol. 26, 1005-1024.
Bennett, A.F., K.M. Dao and R.E. Lenski. 1990. Rapid evolution in response to high-temperature selection. Nature. 346, 79-81.
81
Bentham, G. 1875. Revision of the suborder Mimoscae. Transactions of the Linnean Society of London. 30, 335-664.
Bergersen, F.J. 1997. Regulation of nitrogen fi xation in infected cells of leguminous root nodules in relation to O2 supply. Plant Soil. 191, 189-203.
Bernard, T., J.-A. Pocard, B. Perroud and D. Le Rudulier. 1986. Variations in the response of salt-stressed Rhizobium strains to betaines. Arch. Microbiol. 143, 359-364.
Bernhard-Reversat, F. 1982. Biogeochemical cycle of nitrogen in a semi-arid savanna. Oikos. 38, 321-332.
Bhojvaid, P.P. and V.R. Timmer. 1998. Soil dynamics in an age sequence of Prosopis julifl ora planted for sodic soil restoration in India. For. Ecol. Manage. 106, 181-193.
Bhuvaneswari, T.V., A.A. Bhagwat and W.D. Bauer. 1981. Transient susceptibility of root cells in four common legu-mes to nodulation by rhizobia. Plant Physiol. 68, 1144-1149.
Binnerup, S.J., D.F. Jensen, H. Thordal-Christensen and J. Sorensen. 1993. Detection of viable, but non-culturable Pseudomonas fl uorescens DF57 in soil using a microcolony epifl uorescence technique. FEMS Microbiol. Ecol. 12, 97-105.
Bloem, J. 1995. Fluorescent staining of microbes for total direct counts, p. 1-12. In Akkermans, A.D.L., J.D. Van Elsas and F.J. de Bruijn, (eds.), Molecular Microbial Ecology Manual 4.1.8. Kluwer Academic Publishers, Dordrecht.
Blunden, G., T. Jenkins and Y.W. Liu. 1997. Enhanced leaf clorophyll levels in plants treated with seaweed extract. J. Appl. Phycol. 8, 535-543.
Boncompagni, E., M. Osterås, M.-C. Poggi and D. Le Rudulier. 1999. Occurrence of choline and glycine betaine uptake and metabolism in the family Rhizobiaceae and their roles in osmoprotection. Appl. Environ. Microbiol. 65, 2072-2077.
Boogerd, F.C. and D. van Rossum. 1997. Nodulation of groundnut by Bradyrhizobium: a simple infection process by crack entry. FEMS Microbiol. Rev. 21, 5-27.
Botsford, J.L. and T.A. Lewis. 1990. Osmoregulation in Rhizobium meliloti: production of glutamic acid in response to osmotic stress. Appl. Environ. Microbiol. 56, 488-494.
Bray, E.A. 1997. Plant responses to water defi cit. Trends Plant Sci. 2, 48-54.
Breedveld, M.W., L.P.T.M. Zevenhuizen and A.J.B. Zehnder. 1991. Osmotically-regulated trehalose accumulation and cyclic beta-(1,2)-glucan excretion by Rhizobium leguminosarum biovar trifolii TA-1. Arch. Microbiol. 156, 501-506.
Breedveld, M.W., C. Dijkema, L.P.T.M. Zevenhuizen and A.J.B. Zehnder. 1993. Response of intracellular carbohyd-rates to a NaCl shock in Rhizobium leguminosarum biovar trifolii TA-1 and Rhizobium meliloti SU-47. J. Gen. Microbiol. 139, 3157-3163.
Brenan, J.P.M. 1965. The geographical relationships of the genera of Leguminosae in tropical Africa. Webbia. 19, 545-559.
Brewin, N.J. 1991. Development of the legume root nodule. Annu. Rev. Cell Biol. 7, 191-226.
Brewin, N.J. 1998. Tissue and cell invasion by Rhizobium: the structure and development of infection threads and symbiosomes, p. 417-429. In Spaink, H.P., A. Kondorosi and P.J.J. Hooykaas, (eds.), The Rhizobiaceae - Molecular Biology of Model Plant Associated Bacteria. Kluwer Academic Publishers, Dordrecht.
Brockwell, J., P.J. Bottomley and J.E. Thies. 1995. Manipulation of rhizobia microfl ora for improving legume produc-tivity and soil fertility: A critical assessment. Plant Soil. 174, 143-180.
Bronstein, I., J. Fortin, P.E. Stanley, G.S.A.B. Stewart and L.J. Kricka. 1994. Chemiluminescent and bioluminescent reporter gene assay. Anal. Biochem. 219, 169-181.
Bueno, P., M.J. Soto, M.P. Rodriguez-Rosales, J. Sanjuan, J. Olivares and J.P. Donaire. 2001. Time-course of lipoxy-genase, antioxidant enzyme activities and H2O2 accumulation during the early stages of Rhizobium-legume sym-biosis. New Phytol. 152, 91-96.
Bukhari, Y.M. 1997a. Cytoevolution of taxa in Acacia and Prosopis (Mimosaceae). Aust. J. Bot. 45, 879-891.
Bukhari, Y.M. 1997b. Nuclear DNA amounts in Acacia and Prosopis (Mimosaceae) and their evolutionary implica-tions. Hereditas. 126, 45-51.
Bukhari, Y.M. 1998. Tree-root infl uence on soil physical conditions, seedling establishment and natural thinning of Acacia seyal var. seyal on clays of Central Sudan. Agrofor. Syst. 42, 33-43.
82
Bukhari, Y.M., K. Koivu and A. Kanerva. 1998. Seed storage protein size variations and their signifi cance in the evolu-tion and the systematics of Acacia and Prosopis (Mimosaceae). Silvae Genetica. 47, 241-249.
Bukhari, Y.M., K. Koivu and P.M.A. Tigerstedt. 1999. Phylogenetic analysis of Acacia (Mimosaceae) as revealed from chloroplast RFLP data. Theor. Appl. Genet. . 98, 291-298.
Burdon, J.J., A.H. Gibson, S.D. Searle, M.J. Woods and J. Brockwell. 1999. Variation in the effectiveness of symbiotic associations between native rhizobia and temperate Australian Acacia: within-species interactions. J. Appl. Ecol. 36, 398-408.
Buresh, R.J. and G. Tian. 1998. Soil improvement by trees in sub-Saharan Africa. Agrofor. Syst. 38, 51-76.
Burkart, A. 1976. A monograph of the genus Prosopis (Leguminosae Subfam. Mimosoideae. J. Arnold Arbor. 57, 216-249.
Bushby, H.V.A. and K.C. Marshall. 1977. Desiccation-induced damage to the cell envelope of root-nodule bacteria. Soil. Biol. Biochem. 9, 149-152.
Cermola, M., E. Federova, R. Tate, A. Riccio, R. Favre and E.J. Patriarca. 2000. Nodule invasion and symbiosome differentiation during Rhizobium etli-Phaseolus vulgaris symbiosis. Mol. Plant-Microbe Interact. 13, 733-741.
Chandler, M.R. 1978. Some observation on infection of Arachis hypogaea L. by Rhizobium. J. Exp. Bot. 29, 749-755.
Chandler, M.R., R.A. Date and R.J. Roughley. 1982. Infection and root-nodule development in Stylosanthes species by Rhizobium. J. Exp. Bot. 33, 47-57.
Chappill, J.A. 1995. Cladistic analysis of the Leguminosae: the development of an explicit phylogenetic hypothesis, p. 1-9. In Crisp, M. and J.J. Doyle, (eds.), Advances in Legume Systematics 7: Phylogeny. Royal Botanic Gardens, Kew.
Chappill, J.A. and B.R. Maslin. 1995. A phylogenetic assessment of tribe Acacieae, p. 77-95. In Crisp, M. and J.J. Doyle, (eds.), Advances in Legume Systematics 7: Phylogeny. Royal Botanic Gardens, Kew.
Chatel, D.L. and C.A. Parker. 1973. Survival of fi eld-grown rhizobia over the dry summer period in Western Australia. Soil. Biol. Biochem. 5, 415-423.
Chen, W., E. Wang, S. Wang, Y. Li, X. Chen and Y. Li. 1995. Characteristics of Rhizobium thianshanense sp. nov., a moderately and slowly growing root nodule bacterium isolated from an arid and saline environment in Xinjiang, People’s Republic of China. Int. J. Syst. Bacteriol. 45, 153-159.
Chen, W.M., S. Laevens, T.M. Lee, T. Coenye, P. De Vos, M. Mergeay and P. Vandamme. 2001. Ralstonia taiwanensis sp. nov., isolated from root nodules of Mimosa species and sputum of cystis fi brosis patient. Int. J. Syst. Evol. Microbiol. 51, 1729-1735.
Colonna, J.P., D. Thoen, M. Ducousso and S. Badji. 1991. Comparative effects of Glomus mosseae and P fertilizer on foliar mineral composition of Acacia senegal seedlings inoculated with Rhizobium. Mycorrhiza. 1, 35-38.
Corby, H.D.L. 1988. Types of rhizobial nodules and their distribution among the Leguminosae. Kirkia. 13, 53-123.
Cornet, F., C. Otto, G. Ridaudo, H.G. Diem and Y. Dommergues. 1985. Nitrogen fi xation by Acacia holosericea grown in fi eld simulated conditions. Acta Oecol. 6, 211-218.
Craig, G.F., C.A. Atkins and D.T. Bell. 1991. Effect of salinity on growth of four strains of Rhizobium and their infec-tivity and effectiveness on two species of Acacia. Plant Soil. 133, 253-262.
Csonka, L.N. 1989. Physiological and genetic responses of bacteria to osmotic stress. Microbiol. Rev. 53, 121-147.
Dakora, F.D. and S.O. Keya. 1997. Contribution of legume nitrogen fi xation to sustainable agriculture in Sub-Saharan Africa. Soil. Biol. Biochem. 29, 809-817.
Dart, P. 1977. Infection and development of leguminous nodules, p. 367-472. In Hardy, R.W.F. and W.S. Silver, (eds.), A treatise on dinitrogen fi xation. A Wiley-Interscience Publication, John Wiley & Sons, New York.
Deans, J.D., O. Diagne, D.K. Lindley, M. Dione and J.A. Parkinson. 1999. Nutrient and organic-matter accumulation in Acacia senegal fallows over 18 years. For. Ecol. Manage. 124, 153-167.
de Lajudie, P., A. Willems, B. Pot, D. Dewettinck, G. Maestrojuan, M. Neyra, M.D. Collins, B. Dreyfus, K. Kersters and M. Gillis. 1994. Polyphasic taxonomy of rhizobia: emendation of the genus Sinorhizobium and description of Sinorhizobium meliloti comb. nov., Sinorhizobium saheli sp. nov., and Sinorhizobium teranga sp. nov. Int. J. Syst. Bacteriol. 44, 715-733.
83
de Lajudie, P., A. Willems, G. Nick, S.H. Mohamed, U. Torck, R. Coopman, A. Filali-Maltouf, K. Kersters, B. Dreyfus, K. Lindström and M. Gillis. 1999. Agrobacterium bv. 1 strains isolated from nodules of tropical legumes. Syst. Appl. Microbiol. 22, 119-132.
Delgado, M.J., J.M. Garrido, F. Ligero and C. Lluch. 1993. Nitrogen fi xation and carbon metabolism by nodules and bacteroids of pea plants under sodium chloride stress. Physiol. Plant. 89, 824-829.
Delgado, M.J., F. Ligero and C. Lluch. 1994. Effects of salt stress on growth and nitrogen fi xation by pea, faba-bean, common bean and soybean plants. Soil. Biol. Biochem. 26, 371-376.
De Troch, P. and J. Vanderleyden. 1996. Surface properties and motility of Rhizobium and Azospirillum in relation to plant root attachment. Microb. Ecol. 32, 149-169.
D’Haeze, W., C. Chaparro, A. De Keyser, S. Deleu, R. De Rycke, S. Goormachtig, S. Lievens, R. Mathis, K. Schroey-ers, W. Van de Velde, D. Vereecke, C. Verplancke and M. Holsters. 2002. Role of reactive oxygen species and ethylene in programmed cell death during nodule initiation on Sesbania rostrata, p. 137-141. In Finan, T.M., M.R. O’Brian, D.B. Layzell, J.K. Vessey and W. Newton, (eds.), Nitrogen Fixation: Global Perspectives. Proceedings of the 13th International Congress on Nitrogen Fixation, Hamilton, Ontario, Canada, 2-7 July 2001. CABI Publishing, Wallingford.
Diagne, O. and D.D. Baker. 1994. Quantifi cation of symbiotic N fi xation by Prosopis julifl ora (Swartz) D.C. using 15N-isotope dilution methodology. Soil. Biol. Biochem. 26, 1709-1710.
Diaz-Zorita, M., M.V. Fernandez-Canigia and G.A. Grosso. 2001. Applications of foliar fertilizers containing glycine-betaine improve wheat yields. J. Agr. Crop Sci. 186, 209-215.
Ditta, G., S. Stanfi eld, D. Corbin and D.R. Helinski. 1980. Broad host range DNA cloning system for Gram-negative bacteria: construction of a gene bank of Rhizobium meliloti. Proc. Natl. Acad. Sci. USA. 12, 7347-7351.
Doyle, J.J. 1994. Phylogeny of the legume family: an approach to understanding the origins of nodulation. Annu. Rev. Ecol. Syst. 25, 325-349.
Dreyfus, B.L. and Y.R. Dommergues. 1981. Nodulation of Acacia species by fast- and slow-growing tropical strains of Rhizobium. Appl. Environ. Microbiol. 41, 97-99.
Dupuy, N.C. and B.L. Dreyfus. 1992. Bradyrhizobium populations occur in deep soil under the leguminous tree Acacia albida. Appl. Environ. Microbiol. 58, 2415-2419.
Dupuy, N., A. Willems, B. Pot, D. Dewettinck, I. Vandenbruaene, G. Maestrojuan, B. Dreyfus, K. Kersters, M.D. Col-lins and M. Gillis. 1994. Phenotypic and genotypic characterization of bradyrhizobia nodulating the leguminous tree Acacia albida. Int. J. Syst. Bacteriol. 44, 461-473.
Ehleringer, J.R., H.A. Mooney, P.W. Rundel, R.D. Evans, B. Palma and J. Delatorre. 1992. Lack of nitrogen cycling in the Atacama Desert. Nature. 359, 316-318.
Elsheikh, E.A.E. and M. Wood. 1990. Rhizobia and bradyrhizobia under salt stress: possible role of trehalose in osmo-regulation. Letters in Applied Microbiology. 10, 127-129.
Erskine, P.D., G.R. Stewart, S. Schmidt, M.H. Turnbull, M. Unkovich and J.S. Pate. 1996. Water availability - a phy-siological constraint on nitrate utilization in plants of Australian semi-arid mulga woodlands. Plant Cell Environ. 19, 1149-1159.
Etzler, M.E., G. Kalsi, N.N. Ewing, N.J. Roberts, R.B. Day and J.B. Murphy. 1999. A nod factor binding lectin with apyrase activity from legume roots. Proc. Natl. Acad. Sci. USA. 96, 5856-5861.
Faria, S.M.de, J.M. Sutherland and J.I. Sprent. 1986. A new type of infected cell in root nodules of Andira spp. (Leguminosae). Plant Science. 45, 143-147.
Faria, S.M.de, S.G. McInroy and J.I. Sprent. 1987. The occurrence of infected cells, with persistent infection threads, in legume root nodules. Can. J. Bot. 65, 553-558.
Faria, S.M.de, G.T. Hay and J.I. Sprent. 1988. Entry of rhizobia into roots of Mimosa scabrella Bentham occurs bet-ween epidermal cells. J. Gen. Microbiol. 134, 2291-2296.
Felker, P., P.R. Clark, A.E. Laag and P.F. Pratt. 1981. Salinity tolerance of the tree legumes: mesquite (Prosopis glan-dulosa var. Torreyana, P. velutina and P. articulata), algarrobo (P. chilensis), kiawe (P. pallida) and tamarugo (P. tamarugo) grown in sand culture on nitrogen-free media. Plant Soil. 61, 311-317.
Feng, L., R.J. Roughley and L. Copeland. 2002. Morphological changes of rhizobia in peat cultures. Appl. Environ. Microbiol. 68, 1064-1070.
84
Finan, T.M., S. Weidner, K. Wong, J. Buhrmester, P. Chain, F.J. Vorhölter, I. Hernandez-Lucas, A. Becker, A. Cowie, J. Gouzy, B. Golding and A. Puhler. 2001. The complete sequence of the 1.683-kb pSymB megaplasmid from the N2-fi xing endosymbiont Sinorhizobium meliloti. Proc. Natl. Acad. Sci. USA. 98, 9889-9894.
Fisher, C.E. 1977. Mesquite and modern man in southwestern North America, p. 177-188. In Simpson, B.B., (ed.), Mesquite - Its Biology in Two Desert Scrub Ecosystems. US/IBP Synthesis Series 4. Dowden, Hutchinson & Ross Inc., Stroudsburg, PA.
Fisher, R.F. and S.R. Long. 1992. Rhizobium-plant signal exchange. Nature. 357, 655-660.
Fougere, F. and D. Le Rudulier. 1990a. Glycine betaine biosynthesis and catabolism in bacteroids of Rhizobium meliloti effect of salt stress. J. Gen. Microbiol. 136, 2503-2510.
Fougere, F. and D. Le Rudulier. 1990b. Uptake of glycine betaine and its analogues by bacteroids of Rhizobium meliloti. J. Gen. Microbiol. 136, 157-163.
Fougere, F., D. Le Rudulier and J.G. Streeter. 1991. Effects of salt stress on amino acid, organic acid, and carbohydrate composition of roots, bacteroids, and cytosol of alfalfa (Medicago sativa L.). Plant Physiol. 96, 1228-1236.
Franson, R.L., M.S. Brown and G.J. Bethlenfalvay. 1991. The Glycine-Glomus-Bradyrhizobium symbiosis. XI. Nodule gas exchange and effi ciency as a function of soil and root water status in mycorrhizal soybean. Physiol. Plant. 83, 476-482.
Fred, E.B., I.L. Baldwin and E. McCoy. 1939. Root Nodule Bacteria and Leguminous Plants. The University of Wis-consin Press, Madison. 383 pages.
Galiana, A., Y. Prin, B. Mallet, G.-M. Gnahoua, M. Poitel and H.G. Diem. 1994. Inoculation of Acacia mangium with alginate beads containing selected Bradyrhizobium strains under fi eld conditions: long-term effect on plant growth and persistence of the introduced strains in soil. Appl. Environ. Microbiol. 60, 3974-3980.
Galinski, E.A. and H.G. Truper. 1994. Microbial behaviour in salt-stressed ecosystems. FEMS Microbiol. Rev. 15, 95-108.
Ghittoni, N.E. and M.A. Bueno. 1995. Peanut rhizobia under salt stress: role of trehalose accumulation in strain ATCC 51466. Can. J. Microbiol. 41, 1021-1030.
Giller, K.E. and K.J. Wilson. 1991. Nitrogen fi xation in tropical cropping systems. CAB International, Melksham. 313 pages.
Goddijn, O.J.M. and K. van Dun. 1999. Trehalose metabolism in plants. Trends Plant Sci. 4, 315-319.
Gouffi , K., V. Pichereau, J.-P. Rolland, D. Thomas, T. Bernard and C. Blanco. 1998. Sucrose is a nonaccumulated osmoprotectant in Sinorhizobium meliloti. J. Bacteriol. 180, 5044-5051.
Gouffi , K., N. Pica, V. Pichereau and C. Blanco. 1999. Disaccharides as a new class of nonaccumulated osmoprotec-tants for Sinorhizobium meliloti. Appl. Environ. Microbiol. 65, 1491-1500.
Gouffi , K., T. Bernard and C. Blanco. 2000. Osmoprotection by pipecolic acid in Sinorhizobium meliloti: specifi c effects of D and L isomers. Appl. Environ. Microbiol. 66, 2358-2364.
Guerin, V., J.-C. Trinchant and J. Rigaud. 1990. Nitrogen fi xation (C2H2 reduction) by broad bean (Vicia faba L.) nodules and bacteroids under water-restricted conditions. Plant Physiol. 92, 595-601.
Guerin, V., D. Pladys, J.-C. Trinchant and J. Rigaud. 1991. Proteolysis and nitrogen fi xation in faba-bean Vicia faba nodules under water stress. Physiol. Plant. 82, 360-366.
Gupta, R.J. and I.P. Abrol. 1990. Salt-affected soils: their reclamation and management for crop production. Adv. Soil Sci. 5, 223-288.
Hadri, A.-E. and T. Bisseling. 1998. Responses of the plant to Nod factors, p. 403-416. In Spaink, H.P., A. Kondorosi and P.J.J. Hooykaas, (eds.), The Rhizobiaceae - Molecular Biology of Model Plant Associated Bacteria. Kluwer Academic Publishers, Dordrecht.
Hadri, A.-E., H.P. Spaink, T. Bisseling and N.J. Brewin. 1998. Diversity of root nodulation and rhizobial infection process, p. 347-360. In Spaink, H.P., A. Kondorosi and P.J.J. Hooykaas, (eds.), The Rhizobiaceae - Molecular Biology of Model Plant-Associated Bacteria. Kluwer Academic Publishers, Dordrecht.
Hafeez, F.Y., S. Asad and K.A. Malik. 1991. The effect of high temperature on Vigna radiata nodulation and growth with different bradyrhizobial strains. Environ. Exp. Bot. 31, 285-294.
85
Hansen, A.P. and J.S. Pate. 1987. Comparative growth and symbiotic performance of seedlings of Acacia spp. in defi ned pot culture or as natural understorey components of a eucalypt forest ecosystem in S.W. Australia. J. Exp. Bot. 38, 13-25.
Hasegawa, P.M., R.A. Bressan, J.-K. Zhu and H.J. Bohnert. 2000. Plant cellular and molecular responses to high salinity. Annu. Rev. Plant Physiol. Plant Mol. Biol. 51, 463-499.
Hashem, F.M., D.M. Swelim, L.D. Kuykendall, A.I. Mohamed, S.M. Abdel-Wahab and N.I. Hegazi. 1998. Identifi ca-tion and characterization of salt- and thermo-tolerant Leucaena-nodulating Rhizobium strains. Biol. Fertil. Soils. 27, 335-341.
Hatimi, A. 1999. Effect of salinity on the association between root symbionts and Acacia cyanophylla Lind.: growth and nutrition. Plant Soil. 216, 93-101.
Haukka, K., K. Lindström and J.P.W. Young. 1998. Three phylogenetic groups of nodA and nifH genes in Sinorhizo-bium and Mesorhizobium isolates from leguminous trees growing in Africa and Latin America. Appl. Environ. Microbiol. 64, 419-426.
Heijnen, C.E., S. Page and J.D. van Elsas. 1995. Metabolic activity of Flavobacterium strain P25 during starvation and after introduction into bulk soil and the rhizosphere of wheat. FEMS Microbiol. Ecol. 18, 129-138.
Henderson, I.R., P. Owen and J.P. Nataro. 1999. Molecular switches - the ON and OFF of bacterial phase variation. Mol. Microbiol. 33, 919-932.
Hirsch, A.M. 1992. Transley Review No. 40. Developmental biology of legume nodulation. New Phytol. 122, 211-237.
Hirsch, A.M. 1999. Role of lectins (and rhizobial exopolysaccharides) in legume nodulation. Curr. Opin. Plant Biol. 2, 320-326.
Hocking, D. Ed. 1993. Trees for drylands. International Science Publisher, New York. 370 pages.
Högberg, P. 1986. Nitrogen-fi xation and nutrient relations in savanna woodland trees (Tanzania). J. Appl. Ecol. 23, 675-688.
Hua, S.S.T., V.Y. Tsai, G.M. Lichens and A.T. Noma. 1982. Accumulation of amino acid in Rhizobium sp. strain WR1001 in response to sodium chloride salinity. Appl. Environ. Microbiol. 44, 135-140.
Hungria, M. and A.A. Franco. 1993. Effects of high temperature on nodulation and nitrogen fi xation by Phaseolus vulgaris L. Plant Soil. 149, 95-102.
Hungria, M. and M.A.T. Vargas. 2000. Environmental factors affecting N2 fi xation in grain legumes in the tropics, with an emphasis on Brazil. Field Crops Res. 65, 151-164.
James, E.K., J.I. Sprent, J.M. Sutherland, S.G. McInroy and F.R. Minchin. 1992. The structure of nitrogen fi xing root nodules on the aquatic mimosoid legume Neptunia plena. Annals Bot. 69, 173-180.
Jarvis, B.D.W., P. van Berkum, W.X. Chen, S.M. Nour, M.P. Fernandez, J.C. Cleyet-Marel and M. Gillis. 1997. Transfer of Rhizobium loti, Rhizobium huakuii, Rhizobium ciceri, Rhizobium mediterraneum, and Rhizobium tiansha-nense to Mesorhizobium gen. nov. Int. J. Syst. Bacteriol. 47, 895-898.
Jenkins, M.B., R.A. Virginia and W.M. Jarrel. 1987. Rhizobial ecology of the woody legume mesquite Prosopis glan-dulosa in the Sonoran desert. Appl. Environ. Microbiol. 53, 36-40.
Jenkins, M.B., R.A. Virginia and W.M. Jarrel. 1988a. Depth distribution and seasonal populations of mesquite-nodu-lating rhizobia in warm desert ecosystems. Soil Sci. Soc. Am. J. 52, 1644-1650.
Jenkins, M.B., R.A. Virginia and W.M. Jarrel. 1988b. Rhizobial ecology of the woody legume Psorothamnus spinosus in a Sonoran Desert Arroyo. Plant Soil. 105, 113-120.
Jenkins, M.B., R.A. Virginia and W.M. Jarrel. 1989. Ecology of fast-growing and slow-growing mesquite-nodulating rhizobia in Chihuahuan and Sonoran desert ecosystems. Soil Sci. Soc. Am. J. 53, 543-549.
Jimenez, J. and J. Casadesus. 1989. An altruistic model of the Rhizobium-legume association. J. Hered. 80, 335-337.
Johnson, H.B. and H.S. Mayeux Jr. 1990. Prosopis glandulosa and the nitrogen balance of rangelands: extent and occurrence of nodulation. Oecologia. 84, 176-185.
Kaminski, P.A., J. Batut and P. Boistard. 1998. A survey of symbiotic nitrogen fi xation by rhizobia, p. 431-460. In Spaink, H.P., A. Kondorosi and P.J.J. Hooykas, (eds.), The Rhizobiaceae - Molecular Biology of Model Plant-Associated Bacteria. Kluwer Academic Publishers, Dordrecht.
86
Kannenberg, E.L., B.L. Reuhs, L.S. Forsberg and R.W. Carlson. 1998. Lipopolysaccharides and K-antigenes: their structures, biosynthesis and functions, p. 119-154. In Spaink, H.P., A. Kondorosi and P.J.J. Hooykaas, (eds.), The Rhizobiaceae - Molecular Biology of Model Plant-Associated Bacteria. Kluwer Academic Publisher, Dordrecht.
Karsisto, M. and K. Lindström. 1992. Selection of suitable inoculation methods, microbe strains, and carrier materials for nitrogen fi xing trees in the Sudan (in Finnish). Puuntaimien typensidonta Sudanissa. Loppuraportti Suomen Akatemialle, Helsinki. 66 pages.
Keay, R.W.J. 1989. Trees of Nigeria. Clarendon Press Oxford, Suffolk. 476 pages.
Kell, D.B., A.S. Kaprelyants, D.H. Weichart, C.R. Harwood and M.R. Barer. 1998. Viability and activity in readily culturable bacteria: a review and discussion of the practical issues. Antonie Van Leeuwenhoek. 73, 169-187.
Khbaya, B., M. Neyra, P. Normand, K. Zerhari and A. Filali-Maltouf. 1998. Genetic diversity and phylogeny of rhi-zobia that nodulate Acacia spp. in Morocco assessed by analysis of rRNA genes. Appl. Environ. Microbiol. 64, 4912-4917.
Khurana, E. and J.S. Singh. 2001. Ecology of seed and seedling growth for conservation and restoration of tropical dry forest: a review. Environ. Conserv. 28, 39-52.
Kijne, J.W. 1992. The Rhizobium infection process, p. 349-398. In Stacey, G., R.H. Burris and H.J. Evans, (eds.), Biological Nitrogen Fixation. Chapman and Hall, New York, NY.
Kirmse, R.D. and B.E. Norton. 1984. The potential of Acacia albida for desertifi cation control and increased produc-tivity in Chad. Biol. Conserv. 29, 121-141.
Kishinevsky, B.D., D. Sen and R.W. Weaver. 1992. Effect of high root temperature on Bradyrhizobium-peanut sym-biosis. Plant Soil. 143, 275-282.
Kolter, R., D.A. Siegele and A. Tormo. 1993. The stationary phase of the bacterial life cycle. Annu. Rev. Microbiol. 47, 855-874.
Kozlowski, T.T. and S.G. Pallardy. 1997. Physiology of Woody Plants. Academic Press, San Diego. 411 pages.
Kramer, P.J. 1980. Drought, stress and the origin of adaptations, p. 7-20. In Turner, N.C. and P.J. Kramer, (eds.), Adap-tation of Plants to Water and High Temperature Stress. A Wiley-Interscience Publication, John Wiley & Sons, New York.
Kumar, H., N.K. Arora, V. Kumar and D.K. Maheshwari. 1999. Isolation, characterization, and selection of salt tolerant rhizobia nodulating Acacia catechu and A. nilotica. Symbiosis. 26, 279-288.
Lafay, B. and J.J. Burdon. 1998. Molecular diversity of rhizobia occurring on native shrubby legumes in southeastern Australia. Appl. Environ. Microbiol. 64, 3989-3997.
Lafay, B. and J.J. Burdon. 2001. Small-subunit tRNA genotyping of rhizobia nodulating Australian Acacia spp. Appl. Environ. Microbiol. 67, 396-402.
Lakshmi-Kumari, M., C.S. Singh and N.S. Subba Rao. 1974. Root hair infection and nodulation in lucerne (Medicago sativa L.) as infl uenced by salinity and alkalinity. Plant Soil. 40, 261-268.
Lal, B. and S. Khanna. 1996. Long term fi eld study shows increased biomass production in tree legumes inoculated with Rhizobium. Plant Soil. 184, 111-116.
Lal, R. 2000. Physical management of soils of the tropics: priorities for the 21st century. Soil Sci. 165, 191-207.
Langkamp, P.J., L.B. Swinden and M.J. Dalling. 1979. Nitrogen fi xation (acetylene reduction) by Acacia pellita on areas restored after mining at Groote Eylandt, Northern Territory. Aust. J. Bot. 27, 353-361.
Langkamp, P.J., G.K. Farnell and M.J. Dalling. 1982. Nutrient cycling in a stand of Acacia holosericea A. Cunn. ex G. Don. I. Measurements of precipitation interception, seasonal acetylene reduction, plant growth and nitrogen requirement. Aust. J. Bot. 30, 87-106.
Lawrie, A.C. 1983. Relationships among rhizobia from native Australian legumes. Appl. Environ. Microbiol. 45, 1822-1828.
Letouzey, A. 1982. Manuel de Botanique Forestiere: Afrique Tropicale. Centre technique forestier tropical. 864 pages.
Lindström, K. 1984a. Analysis of factors affecting in situ nitrogenase (C2H2) activity of Galega orientalis, Trifolium pratense and Medicago sativa in temperate conditions. Plant Soil. 79, 329-341.
Lindström, K. 1984b. Effect of various Rhizobium trifolii strains in nitrogenase (C2H2) activity profi les of red clover (Trifolum pratense cv. Venla. Plant Soil. 80, 79-89.
87
Lippi, D., M.R. De Paolis, E. Di Mattia, S. Grego, T. Pietrosanti and I. Cacciari. 2000. Effect of salinity on growth and starvation-survival of a tropical Rhizobium strain. Biol. Fertil. Soils. 30, 276-283.
Lipsanen, P. and K. Lindström. 1988. Infection and root nodule structure in the Rhizobium galegae sp. nov. -Galega symbiosis. Symbiosis. 6, 81-96.
Lloret, J., B.B.H. Wulff, J.M. Rubio, J.A. Downie, I. Bonilla and R. Rivilla. 1998. Exopolysaccharide II production is regulated by salt in the halotolerant strain Rhizobium meliloti EFB1. Appl. Environ. Microbiol. 64, 1024-1028.
Lloret, J., M. Martin, R.I. Oruezabal, I. Bonilla and R. Rivilla. 2002. MucR and MucS activate exp genes transcription and galactoglucan production in Sinorhizobium meliloti EFB1. Mol. Plant-Microbe Interact. 15, 54-59.
Lortet, G., N. Mear, J. Lorquin, B. Dreyfus, P. de Lajudie, C. Rosenberg and C. Boivin. 1996. Nod factor thin-layer chromatography profi ling as a tool to characterize symbiotic specifi city of rhizobial strains: application to Sinor-hizobium saheli, S. teranga, and Rhizobium sp. strains isolated from Acacia and Sesbania. Mol. Plant-Microbe Interact. 9, 736-747.
Lowder, M., A. Unge, N. Maraha, J.K. Jansson, J. Swigget and J.D. Oliver. 2000. Effect of starvation and the viable-but-nonculturable state on green fl uorescent protein (GFP) fl uorescence in GFP-tagged Pseudomonas fl uorescens A506. Appl. Environ. Microbiol. 66, 3160-3165.
Lupwayi, N.Z., P.E. Olsen, E.S. Sande, H.H. Keyser, M.M. Collins, P.W. Singleton and W.A. Rice. 2000. Inoculant quality and its evaluation. Field Crops Res. 65, 259-270.
Mäkelä, P., P. Peltonen-Sainio, K. Jokinen, E. Pehu, H. Setälä, R. Hinkkanen and S. Somersalo. 1996. Uptake and translocation of foliar-applied glycinebetaine in crop plants. Plant Science. 121, 221-230.
Maniatis, T., E.F. Fritsch and J. Sambrook. 1982. Molecular Cloning: A Laboratory Manual. Cold Spring Harbor Laboratory, Cold Spring Harbor, New York. N pages.
Marcar, N.E., P. Dart and C. Sweeney. 1991. Effect of root-zone salinity on growth and chemical composition of Acacia ampliceps B. R. Maslin, A. auriculiformis A. Cunn ex Benth. and A. mangium Willd. at two nitrogen levels. New Phytol. 119, 567-573.
Marsudi, N.D.S., A.R. Glenn and M.J. Dilworth. 1999. Identifi cation and characterization of fast- and slow-growing root nodule bacteria from South-Western Australian soils able to nodulate Acacia saligna. Soil. Biol. Biochem. 31, 1229-1238.
Maslin, B.R. and C.H. Stirton. 1997. Generic and infrageneric classifi cation in Acacia (Leguminosae: Mimosoideae): a list of critical species on which to build a comparative data set. Bull. Internat. Group for the Study of Mimoso-ideae. 20, 22-44.
Mateos, P.F., D.L. Baker, M. Petersen, E. Velazquez, J.I. Jimenez-Zurdo, E. Martinez-Molina, A. Squartini, G. Orgam-bide, D.H. Hubbell and F.B. Dazzo. 2001. Erosion of root epidermal cell walls by Rhizobium polysaccharide-degrading enzymes as related to primary host infection in the Rhizobium-legume symbiosis. Can. J. Microbiol. 47, 475-487.
Matthysse, A.G. and J.W. Kijne. 1998. Attachment of Rhizobiaceae to plant cells, p. 235-249. In Spaink, H.P., A. Kondorosi and P.J.J. Hooykaas, (eds.), The Rhizobiaceae - Molecular Biology of Model Plant-Associated Bacteria. Kluwer Academic Publishers, Dordrecht.
Maydell, H.-J.von. 1990. Trees and Shrubs of the Sahel: Their Characteristics and Uses. Verlag Josef Margraf, Weiker-sheim. 525 pages.
McDougald, D., S.A. Rice, D. Weichart and S. Kjelleberg. 1998. Nonculturability: adaptation or debilitation? FEMS Microbiol. Ecol. 25, 1-9.
McInroy, S.G., C.D. Cambell, K.E. Haukka, D.W. Odee, J.I. Sprent, W.-J. Wang, J.P.W. Young and J.M. Sutherland. 1999. Characterization of rhizobia from African acacias and other tropical woody legumes using BiologTM and partial 16S rRNA sequencing. FEMS Microbiol. Lett. 170, 111-117.
McNeil, S.D., M.L. Nuccio and A.D. Hanson. 1999. Betaines and related osmoprotectants. Targets for metabolic engineering of stress resistance. Plant Physiol. 120, 945-949.
Mellor, R.B. and D. Werner. 1990. Legume nodule biochemistry and function, p. 111-129. In Gresshoff, M., (ed.), The Molecular Biology of Symbiotic Nitrogen Fixation. CRC Press, Boca Raton, Florida.
Michelsen, A. and S. Rosendahl. 1990. The effect of VA mycorrhizal fungi, phosphorus and drought stress on the growth of Acacia nilotica and Leucaena leucocephala seedlings. Plant Soil. 124, 7-13.
88
Michiels, J., C. Verreth and J. Vanderleyden. 1994. Effects of temperature stress on bean-nodulating Rhizobium strains. Appl. Environ. Microbiol. 60, 1206-1212.
Miller, K.J. and J.M. Wood. 1996. Osmoadaptation by rhizosphere bacteria. Annu. Rev. Microbiol. 50, 101-136.
Milnitsky, F., L. Frioni and F. Agius. 1997. Characterization of rhizobia that nodulate native legume trees from Uru-guay. Soil. Biol. Biochem. 29, 989-992.
Möller, A. and J.K. Jansson. 1998. Detection of fi refl y luciferase- tagged bacteria in environmental samples, p. 269-283. In LaRossa, R.A., (ed.), Methods in Molecular Biology: Bioluminescence Methods and Protocols. Humana Press Inc., Totowa, NJ.
Monk, D., J.S. Pate and W.A. Loneragan. 1981. Biology of Acacia pulchella R.Br. with special reference to symbiotic nitrogen fi xation. Aust. J. Bot. 29, 579-592.
Montagu, K.D. and K.C. Woo. 1999. Recovery of tree photosynthetic capacity from seasonal drought in the wet-dry tropics: the role of phyllode and canopy processes in Acacia auriculiformis. Aust. J. Plant Physiol. 26, 135-145.
Moreira, F.M.S., K. Haukka and J.P.W. Young. 1998. Biodiversity of rhizobia isolated from a wide range of forest legumes in Brazil. Mol. Ecol. 7, 889-895.
Moulin, L., A. Munive, B. Dreyfus and C. Boivin-Masson. 2001. Nodulation of legumes by members of the the beta-subclass of Proteobacteria. Nature. 411, 948-950.
Munchbach, M., A. Nocker and F. Narberhaus. 1999. Multiple small heat shock proteins in rhizobia. J. Bacteriol. 181, 83-90.
Munro, P.M., G.N. Flatau, R.L. Clement and M.J. Gauthier. 1995. Infl uence of the RpoS (KatF) sigma factor on main-tenance of viability and culturability of Escherichia coli and Salmonella typhimurium in seawater. Appl. Environ. Microbiol. 61, 1853-1858.
NAS. 1979. Tropical legumes: Recources for the future. National Academy of Sciences, Washington, D.C. 331 pages.
Nativ, R., J.E. Ephrath, P.R. Berliner and Y. Saranga. 1999. Drought resistance and water use effi ciency in Acacia saligna. Aust. J. Bot. 47, 577-586.
Ndoye, I., F. de Billy, J. Vasse, B. Dreyfus and G. Truchet. 1994. Root nodulation of Sesbania rostrata. J. Bacteriol. 176, 1060-1068.
Newcomb, W. and L. McIntyre. 1981. Development of root nodules of mung bean (Vigna radiata): a reinvestigation of endocytosis. Can. J. Bot. 59, 2478-2499.
Nick, G. (1998): Polyphasic taxonomy of rhizobia isolated from tropical tree legumes. Thesis, University of Helsinki, Finland.
Nick, G., M. Jussila, B. Hoste, R.M. Niemi, S. Kaijalainen, P. de Lajudie, M. Gillis, F.J. de Bruijn and K. Lindström. 1999a. Rhizobia isolated from root nodules of tropical leguminous trees characterized using DNA-DNA dot-blot hybridisation and rep-PCR genomic fi ngerprinting. Syst. Appl. Microbiol. 22, 287-299.
Nick, G., P. de Lajudie, B.D. Eardly, S. Suomalainen, L. Paulin, X. Zhang, M. Gillis and K. Lindström. 1999b. Sinor-hizobium arboris sp. nov. and Sinorhizobium kostiense sp. nov., isolated from leguminous trees in Sudan and Kenya. Int. J. Syst. Bacteriol. 49, 1359-1368.
Nilsen, E.T., M.R. Sharifi , P.W. Rundel, W.M. Jarrel and R.A. Virginia. 1983. Diurnal and seasonal water relations of the desert phreatophyte Prosopis glandulosa (honey mesquite) in the Sonoran Desert of California. Ecology. 64, 1381-1393.
Nilsen, E.T., R.A. Virginia and W.M. Jarrel. 1986. Water relations and growth characteristic of Prosopis glandulosa var. torreyana in a simulated phreatophytic environment. Amer. J. Bot. 73, 427-433.
Nocker, A., N.-P. Krstulovic, X. Perret and F. Narberhaus. 2001. ROSE elements occur in disparate rhizobia and are funtionally interchangeable between species. Arch. Microbiol. 176, 44-51.
Nogales, J., R. Campos, H. BenAbdelkhalek, J. Olivares, C. Lluch and J. Sanjuan. 2002. Rhizobium tropici genes involved in free-living salt tolerance are required for the establishment of effi cient nitrogen-fi xing symbiosis with Phaseolus vulgaris. Mol. Plant-Microbe Interact. 15, 225-232.
Norris, D.O. 1956. Legumes and the Rhizobium symbiosis. Emp. J. Exp. Agric. 24, 247-270.
Odee, D.W., J.M. Sutherland, J.M. Kimiti and J.I. Sprent. 1995. Natural rhizobial populations and nodulation status of woody legumes growing in diverse Kenyan conditions. Plant Soil. 173, 211-224.
89
Odee, D.W., J.M. Sutherland, E.T. Makatiani, S.G. McInroy and J.I. Sprent. 1997. Phenotypic characteristics and composition of rhizobia associated with woody legumes growing in diverse Kenyan conditions. Plant Soil. 188, 65-75.
Odee, D.W., K. Haukka, S.G. McInroy, J.I. Sprent, J.M. Sutherland and J.P.W. Young. 2002a. Genetic and symbiotic characterization of rhizobia isolated from tree and herbaceous legumes grown in soils from ecologically diverse sites in Kenya. Soil. Biol. Biochem. 34, 801-811.
Odee, D.W., S.A. Indieka and D. Lesueur. 2002b. Evaluation of inoculation procedures for Calliandra calothyrsus Meisn. grown in tree nurseries. Biol. Fertil. Soils. 36, 124-128.
Olivares, J., M.A. Herrera and E.J. Bedmar. 1988. Woody legumes in arid and semi-arid zones: the Rhizobium-Prosopis chilensis symbiosis, p. 65-72. In Beck, D.P. and L.A. Materon, (eds.), Nitrogen Fixation by Legumes in Mediterranean Agriculture. ICARDA, Netherlands.
Olivieri, I. and S.A. Frank. 1994. The evolution of nodulation in Rhizobium: altruism in the rhizosphere. J. Hered. 85, 46-47.
Otieno, D.O., J.I. Kinyamario and T.O. Omenda. 2001. Growth features of Acacia tortilis and Acacia xanthophloea seedlings and their response to cyclic soil drought stress. Afr. J. Ecol. 39, 126-132.
Paau, A.S., C.B. Bloch and W.J. Brill. 1980. Developmental fate of Rhizobium meliloti bacteroids in alfalfa nodules. J. Bacteriol. 143, 1480-1490.
Pankhurst, C.E. and A.S. Craig. 1978. Effect of oxygen concentration, temperature and combined nitrogen on the mor-phology and nitrogenase activity of Rhizobium sp. strain 32H1 in agar culture. J. Gen. Microbiol. 106, 207-219.
Pankhurst, C.E. and A.H. Gibson. 1973. Rhizobium strain infl uence on disruption of clover nodule development at high root temperature. J. Gen. Microbiol. 74, 219-231.
Pararajasingham, S. and D.P. Knievel. 1990. Nitrogenase activity of cowpea (Vigna unguiculata (L.) Walp.) during and after drought stress. Can. J. Plant. Sci. 70, 163-171.
Pate, J.S., M.J. Unkovich, P.D. Erskine and G.R. Stewart. 1998. Australian mulga ecosystems - 13C and 15N natural abundances of biota components and their ecophysiological signifi cance. Plant Cell Environ. 21, 1231-1242.
Peoples, M.B. and D.F. Herridge. 1990. Nitrogen fi xation by legumes in tropical and subtropical agriculture. Adv. Agron. 44, 155-223.
Phillips, D.A. 1999. Using rhizobia in the 21st century. Symbiosis. 27, 83-102.
Pichereau, V., J.-A. Pocard, J. Hamelin, C. Blanco and T. Bernard. 1998. Differential effects of dimethylsulfoniopro-pionate, dimethylsulfonioacetate, and other S-methylated compounds on the growth of Sinorhizobium meliloti at low and high osmolarities. Appl. Environ. Microbiol. 64, 1420-1429.
Pocard, J.-A., T. Bernard and D. Le Rudulier. 1991. Translocation and metabolism of glycine betaine in nodulated alfalfa plants subjected to salt stress. Physiol. Plant. 81, 95-102.
Poolman, B. and E. Glaasker. 1998. Regulation of compatible solute accumulation in bacteria. Mol. Microbiol. 29, 397-407.
Potts, M. 1994. Desiccation tolerance of prokaryotes. Microbiol. Rev. 58, 755-805.
Pueppke, G.S. and W.J. Broughton. 1999. Rhizobium sp. strain NGR234 and R. fredii USDA257 share exceptionally broad, nested host ranges. Mol. Plant-Microbe Interact. 12, 293-318.
Purwantari, N.D., R.A. Date and P.J. Dart. 1995. Nodulation and N2 by Calliandra calothyrsus and Sesbania sesban grown at different root temperatures. Soil. Biol. Biochem. 27, 421-425.
Räsänen, L.A. and K. Lindström. 1997. Stability of short and long O-chain lipopolysaccharide types in Rhizobium galegae and their correlation with symbiotic properties and growth conditions, tolerance of low pH, aluminum and salt in the growth medium. FEMS Microbiol. Lett. 155, 17-22.
Räsänen, L.A., U. Heikkilä-Kallio, L. Suominen, P. Lipsanen and K. Lindström. 1991. Expression of Rhizobium gale-gae common nod clones in various backgrounds. Mol. Plant-Microbe Interact. 6, 535-544.
Rajakylä, E. and M. Paloposki. 1983. Determination of sugars (and betaine) in molasses by high-performance liquid chromatography. Comparison of the results with those obtained by the classical lane-eynon method. J. Chroma-tography. 282, 595-602.
Rao, A.V. and B. Venkateswarlu. 1983. Microbial ecology of the soils of Indian desert. Agriculture, Ecosystems and Environment. 10, 361-369.
90
Reddel, P. 1993. Soil constraints to the growth of nitrogen-fi xing trees in tropical environments, p. 65-83. In Subba Rao, N.S. and C. Rodriguez-Barrueco, (eds.), Symbioses in nitrogen-fi xing trees. International Science Publisher, New York.
Requena, N., E. Perez-Solis, C. Azcon-Aguilar, P. Jeffries and J.M. Barea. 2001. Management of indigenous plant-microbe symbioses aids restoration of desertifi ed ecosystems. Appl. Environ. Microbiol. 67, 495-498.
Rhodes, D. and A.D. Hanson. 1993. Quaternary ammonium and tertiary sulfonium compounds in higher plants. Annu. Rev. Plant Physiol. Plant Mol. Biol. 44, 357-384.
Rivas, R., E. Velazquez, A. Willems, N. Vizcaino, N.S. Subba-Rao, P.F. Mateos, M. Gillis, F.B. Dazzo and E. Martinez-Molina. 2002. A new species of Devosia that forms a unique nitrogen-fi xing root-nodule symbiosis with the aquatic legume Neptunia natans (L.f.) Druce. Appl. Environ. Microbiol. 68, 5217-5222.
Roberson, E.B. and M.K. Firestone. 1992. Relationship between desiccation and exopolysaccharide production in a soil Pseudomonas sp. Appl. Environ. Microbiol. 58, 1284-1291.
Roberts, N.J., J. Brigham, B. Wu, J.B. Murphy, H. Volpin, D.A. Phillips and M.E. Etzler. 1999. A nod factor-binding lectin is a member of a distinct class of apyrases that may be unique to the legumes. Mol. Gen. Genet. 262, 261-267.
Robertson, J.G. and K.J.F. Farnden. 1980. Ultrastructure and metabolism of the developing legume root nodule, p. 65-112. In Mifl in, B.J., (ed.), The Biochemistry of Plants. Amino acids and Derivates. Academic Press, Inc., .
Robinson, J. and S.A. Harris. 2000. A plastid DNA phylogeny of the genus Acacia Miller (Acacieae, Leguminoseae). Bot. J. Linn. Soc. 132, 195-222.
Rosendahl, L., P. Mouritzen and A. Rudbeck. 2001. Nitrogen transfer in the interface between the symbionts in pea root nodules. Plant Soil. 230, 31-37.
Roshetko, J.M. Ed. 2001. Agroforestry species and technologies: a compilation of the highlights and factsheets pub-lished by NFTA and FACT Net 1895-1999. A Publication of Winrock International, Morrilton, Arkansas. 232 pages.
Roth, B.L., M. Poot, S.T. Yue and P.J. Millard. 1997. Bacterial viability and antibiotic susceptibility testing with SYTOX Green nucleic acid stain. Appl. Environ. Microbiol. 63, 2421-2431.
Roth, L.E. and G. Stacey. 1989. Bacterium release into host cells of nitrogen-fi xing soybean nodules: the symbiosome membrane comes from three sources. Eur. J. Cell Biol. 49, 13-23.
Ruiz-Lozano, J.M., C. Collados, J.M. Barea and R. Azcon. 2001. Arbuscular mycorrhizal symbiosis can alleviate drought-induced nodule senescence in soybean plants. New Phytol. 151, 493-502.
Rundel, P.W., E.T. Nilsen, M.R. Sharifi , R.A. Virginia, W.M. Jarrel, D.H. Kohl and G.B. Shearer. 1982. Seasonal dyna-mics of nitrogen cycling for a Prosopis woodland in the Sonoran Desert. Plant Soil. 67, 343-353.
Sanchez, P.A. and T.J. Logan. 1992. Myths and science about the chemistry and fertility of soils in the tropics, p. 35-46. In Lal, R. and P.A. Sanchez, (eds.), Myths and Science of Soils of the Tropics. SSSA Special Publication No 29, Madison.
Sanginga, N., G.D. Bowen and S.K.A. Danso. 1990. Assessment of genetic variability for N2 fi xation and within pro-venances of Leucaena leococephala and Acacia albida estimated by 15N labelling techniques. Plant Soil. 127, 169-178.
Santos, R., D. Herouart, S. Sigaud, D. Touati and A. Puppo. 2001. Oxidative burst in alfalfa-Sinorhizobium meliloti symbiotic interaction. Mol. Plant-Microbe Interact. 14, 86-89.
Sauvage, D., J. Hamelin and F. Larher. 1983. Glycine betaine and other structurally related compounds improve the salt tolerance of Rhizobium meliloti. Plant Sci. Lett. 31, 291-302.
Schnider-Keel, U., K. Bang Lejbolle, E. Baehler, D. Haas and C. Keel. 2001. The sigma factor AlgU (AlgT) controls exopolysaccharide production and tolerance towards desiccation and osmotic stress in the biocontrol agent Pseu-domonas fl uorescens CHA0. Appl. Environ. Microbiol. 67, 5683-5693.
Schulze, E.-D., G. Gebauer, H. Ziegler and O.L. Lange. 1991. Estimates of nitrogen fi xation by trees on an aridity gradient in Namibia. Oecologia. 88, 451-455.
Serraj, R., T.R. Sinclair and L.H. Allen. 1998. Soybean nodulation and N2 fi xation response to drought under carbon dioxide enrichment. Plant Cell Environ. 21, 491-500.
Shannon, M.C. 1997. Adaptation of plants to salinity. Adv. Agron. 60, 75-120.
91
Shearer, G., D.H. Kohl, R.A. Virginia, B.A. Bryan, J.L. Skeeters, E.T. Nilsen, M.R. Sharifi and P.W. Rundel. 1983. Estimates of N-fi xation from variation in the natural abundance of 15N in Sonoran Desert ecosystems. Oecologia. 56, 365-373.
Singleton, P.W. and B.B. Bohlool. 1984. Effect of salinity on nodule formation by soybean. Plant Physiol. 74, 72-76.
Sleator, R.D. and C. Hill. 2001. Bacterial osmoadaptation: the role of osmolytes in bacterial stress and virulence. FEMS Microbiol. Rev. 26, 49-71.
Smith, J. and K. Smith. 1992. The ABCs of MPTs: The multiple uses of leguminous trees. Ceres. 133, 38-43.
Smith, L.T. and G.M. Smith. 1989. An osmoregulated dipeptide in stressed Rhizobium meliloti. J. Bacteriol. 171, 4714-4717.
Smith, L.T., J.-A. Pocard, T. Bernard and D. Le Rudulier. 1988. Osmotic control of glycine betaine biosynthesis and degradation in Rhizobium meliloti. J. Bacteriol. 170, 3142-3149.
Smith, L.T., G.M. Smith and M.A. Madkour. 1990. Osmoregulation in Agrobacterium tumefaciens: accumulation of a novel disaccharide is controlled by osmotic strength and glycine betaine. J. Bacteriol. 172, 6849-6955.
Smith, L.T., G.M. Smith, M.R. D’souza, J.-A. Pocard, D. Le Rudulier and M.A. Madkour. 1994. Osmoregulation in Rhizobium meliloti: mechanism and control by other environmental signals. J. Exp. Zool. 268, 162-165.
Soil Survey Staff. 1975. Soil taxonomy: a basic system of soil classifi cation for making and interpreting soil surveys. USDA Handb. 436. U.S. Gov. Print. Offi ce, Washington, DC. pages.
Sprent, J.I. 1984. Effects of drought and salinity on heterotrophic nitrogen fi xing bacteria and on infection of legumes by rhizobia, p. 295-302. In Veeger, C. and W.E. Newton, (eds.), Advances in Nitrogen Fixation Research. Procee-dings of the 5th International Symposium on Nitrogen Fixation, Noordwijkerhout, The Netherlands, August 28 - September 3, 1983. Martinus Nijhof / Dr. W. Junk Publishers, Wageningen.
Sprent, J.I. 2001. Nodulation in legumes. Royal Botanic Gardens, Kew, Great Britain. 146 pages.
Sprent, J.I. and S.M.de Faria. 1988. Mechanisms of infection of plants by nitrogen fi xing organisms. Plant Soil. 110, 157-165.
Sprent, J.I. and P. Sprent. 1990. Nitrogen Fixing Organisms. Pure and Applied Aspects. Chapman and Hall, Cam-bridge. 256 pages.
Sprent, J.I., J.M. Sutherland and S.M.de Faria. 1989. Structure and function of root nodules from woody legumes, p. 559-578. In Stirton, C.H. and J.L. Zarucchi, (eds.), Advances in Legume Biology. Monogr. Syst. Bot. Missouri Bot. Gard., Missouri.
Subba-Rao, N.S., P.F. Mateos, D. Baker, H.S. Pankratz, J. Palma and F.B. Dazzo. 1995. The unique root-nodule sym-biosis between Rhizobium and the aquatic legume, Neptunia natans (L. f.) Druce. Planta. 196, 311-320.
Surange, S., A.G. Wollum II, N. Kumar and C.S. Nautiyal. 1997. Characterization of Rhizobium from root nodules of leguminous trees growing in alkaline soils. Can. J. Microbiol. 43, 891-894.
Sutherland, J.M. and J.I. Sprent. 1993. Nitrogen fi xation by legume trees, p. 33-63. In Subba Rao, N.S. and C. Rodriguez-Barrueco, (eds.), Symbiosis in Nitrogen-fi xing Trees. International Science Publisher, New York.
Sutherland, J.M., D.W. Odee, G.M. Muluvi, S.G. McInroy and A. Patel. 2000. Single and multi-strain rhizobial ino-culation of African acacias in nursery conditions. Soil. Biol. Biochem. 32, 323-333.
Swift, S., J.A. Downie, N.A. Whitehead, A.M.L. Barnard, G.P.C. Salmond and P. Williams. 2001. Quorum sensing as a population-density-dependent determinant of bacterial physiology. Adv. Microb. Physiol. 45, 199-270.
Sy, A., E. Giraud, P. Jourand, N. Garcia, A. Willems, P. de Lajudie, Y. Prin, M. Neyra, M. Gillis, C. Boivin-Masson and B. Dreyfus. 2001. Methylotrophic Methylobacterium bacteria nodulate and fi x nitrogen in symbioses with legumes. J. Bacteriol. 183, 214-220.
Taiz, L. and E. Zeiger. 1998. Plant Physiology. Sinauer Associates, Inc., Publishers, Massachusetts. 792 pages.
Talibart, R., M. Jebbar, G. Gousbet, S. Himdi-Kabbab, H. Wroblewski, C. Blanco and T. Bernard. 1994. Osmoadap-tation in rhizobia: ectoine-induced salt tolerance. J. Bacteriol. 176, 5210-5217.
Talibart, R., M. Jebbar, K. Gouffi , V. Pichereau, G. Gouesbet, C. Blanco, T. Bernard and J.-A. Pocard. 1997. Transient accumulation of glycine betaine and dynamics of endogenous osmolytes in salt-stressed cultures of Sinorhizo-bium meliloti. Appl. Environ. Microbiol. 63, 4657-4663.
Tate, R., E.J. Patriarca, A. Riccio, R. Defez and M. Iaccarino. 1994. Development of Phaseolus vulgaris root nodules. Mol. Plant-Microbe Interact. 7, 582-589.
92
Thomas, P.M., K.F. Golly, J.W. Zyskind and R.A. Virginia. 1994. Variation of clonal, mesquite-associated rhizobial and bradyrhizobial populations from surface and deep soils by symbiotic gene region restriction fragment length polymorphism and plasmid profi le analysis. Appl. Environ. Microbiol. 60, 1146-1153.
Thrall, P.H., J.J. Burdon and M.J. Woods. 2000. Variation in the effectiveness of symbiotic associations between native rhizobia and temperate Australian legumes: interactions within and between genera. J. Appl. Ecol. 37, 52-65.
Timmers, A.C.J., E. Soupene, M.-C. Auriac, F. de Billy, J. Vasse, P. Boistard and G. Truchet. 2000. Saprophytic intra-cellular rhizobia in alfalfa nodules. Mol. Plant-Microbe Interact. 13, 1204-1213.
Tsuji, T., Y. Kawasaki, S. Takeshima, T. Sekiya and S. Tanaka. 1995. A new fl uorescence staining assay for visualizing living microorganisms in soil. Appl. Environ. Microbiol. 61, 3415-3421.
Tu, J.C. 1981. Effects of salinity on Rhizobium-root-hair interaction, nodulation and growth of soybean. Can. J. Plant. Sci. 61, 231-239.
Turgeon, B.G. and W.D. Bauer. 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.
Turgeon, B.G. and W.D. Bauer. 1985. Ultrastructure of infection-thread development during the infection of soybean by Rhizobium japonicum. Planta. 163, 328-349.
Turk, D., H.H. Keyser and P.W. Singleton. 1993. Response of tree legumes to rhizobial inoculation in relation to the population density of indigenous rhizobia. Soil. Biol. Biochem. 25, 75-81.
Turner, N.C. 1986. Crop water defi cits: a decade of progress. Adv. Agron. 39, 1-51.
Turner, N.C., G.C. Wright and K.H.M. Siddique. 2000. Adaptation of grain legumes (pulses) to water-limited environ-ments. Adv. Agron. 71, 193-231.
Tyerman, S.D., L.F. Whitehead and D.A. Day. 1995. A channel-like transporter for NH4+ on the symbiotic interface of
N2-fi xing plants. Nature. 378, 629-632.
Ullmann, I. 1989. Stomatal conductance and transpiration of Acacia under fi eld conditions: similarities and differen-ces between leaves and phyllodes. Trees. 3, 45-56.
Unkovich, M.J. and J.S. Pate. 2000. An appraisal of recent fi eld measurements of symbiotic N2 fi xation by annual legumes. Field Crops Res. 65, 211-226.
Urban, J.E. and F.B. Dazzo. 1982. Succinate-induced morphology of Rhizobium trifolii 0403 resembles that of bac-teroids in clover nodules. Appl. Environ. Microbiol. 44, 219-226.
Vance, C.P. 1998. Legume symbiotic nitrogen fi xation: agronomic aspects, p. 509-530. In Spaink, H.P., A. Kondorosi and P.J.J. Hooykaas, (eds.), The Rhizobiaceae-Molecular Biology of Model Plant-Associated Bacteria. Kluwer Academic Publishers, Dordrecht.
Vance, C.P., L.E.B. Johnson, A.M. Halvorsen, G.H. Heichel and D.K. Barnes. 1980. Histological and ultrastructural observations of Medicago sativa root nodule senescence after foliage removal. Can. J. Bot. 58, 295-309.
Vance, C.P., L.E.B. Johnson, S. Stade and R.G. Groat. 1982. Birdsfoot trefoil (Lotus corniculatus) root nodules: mor-phogenesis and the effect of forage harvest on structure and function. Can. J. Bot. 60, 505-518.
Vandenbosch, K.A., K.D. Noel, Y. Kaneko and E.H. Newcomb. 1985. Nodule initiation elicited by noninfective mutants of Rhizobium phaseoli. J. Bacteriol. 162, 950-959.
Vassal, J. 1972. Apport des recherches ontogeniques et seminologiques a l’etude morpholoqique, taxonomique et phylogenique du genre Acacia. Bull. Soc. Hist. Nat. Toulose. 108, 105-247.
Vasse, J.M. and G.L. Truchet. 1984. The Rhizobium-legume symbiosis: Observation of root infection by bright-fi eld microscopy after staining with methylene blue. Planta. 161, 487-489.
Vasse, J., F. de Billy, S. Camut and G. Truchet. 1990. Correlation between ultrastructural differentation of bacteroids and nitrogen fi xation in alfalfa nodules. J. Bacteriol. 172, 4295-4306.
Vasse, J., F. de Billy and G. Truchet. 1993. Abortion of infection during the Rhizobium meliloti-alfalfa symbiotic interaction is accompanied by a hypersensitive reaction. Plant J. 4, 555-566.
Vega-Hernandez, M.C., R. Perez-Galdona, F.B. Dazzo, A. Jarobo-Lorenzo, M.C. Alfayate and M. Leon-Barrios. 2001. Novel infection process in the indeterminate root nodule symbiosis between Chamaecytisus proliferus (tagasaste) and Bradyrhizobium. New Phytol. 150, 707-721.
Velazquez, E., J.M. Igual, A. Willems, M.P. Fernandez, E. Munoz, P.F. Mateos, A. Abril, N. Toro, P. Normand, E. Cervantes, M. Gillis and E. Martinez-Molina. 2001. Mesorhizobium chacoense sp. nov., a novel species that nodulates Prosopis alba in the Chaco Arido region (Argentina). Int. J. Syst. Evol. Microbiol. 51, 1011-1021.
93
Verma, D.P.S. and Z. Hong. 1996. Biogenesis of the peribacteroid membrane in root nodules. Trends in Microbiology. 4, 364-368.
Virginia, R.A. and W.M. Jarrel. 1983. Soil properties in a mesquite-dominated Sonoran desert ecosystem. Soil Sci. Soc. Am. J. 47, 138-144.
Virginia, R.A., L.M. Baird, J.S. La Favre, W.M. Jarrell, B.A. Bryan and G. Shearer. 1984. Nitrogen fi xation effi ciency, natural 15N abundance, and morphology of mesquite (Prosopis glandulosa) root nodules. Plant Soil. 79, 273-284.
Virginia, R.A., M.B. Jenkins and W.M. Jarrel. 1986. Depth of root symbiont occurrence in soil. Biol. Fertil. Soils. 2, 127-130.
Waldon, H.B., M.B. Jenkins, R.A. Virginia and E.E. Harding. 1989. Characteristics of woodland rhizobial populations from surface- and deep-soil environments of the Sonoran Desert. Appl. Environ. Microbiol. 55, 3058-3064.
Walsh, K.B. 1995. Physiology of the legume nodule and its response to stress. Soil. Biol. Biochem. 27, 637-655.
Wang, E.T. and E. Martinez-Romero. 2000. Phylogeny of root- and stem-nodule bacteria associated with legumes, p. 237-250. In Triplett, E.W., (ed.), Prokaryotic Nitrogen Fixation. A modell system for the analysis of a biological process. Horizon Scientifi c Press, Wymondham, UK.
Welsh, D.T. 2000. Ecological signifi cance of compatible solute accumulation by micro-organisms: from single cells to global climate. FEMS Microbiol. Rev. 24, 263-290.
Whitehead, N.A., A.M.L. Barnard, H. Slater and N.J.L. Simpson. 2001. Quorum-sensing in Gram-negative bacteria. FEMS Microbiol. Rev. 25, 365-404.
Williams, P.M. and M.S. Sicardi de Mallorca. 1984. Effect of osmotically induced leaf moisture stress on nodulation and nitrogenase activity of Glycine max. Plant Soil. 80, 267-283.
Wilson, K.J., A. Sessitsch, J.C. Corbo, K.E. Giller, A.D.L. Akkermans and R.A. Jefferson. 1995. Glucuronidase (GUS) transposons for ecological and genetic studies of rhizobia and other Gram-negative bacteria. Microbiol. 141, 1691-1705.
Winding, A., S.J. Binnerup and J. Sorensen. 1994. Viability of indigenous soil bacteria assayed by respiratory activity and growth. Appl. Environ. Microbiol. 60, 2869-2875.
Wood, S.M. and W. Newcomb. 1989. Nodule morphogenesis: the early infection of alfalfa (Medicago sativa) root hairs by Rhizobium meliloti. Can. J. Bot. 67, 3108-3122.
Worral, V.S. and R.J. Roughley. 1976. The effect of moisture stress on infection of Trifolium subterraneum L. by Rhi-zobium trifolii Dang. J. Exp. Bot. 27, 1233-1241.
Xing, W. and C.B. Rajashekar. 1999. Alleviation of water stress in beans by exogenous glycine betaine. Plant Science. 148, 185-195.
Yancey, P.H., M.E. Clark, S.C. Hand, R.D. Bowlus and G.N. Somero. 1982. Living with water stress: evolution of osmolyte systems. Science. 217, 1214-1222.
Yost, C.K. and M.F. Hynes. 2000. Rhizobial motility and chemotaxis: molecular biology and ecological role, p. 237-250. In Triplett, E.W., (ed.), Prokaryotic Nitrogen Fixation. A model System for the Analysis of a Biological Process. Horizon Scientifi c Press, Wymondham, UK.
