Journal of Plant Physiology 165 (2008) 641—650 Growth and nitrogen fixation in Lotus japonicus and Medicago truncatula under NaCl stress: Nodule carbon metabolism Miguel Lo ´pez , Jose A. Herrera-Cervera, Carmen Iribarne, Noel A. Tejera, Carmen Lluch Departamento de Fisiologı ´a Vegetal, Facultad de Ciencias, Universidad de Granada, Campus de Fuentenueva s/n, 18071 Granada, Spain Received 26 December 2006; received in revised form 9 May 2007; accepted 10 May 2007 KEYWORDS Carbon metabolism; Lotus japonicus; Medicago truncatula; Salt stress; Trehalose Summary Lotus japonicus and Medicago truncatula model legumes, which form determined and indeterminate nodules, respectively, provide a convenient system to study plant–Rhizobium interaction and to establish differences between the two types of nodules under salt stress conditions. We examined the effects of 25 and 50 mM NaCl doses on growth and nitrogen fixation parameters, as well as carbohydrate content and carbon metabolism of M. truncatula and L. japonicus nodules. The leghemoglo- bin (Lb) content and nitrogen fixation rate (NFR) were approximately 10.0 and 2.0 times higher, respectively, in nodules of L. japonicus when compared with M. truncatula. Plant growth parameters and nitrogenase activity decreased with NaCl treatments in both legumes. Sucrose was the predominant sugar quantified in nodules of both legumes, showing a decrease in concentration in response to salt stress. The content of trehalose was low (less than 2.5% of total soluble sugars (TSS)) to act as an osmolyte in nodules, despite its concentration being increased under saline conditions. Nodule enzyme activities of trehalose-6-phosphate synthase (TPS) and trehalase (TRE) decreased with salinity. L. japonicus nodule carbon metabolism proved to be less sensitive to salinity than in M. truncatula, as enzymatic activities responsible for the carbon supply to the bacteroids to fuel nitrogen fixation, such as ARTICLE IN PRESS www.elsevier.de/jplph 0176-1617/$ - see front matter & 2007 Elsevier GmbH. All rights reserved. doi:10.1016/j.jplph.2007.05.009 Abbreviations: AI, alkaline invertase; ANA, apparent nitrogenase activity; HK, hexokinase; ICDH, isocitrate dehydrogenase; Lb, leghemoglobin; MDH, malate dehydrogenase; NDW, nodule dry weight; NFR, nitrogen fixation rate; PDW, plant dry weight; PEPC, phosphoenolpyruvate carboxylase; RDW, root dry weight; SS, sucrose synthase; TPP, trehalose-6-phosphate phosphatase; TPS, trehalose-6-phosphate synthase; TSS, total soluble sugars; TRE, trehalase Corresponding author. Tel.: +34958 243382; fax: +34958 248995. E-mail address: [email protected](M.Lo´pez).
SummaryLotus japonicus and Medicago truncatula model legumes, which form determinedand indeterminate nodules, respectively, provide a convenient system to studyplant–Rhizobium interaction and to establish differences between the two types ofnodules under salt stress conditions. We examined the effects of 25 and 50mM NaCldoses on growth and nitrogen fixation parameters, as well as carbohydrate contentand carbon metabolism of M. truncatula and L. japonicus nodules. The leghemoglo-bin (Lb) content and nitrogen fixation rate (NFR) were approximately 10.0 and 2.0times higher, respectively, in nodules of L. japonicus when compared withM. truncatula. Plant growth parameters and nitrogenase activity decreased withNaCl treatments in both legumes. Sucrose was the predominant sugar quantified innodules of both legumes, showing a decrease in concentration in response to saltstress. The content of trehalose was low (less than 2.5% of total soluble sugars (TSS))to act as an osmolyte in nodules, despite its concentration being increased undersaline conditions. Nodule enzyme activities of trehalose-6-phosphate synthase (TPS)and trehalase (TRE) decreased with salinity. L. japonicus nodule carbon metabolismproved to be less sensitive to salinity than in M. truncatula, as enzymatic activitiesresponsible for the carbon supply to the bacteroids to fuel nitrogen fixation, such as
sucrose synthase (SS), alkaline invertase (AI), malate dehydrogenase (MDH) andphosphoenolpyruvate carboxylase (PEPC), were less affected by salt than thecorresponding activities in barrel medics. However, nitrogenase activity was onlyinhibited by salinity in L. japonicus nodules.& 2007 Elsevier GmbH. All rights reserved.
Introduction
Roots of legumes have the ability to form asymbiotic relationship with nitrogen-fixing rhizobialbacteria to form nodules, highly specialized organsin which atmospheric dinitrogen is reduced toammonia. Two types of legume nodules havebeen defined: determined and indeterminate no-dules (Hadri and Bisseling, 1998). Differencesbetween determined and indeterminate nodulescould occur because of different vascular bundlearrangements and other anatomical features(Brown and Walsh, 1994). In many indeterminatenodule types, starch is stored in both infected anduninfected cells, whereas it is rarely or never foundin infected cells in determined nodules (Gordonet al., 1992). In addition, in indeterminate nodules,only a part of the tissue fixes nitrogen and containsleghemoglobin (Lb), the major nodule-specificprotein that facilitates O2 supply to the bacteroidsin order to prevent nitrogenase inactivation by O2
(Sprent, 1980; Hirsch, 1992). Lotus and Medicagodevelop determined and indeterminate nodules,and they are nodulated by Mesorhizobium loti andSynorhizobium meliloti, respectively. Legume re-search has been focused on these two speciesbecause they provide convenient and powerfulsystems to study plant–Rhizobium interactions(Stougaard, 2001).
Nitrogen symbiotic fixation in legumes hasproved to be limited, especially in semiarid condi-tions where salinity is often a temporary problembecause of the poor quality of water irrigationduring the dry season. Legumes are classified assalt-sensitive crop species (Lauchli, 1984) and theirlimitation in productivity is associated with lowergrowth of the host plant, poor symbiotic develop-ment of root-nodule bacteria (Georgiev and Atkins,1993) and, consequently, a reduction in the nitro-gen-fixation capacity (Delgado et al., 1993). Understress conditions, biochemical and physiologicalmechanisms such as accumulation of compatibleosmolytes proline, glycine betaine, sucrose andmannitol (Zhu, 2002) are switched on to protectmajor processes such as cell respiration, photosyn-thetic activity, nutrient transport, and nitrogen andcarbon metabolism.
One such compound is trehalose (a-D-gluco-pyranosyl-1,1-a-D-glucopyranoside), a non-reducingdisaccharide that has been found in a wide varietyof organisms such as yeast, fungi, bacteria, plants,insects and other invertebrates (Crowe et al., 1984;Wiemken, 1990). This molecule plays an importantrole as an abiotic stress protectant, stabilizingdehydrated enzymes and membranes as well asprotecting biological structures from desiccationdamage. In higher vascular plants, trehalose is arare sugar (Hoekstra et al., 1992) but may occur inplants with diseases (Keen and Williams, 1969) orin plants colonized by microorganisms, for examplein mycorrhizal roots (Harley and Smith, 1983),nitrogen-fixing nodules (Streeter, 1985) and acti-norhizal nodules (Lopez and Torrey, 1985). Inlegumes, Muller et al. (1994) detected a largeamount of trehalose in root nodules of soybean,primarily localized (70%) in the bacteroid (Streeter,1987). Synthesis of trehalose is a two-step processcatalyzed by sequential action of trehalose-6-phosphate synthase (TPS; EC 2.4.1.15) and treha-lose-6-phosphate phosphatase (TPP; EC 3.1.3.12)activities, called the synthase complex (Bell et al.,1998). It has been suggested that geneticallyengineered trehalose accumulation in crop plantscould improve their tolerance to drought andsalinity (Romero et al., 1997). However, limitedamounts of trehalose were found to accumulate,likely because of the ubiquitous presence of theenzyme trehalase (TRE) in plants (Muller et al.,2001). TRE (EC 3.2.1.28), the only enzyme capableof hydrolyzing trehalose to glucose, has beenmainly found in the plant fraction, but it is alsofound in relatively low levels in bacteroids(Streeter, 1982).
Nodule nitrogen fixation depends on the supply ofsucrose delivered from the phloem (Gordon et al.,1987). This sucrose may be hydrolyzed by eithersucrose synthase (SS; EC 2.4.1.13) or alkalineinvertase (AI; EC 3.2.1.26). SS appears to be thekey enzyme of sucrose hydrolysis in nodules(Gordon et al., 1992). The relative activities ofboth enzymes vary with developmental stage(Gordon and James, 1997). In addition, it has beenobserved that the reduction of nitrogenase activityunder salt stress was associated with the decrease
in SS activity in soybean nodules (Gordon et al.,1997), while in chick-pea nodules AI could com-pensate the lack of SS hydrolytic activity (Soussiet al., 1998). The unequivocal functions of SS andAI in nodule metabolism remain to be established.
In nitrogen-fixing bacteroids, malate appears tobe the most likely substrate (Vance et al., 1985;Salminen and Streeter, 1992) supplied by the hostcell cytosol; this organic compound is formed byphosphoenolpyruvate carboxylase (PEPC; EC4.1.1.31), which utilizes phosphoenolpyruvate (gly-colytic pathway) and malate dehydrogenase (MDH;EC 1.1.1.37). It has been suggested that isocitratedehydrogenase (ICDH) plays a role that ensures anadequate supply of NADPH for plant defense againstoxidative stress (Hodges et al., 2003), or as afunction to supply 2-oxoglutarate to the glutaminesynthase–glutamate synthase (GS-GOGAT) cycle incarbon-limiting situations such as water stress(Galvez et al., 2005) and salinity (Soussi et al.,1998).
The aim of this work was to compare responsesand adaptation to salt stress in indeterminate anddetermined nodules in the model legumes Lotusjaponicus and Medicago truncatula grown undersymbiotic conditions; this was done by evaluatingthe role of trehalose as an osmoprotectant and itsrelation with changes in carbon metabolism and thecarbohydrate pool under salt stress conditions inthe nodule. Plant growth and nitrogen fixationparameters were obtained in order to determinethe tolerance of the two legumes to salt stress.
Materials and methods
Plant material and experimental treatments
L. japonicus (cv. Gifu) and M. truncatula (var. Jema-long) seeds were scarified by immersion in concentratedH2SO4 for 5min, washed with sterile water, surfacesterilized by immersion in 5% NaClO plus Tween 20 for20min and germinated on 0.8% water–agar plates at 28 1Cin the darkness. After 4 days, Lotus and Medicagoseedlings were transferred to sterile vermiculite andwatered with the Hornum nutrient solution (Handbergand Stougaard, 1992) and a modified Rigaud and Puppo(1975) nutrient solution supplied with NaCl (0, 25 and50mM). Two days later, L. japonicus seedlings wereinoculated with 1mL of a stationary culture of Mesorhi-zobium loti R7A strain and M. truncatula seedlings withSinorhizobium meliloti GR4 strain (ca. 109 cellmL�1)grown in a TY medium (tryptone–yeast extract, Beringer,1974). Plants were grown in a controlled environmentalchamber in individual pots of about 200mL, with a 16/8 hlight–dark cycle, 23/18 1C day–night temperature, rela-tive humidity 55/65% and photosynthetic photon fluxdensity (400–700 nm) of 450 mmolm�2 s�1 supplied by
combined fluorescent and incandescent lamps. Plantswere harvested 12 weeks after inoculation and noduleswere frozen at �80 1C for further analyses. Samples ofleaves, stems, roots and nodules were dried at 70 1C for24 h and dry weights estimated.
Nitrogen fixation
Nitrogenase activity (EC 1.7.9.92) was measured as therepresentative H2 evolution in an open-flow system(Witty and Minchin, 1998) using an electrochemical H2
sensor (Qubit System Inc., Canada). For nitrogenasemeasurements, pots maintained in a controlled environ-mental chamber (as described above) were sealed and H2
production was recorded. Apparent nitrogenase activity(ANA, rate of H2 production in air) was determined underN2:O2 (80%:20%) with a total flow of 0.4 Lmin�1. Afterreaching steady-state conditions, total nitrogenase ac-tivity (TNA) was determined under Ar:O2 (79%:21%). Thenitrogen fixation rate (NFR) was calculated as(TNA�ANA)/3. Standards of high-purity H2 were used tocalibrate the detector.
Leghemoglobin determination
Lb content was determined by the fluorometricprocedure of La Rue and Child (1978) with somemodifications. Crude nodule extracts were obtained bygrinding 0.5 g nodules in 6mL 50mM phosphate buffer(pH 7.4) containing 0.02% K3Fe(CN)6 and 0.1% NaHCO3.The extract was centrifuged at 24,000g for 20min and50 mL of supernatant was mixed with a saturated solutionof oxalic acid and heated at 120 1C for 30min. Fluores-cence of the supernatant was measured in a ShimadzuRF-540 fluorimeter (excitation wavelength 405 nm, emis-sion wavelength 600 nm) using calibration curves ofbovine hemoglobin.
Preparation of extracts and enzyme assays
Extracts were prepared by homogenizing 0.2 g ofnodules in a mortar with 33% (w/w) polyvinyl-polypyrro-lidone and 1.5mL of 50mM phosphate K buffer (pH 8)containing 1mM EDTA and 20% (v/v) ethylene glycol forSS and hexokinase (HK) activities; 1.5mL 100mM maleicKOH buffer (pH 6.8) containing 100mM sucrose, 2% (v/v)b-mercaptoethanol and 20% (v/v) ethylene glycol for(PEPC, ICDH and MDH), 2mL of 100mM MES buffer (pH6.3) containing 2mM EDTA and 2mM PMSF for AI and TREactivities; and 2mL 50mM Tris–HCl buffer (pH 7.5)containing 2.5mM MgCl2, 100mM NaCl and 10mM b-mercaptoethanol for TPS and TPP activities. Extractswere centrifuged at 30,000g for 20min and undesaltedsupernatants were used to determine enzyme activities.All operations were carried out at 4 1C, and enzymeactivities were monitored between 2 and 4 h.
SS activity (EC 2.4.1.13) was measured according toMorell and Copeland (1985). The production of UDP-glucose was coupled to the reduction of NAD+ in thepresence of excess UDP-glucose dehydrogenase. Reaction
mixtures were contained in a volume of 1mL, 100mMbicine KOH buffer (pH 8.5), 100mM sucrose, 2mM UDP,0.025U UDP-glucose dehydrogenase and 1.5mM NAD+.The assay was started by the addition of the enzymeextract. SS activity was spectrophotometrically mea-sured by following the NAD+ reduction at 340 nm.
Alkaline invertase activity (EC 3.2.1.26) was assayedby incubating at 30 1C enzyme aliquots in 20mMphosphate K buffer (pH 7.5) containing 100mM sucrose(modified from Morell and Copeland, 1984) for 30min.The assay was stopped by boiling, and the glucosegenerated was determined with the glucose oxidase–
peroxidase method using a test kit (ATOM, BioSystems).HK activity (EC 2.7.1.1) was assayed according to Levi
and Preiss (1978). The reaction mixtures contained 50mMHEPES (pH 6.8), 1mM NADP+, 0.5mM ATP, 1.2mM glucoseand 3U of glucose-6-phosphate dehydrogenase. Activitywas followed by reduction of NADP+ at 340 nm.
PEPC (EC 4.1.1.31) and MDH (EC 1.1.1.37) activityassays were optimized from the Soussi et al. (1998)method. The reaction mixtures contained 100mM bicineKOH (pH 8.5), 5.0mM MgCl2, 0.2mM NADH, 10mMNaHCO3 2.0mM PEP for PEPC and 1mM oxalacetic acidfor MDH. Activity was followed by oxidation of NADH at340 nm.
ICDH activity (EC 1.1.1.42) was assayed according toChen et al. (1988). The reaction mixture contained100mM bicine KOH (pH 8.5), 5.0mM MgCl2, 1.0mMisocitrate and 0.5mM NADP+. Activity was followed byreduction of NADP+ at 340 nm.
TRE activity (EC 3.2.1.28) was determined colorime-trically according to Muller et al. (1994) by measuring theglucose released. The reaction mixture contained 100mMtrehalose in 50mM MES/KOH (pH 6.3). After incubation at37 1C for 45min, the reaction was stopped by heating at100 1C for 5min. The glucose released was measured bythe glucose oxidase–peroxidase method.
Trehalose-6-phosphate synthetase assay (EC 2.4.1.15)was based on the method of Salminen and Streeter (1986)by determining the release of UDP from UDP-glucose inthe presence of glucose-6-phosphate. The reactionmixture (0.2mL) contained 100mM Tris–HCl buffer (pH7.5), 8mM UDP-glucose, 30mM glucose-6-phosphate,100mM MgCl2, 3mM EDTA and 25mM KCl. The reactionwas started by the addition of the nodule enzyme extract(0.04mL). After 60min at 30 1C, reactions were stoppedby heating at 100 1C for 2min. Samples were centrifugedat 2000g for 10min, and the amount of UDP in thesupernatant was measured in terms of oxidation of NADHin a linked assay with pyruvate kinase and lactic aciddehydrogenase. The assay mixture contained 50mMTris–HCl (pH 7.5), 5mM phosphoenolpyruvate, 0.24mMNADH, 10mM MgCl2, 3.5 U pyruvate kinase and 5U lacticacid dehydrogenase. The decrease in absorbance at340 nm was measured continuously over a period of20min.
TPP activity (EC 3.1.3.12) was assayed by monitoringphosphate release from trehalose-6-phosphate (Padilla etal., 2004). The reaction was carried out in a final volumeof 0.25mL containing 25mM Tris–HCl (pH 7.0), 10mMMgCl2 and 1 mM trehalose-6-phosphate. Samples were
assayed for phosphate by the zinc acetate method(Bencini et al., 1983).
Carbohydrate analysis
Carbohydrates glucose, sucrose, fructose and treha-lose were separated and quantified by gas chromatogra-phy (Streeter and Strimbu, 1998). Samples of nodules(200mg) were ground in methanol (80% v/v) andincubated at 60 1C for 10min, followed by centrifugationat 13,000g for 10min. The pellet was re-extracted threemore times, and the supernatants were collected andvacuum-dried. Solids were dissolved in 125 mL purepyridine plus 125 mL STOX reagent (Pierce Biotechnology,Inc.). The samples were derivatized by adding 200 mLhexamethyldisilazane and 20 mL trifluoroacetic acid60min before analysis. TMS-oxime derivatives wereseparated on a packed column of 3% OV-17 on Chromo-sorb WHP using a Hewlett-Packard 5890 Series II gaschromatograph and peak areas were quantified using aHewlett-Packard 3396A integrator. All reagents werepurchased from Pierce Chemical Co. (Rockford, IL).
Starch and total soluble sugars (TSS) were determinedby the colorimetric methods of Kiniry (1993) and Irigoyenet al. (1992), respectively. Samples of nodules (100mg)were ground in ethanol (70% v/v) and centrifuged at5000g for 10min. For starch analysis, the insolublematerial was dried at 70 1C for 24 h, resuspended in4mL of distilled water and boiled for 1 h. After cooling,1mL of 8.5mM acetate buffer containing amyloglucosi-dase (0.8mgmL�1), pH 4.5, was added. The mixture wasincubated overnight at 50 1C. The glucose released wasdetermined with the kit ATOM as described above for AIactivity assay. The quantification of TSS was performedon ethanolic extracts incubated with anthrone reagent at100 1C for 10min and the absorbance measured at625 nm.
Statistical analyses
The experimental layout was a randomized completeblock design. The growth and nitrogen fixation valueswere means of five replicates per treatment. Threereplicates were performed for the other parametersstudied. All results were subjected to two-way analysis ofvariance with a least significant difference test betweenmeans using Statgraphics 5.0 (Statistical Graphics Corp.,Rockville, MD, USA). The standard error (SE) and simplecorrelation coefficients were also calculated.
Results
Plant biomass and nitrogen fixation
Plant biomass and nitrogen fixation parametersof both species were markedly affected by saltstress conditions (Table 1). At harvest time (flower-ing stage), a decrease of approximately 40% in
a–cMeans followed by the same letter within a column do not differ (Pp0.05) using the LSD test.
Table 2. Concentration of total soluble sugars (TSS, mg glucose g�1 NDW), starch (mg glucose g�1 NDW), sucrose(mg g�1 NDW), glucose (mg g�1 NDW), fructose (mg g�1 NDW) and trehalose (mg g�1 NDW) in nodules of M. truncatula andL. japonicus inoculated with the S. meliloti GR4 and M. loti R7A strains, respectively, and treated with NaCl
Species NaCl (mM) TSS Starch Sucrose Glucose Fructose Trehalose
a–cMeans followed by the same letter within a column do not differ (Pp0.05) using the LSD test.
Nodule carbon metabolism under salt stress 645
plant dry weight (PDW) and root dry weight (RDW)in M. truncatula and L. japonicus with 25mM NaClwas observed. However, for PDW and RDW inL. japonicus, no significant differences were ob-served between 25 and 50mM treatments.
Regarding nitrogen fixation parameters (Table 1),M. truncatula nodule dry weight (NDW) wasunaffected by salt stress, while L. japonicus NDWshowed a 40% decrease under salinity. The Lbcontent in M. truncatula nodules decreased about40% with 25mM and 80% with 50mM NaCl. However,in L. japonicus nodules, a significant increase underboth levels of salt stress was observed (50% and 70%with 25 and 50mM NaCl, respectively). ANA, totalnitrogenase activity (TNA) and NFR showed adecline in M. truncatula under salt stress, althoughno statistically significant differences were ob-served. In contrast, in L. japonicus, only ANA wasaffected by the 50mM salt dose, showing a declineof 55%. Similar results were obtained for TNA andNFR. An important difference in nodule Lb contentand nitrogenase activity was observed betweenboth types of nodules; the amount of Lb was about
10 times higher in L. japonicus control nodules(determined) than in M. truncatula (indetermi-nate) nodules. Similarly, ANA was 2.5 and TNA andNFR 2.2 times higher, respectively, in L. japonicusnodules compared with M. truncatula nodules.Indeed, in ANA as with TNA, there was a positivesignificant correlation with PDW and RDW (Pp0.05)in M. truncatula. These correlation values supportthe close relationship between plant growth andthe symbiotic nitrogen fixation of nodules.
Nodule carbohydrates
Sucrose was the predominant sugar quantified innodules of M. truncatula and L. japonicus (Table 2),showing a decrease of 34% with 25mM and 66% with50mM in concentration in M. truncatula nodules,and a 35% decrease in L. japonicus with 50mMNaCl. The amount of TSS was about 4.5 timeshigher in M. truncatula nodules relative toL. japonicus. The response to salt stress wasdifferent between the two species; the content of
ARTICLE IN PRESS
M. Lopez et al.646
TSS decreased in M. truncatula nodules, while inL. japonicus, it increased by 35% and 60% with 25and 50mM NaCl, respectively. The starch content inM. truncatula nodules was about 20 times highercompared with L. japonicus nodules. However, asignificant starch decrease in M. truncatula nodulesby salt stress was observed (75%). This behaviordiffered in L. japonicus nodules where the starchamount increased by 70% and 190% with 25 and50mM doses, respectively. It is noteworthy thatglucose could not be detected in M. truncatulanodules; however, in L. japonicus nodules, glucosewas found and its concentration increased undersalinity (50mM NaCl). Similarly, fructose was moreabundant in M. truncatula nodules but showed adecline in concentration with 25 and 50mM NaCl,and a 70% increase was observed in L. japonicusnodules with 50mM NaCl. Trehalose concentrationwas similar in both species and increased with salttreatment. In L. japonicus nodules, the increaseobserved was 31% and 56% with 25 and 50mM NaCl,respectively. In M. truncatula nodules, the contentof trehalose increased in concentration by 155%with the lower salt dose and 107% with 50mM NaCl.There was a high positive significant correlationbetween TSS and glucose, fructose and trehalose(rX0.85**) in L. japonicus, and TSS and sucrose andfructose content in M. truncatula (rX0.80*).
Enzyme activities of carbon metabolism
The enzymes SS and AI, involved in sucrosecleavage (Figure 1), showed remarkable inhibitionwith salinity in M. truncatula nodules, where thedecrease in activity was proportional to theincrease of the salt dose. In L. japonicus nodulesboth activities were less sensitive to salinity,showing only a 30% decrease in the SS activity withthe 50mM dose, and a slight inhibition of about 10%in the IA activity with 25 and 50mM NaCl. HKactivity showed a different pattern in M. truncatulathan in L. japonicus nodules (Figure 1), whereactivity increased by 39% and 67% with 25 and50mM NaCl, respectively. In M. truncatula nodules,the activity did not show significant changes. Theenzymatic activities ICDH, MDH and PEPC showed asimilar pattern; in general, enzymes were moresensitive to salinity in M. truncatula nodules. InL. japonicus nodules, no effect of salinity wasobserved in ICDH, in MDH only a 10% decline wasinduced by 50mM NaCl dose and in PEPC thedecline was 15% with 25mM NaCl and 25% with50mM NaCl. It is noteworthy that PEPC activity was5 times higher in L. japonicus than in M. trunca-tula, which is in agreement with the differences
found in the nitrogen fixation parameters. Thecorrelations among PEPC, MDH and ICDH werepositive and significant (Pp0.01) in both species.Carbon metabolism enzymatic activity results sup-port the fact that the decrease in nitrogen fixationmay be due to the inhibition of the carbon flowfrom the plant to the bacteroids to fuel nitrogenfixation.
Enzyme activities of trehalose metabolism
The role of trehalose in the adaptation of nodulemetabolism to salt stress is one of the goals of thiswork, and biosynthetic and catabolic enzymeactivities of trehalose metabolism were studied.Synthesis of trehalose was inhibited by the saltstress in M. truncatula and L. japonicus nodules(Figure 2), showing a reduction of 50% and 35% inthe TPS activity, respectively. Trehalose catabolismmeasured by the TRE activity was inhibited bysalinity, although to a lesser extent than synthesis.
Discussion
In general, salt stress led to a decrease of growthand development of the plant. Nevertheless, it isknown that plants have a suite of morphologicaland physiological strategies to allow themselves tosurvive under salt stress conditions. The adaptationcapacity to salinity may vary considerably betweenspecies and, as a result, sensitivity and/or toler-ance to salt of the host plant cultured undersymbiotic conditions were different in M. trunca-tula and L. japonicus. In this regard, the reductionof plant growth and dry-matter accumulation undersaline conditions of M. truncatula were moreseverely affected than those of L. japonicus.Nitrogen fixation parameters, such as the contentsof Lb, ANA and NFR, showed higher values inL. japonicus nodules than in M. truncatula, whichwould be related to the observation that inindeterminate nodules only a part of the tissuecontains Lb and fixes nitrogen (Sprent, 1980;Hirsch, 1992). The reduction in Lb content inM. truncatula could limit the oxygen supply tothe bacteroids and could affect respiration; asimilar finding was reported by Fernandez-Pascualet al. (1996) in white lupin nodules under saltstress. Despite the fact that the growth of bothlegumes depends on N2 fixation, no significantcorrelation between growth and nitrogen fixationparameters was found, at least at harvest time.Since most of the plant biomass is generatedduring vegetative growth, we do not rule out a
Figure 1. Effect of NaCl treatments on nodule enzymes activities: sucrose synthase (SS), alkaline invertase (AI),hexokinase (HK), isocitrate dehydrogenase (ICDH), malate dehydrogenase (MDH) and phosphoenolpyruvate carboxylase(PEPC) of M. truncatula and L. japonicus. Values of each legume followed by the same letter do not differ significantlyat Pp0.05. Vertical bars represent 7SE (n ¼ 4) of four replicates.
Nodule carbon metabolism under salt stress 647
relationship between these parameters at thisgrowth stage.
Regarding nodule carbohydrate content, themost intriguing effect was the increase of TSS,glucose and fructose in L. japonicus nodules of thesalt-stressed plants. It has been assumed that theaccumulation of organic solutes in response to saltstress is involved in protection mechanisms, such asthe restoration of cell volume and turgor, thereduction of cell damage induced by free radicals,and the protection and stabilization of enzymes andmembrane structures (Chen and Murata, 2002). Thecontent of compatible solutes increased with thesalt dose; therefore, the accumulation of compa-tible solutes is a consequence of damage produced
by salt stress, more than that of a protectivestrategy, as Lutts et al. (1996) reported in Oryzasativa and Soussi et al. (1998) in Cicer arietinum,suggesting an osmoregulatory mechanism of carbo-hydrates in nodules. Similar findings have beenreported in soybean nodules (Muller et al., 1996)and alfalfa nodules exposed to water and salt stress(Fougere et al., 1991). The increase in TSS understress conditions in L. japonicus nodules is consis-tent with the increase in glucose (r ¼ 0.91**),fructose (r ¼ 0.85**) and trehalose (r ¼ 0.99***).Surprisingly, sucrose concentration showed a nega-tive correlation with TSS, probably through theeffect of other carbohydrates that were notanalyzed separately by GC and accounted as TSS,
Figure 2. Effect of NaCl treatments on nodule enzymesactivities: trehalose-6-phosphate synthase (TPS), trehalose-6-phosphate phosphatase (TPP) and trehalase (TRE) of M.truncatula and L. japonicus. Values of each legume followedby the same letter do not differ significantly at Pp0.05.Vertical bars represent 7SE (n ¼ 4) of four replicates.
M. Lopez et al.648
whereas in M. truncatula the TSS content innodules diminished and showed a positive andsignificant correlation with the level of sucrose
Sucrose was the predominant sugar in nodules ofboth species, and its concentration decreased withthe salt dose. Results of chlorophyll fluorescenceand total chlorophyll content (data not shown)suggest a decrease of photosynthetic activity inboth salt-stressed legumes and, consequently, alower sucrose flow from leaves to nodules. Never-theless, other authors have found increased sucrosecontent in root nodules of soybean (Muller et al.,1996; Gordon et al., 1997) exposed to droughtstress, and alfalfa (Fougere et al., 1991) and whitelupin (Fernandez-Pascual et al., 1996) exposed tosalt stress. According to our results, we suggestthat sucrose does not act as a salt stress-relatedosmoprotectant in M. truncatula or in L. japonicusnodules.
The decrease in sucrose levels could be relatedto the increase of starch content in L. japonicusnodules (r ¼ �0.90**) by the formation of ADP-glucose by SS activity (Pozueta-Romero et al.,1999). Zrenner et al. (1995) have suggested thatin some tissues SS is directly involved in starchsynthesis, which may explain why starch levelsdecreased in parallel with the decline in SS activityin M. truncatula (r ¼ 0.95***), where this enzymeshowed a higher sensitivity to salinity. In suchconditions, only SS activity of L. japonicus nodulesshowed a high correlation with nitrogen fixationparameters (ANA r ¼ 0.97***, TNA r ¼ 0.96***, NFRr ¼ 0.85**), in accordance with the finding ofGordon et al. (1997) in soybean nodules.
The enzyme activities responsible for the carbonsupply to the bacteroids by the formation ofdicarboxylates (PEPC, MDH and ICDH), consideredas the primary carbon sources for bacteroids(Streeter, 1981, 1987), showed higher values andwere more tolerant to salinity in L. japonicus thanin M. truncatula nodules, which could account forhigher nitrogen fixation efficiency in L. japonicusnodules.
Our results indicate a significant increase oftrehalose content in M. truncatula and L. japonicusnodules in response to salt stress. These findingssupport the possible role of this disaccharide as anosmoprotectant against abiotic stresses (Sampedroand Uribe, 2004). However, the trehalose concen-tration detected accounted for less than 2.5% ofthe carbohydrate pool in M. truncatula andL. japonicus nodules. Thus, the concentration wastoo low to contribute efficiently to osmoregulation.This result is related to those found by Fougereet al. (1991) and Muller et al. (1994) in Medicagosativa nodules. Nevertheless, trehalose could act tostabilize cell proteins and membranes and protect
them against damage by oxygen radicals understress conditions at low concentrations (Benaroudjet al., 2001; Sampedro and Uribe, 2004). We do notdismiss the possibility that other compatibleosmolytes such as proline (which increased undersalt stress, data not shown) could be involved in theprotection and/or adaptation of L. japonicusnodules to salinity. Previous experiments of Mar-quez (2005), with nitrate-feed plants, reported upto a 12-fold increase of proline concentration inresponse to drought and salt stress.
Interestingly, trehalose accumulation and TREactivity were negatively correlated in L. japonicusnodules (r ¼ �0.97***). Thus, TRE activation innodules (Figure 2) is correlated with nitrogenfixation, rather than with the availability of thissubstrate (Muller et al., 1994). In this sense, TREactivity was higher in L. japonicus nodules relativeto M. truncatula that showed lower nitrogenfixation efficiency. These results are in accordancewith Muller et al. (1994), who reported higher TREactivity in effective soybean nodules than ineffec-tive nodules. We have previously hypothesized thatthe control of trehalose accumulation in nodules ismainly due to TRE activity, since it is a nodule-induced enzyme (Muller et al., 1995). Our resultsshow that the control of nodule trehalose contentmight be influenced by other unknown factors.
Acknowledgments
The authors are grateful to Dr. John G. Streeterfor making valuable suggestions and other contri-butions. Financial support was obtained throughthe Andalusian Research Program (AGR-139) andthe Spanish Ministry of Education and Culture GrantBOS2002-04182-C02-02.
References
Bell W, Sun W, Hohmann S, Wera S, Reinders A, De VirgilioC, et al. Composition and functional analysis of theSaccharomyces cerevisiae trehalose synthase com-plex. J Biol Chem 1998;273:33311–9.
Benaroudj N, Lee DH, Glodberg AL. Trehalose accumula-tion during cellular stress protects cells and cellularproteins from damage by oxygen radicals. J Biol Chem2001;276:24261–7.
Bencini DA, Wild JR, O’Donovan GA. Lincar one-step assayfor the determination of orthophosphate. Anal Bio-chem 1983;132:254–8.
Beringer JE. R factor transfer in Rhizobium leguminosar-um. J Gen Microbiol 1974;84:188–98.
Brown SM, Walsh KB. Anatomy of the legume nodulecortex with respect to nodule permeability. Aust JPlant Physiol 1994;21:49–68.
Chen RD, Marechal JV, Jacquot JP, Gadal P. Purificationand comparative properties of the citosolic isocitratedehydrogenases (NADP) from pea (Pisum sativum L.)root and green leaves. Eur J Biochem 1988;175:565–72.
Chen THH, Murata N. Enhancement of tolerance ofabiotic stress by metabolic engineering of betainesand other compatible solutes. Curr Opin Plant Biol2002;5:250–7.
Crowe J, Crowe L, Chapman D. Preservation of mem-branes in anhydrobiotic organisms: the role oftrehalose. Science 1984;223:701–3.
Delgado MJ, Garrido JM, Ligero F, Lluch C. Nitrogenfixation and carbon metabolism by nodules andbacteroids of pea plants under sodium chloride.Physiol Plant 1993;89:824–9.
Fernandez-Pascual M, De Lorenzo C, De Felipe MR,Rajalakshi S, Gordon AJ, Thomas BJ, et al. Possiblereasons for relative salt stress tolerance in nodulesof white lupin cv. Multolupa. J Exp Bot 1996;47:1709–16.
Fougere F, Rudulier D, Streeter JG. Effects of saltstress on amino acid, organic acid, and carbohydratecomposition of roots, bacteroids, and cytosol ofalfalfa (Medicago sativa L.). Plant Physiol 1991;96:1228–36.
Galvez L, Gonzalez EM, Arrese-Igor C. Evidence forcarbon flux shortage and strong carbon/nitrogeninteractions in pea nodules at early stages of waterstress. J Exp Bot 2005;56:2551–61.
Georgiev GI, Atkins CA. Effects of salinity on N2 fixation,nitrogen metabolism and export and diffusive con-ductance of cowpea root nodules. Symbiosis 1993;15:239–55.
Gordon AJ, James CL. Enzymes of carbohydratesand amino acid metabolism in developing andmature nodules of white clover. J Exp Bot 1997;48:895–903.
Gordon AJ, Mitchel DF, Ryle GJA, Powell CE. Diurnalproduction and utilization of photosynthate in nodu-lated white clover. J Exp Bot 1987;38:84–98.
Gordon AJ, Thomas BJ, Reynolds PHS. Localization ofsucrose synthase in soybean root nodules. New Phytol1992;122:35–44.
Gordon AJ, Minchin FR, Skot L, James CL. Stress induceddecline in soybean N2 fixation are related to nodulesucrose synthase activity. Plant Physiol 1997;114:937–46.
Hadri A, Bisseling T. Responses of the plant to nodfactors. In: Spaink H, Kondorosi A, Hooykaas P, editors.The Rhizobiacea. Dordrecht: Kluwer Academic Pub-lishers; 1998. p. 402–16.
Handberg K, Stougaard J. Lotus japonicus, an autoga-mous, diploid legume species for classical andmolecular genetics. Plant J 1992;2:487–96.
Harley JL, Smith SE. In: Mycorrhizal Symbiosis New York,USA: Academic Press; 1983.
Hirsch A. Developmental biology of legume nodulation.New Phytol 1992;122:211–37.
Hodges M, Flesch V, Galvez S, Bismuth E. Higher plantNADP dependent isocitrate dehydrogenases, ammo-nium assimilation and NADPH production. Plant PhysiolBiochem 2003;41:577–85.
Hoekstra FA, Crowe JH, Crowe LM, Van Roekel T, VermeerT. Do phospholipids and sucrose determine membranephase transitions in dehydrating pollen species? PlantCell Environ 1992;15:601–6.
Irigoyen JJ, Emerich DW, Sanchez-Dıaz M. Water stressinduced changes in concentrations of proline and totalsoluble sugar in nodulated alfalfa (Medicago sativa)plants. Physiol Plant 1992;84:55–60.
Keen NT, Williams PH. Translocation of sugars intoinfected cabbage tissues during clubroot develop-ment. Plant Physiol 1969;44:748–54.
Kiniry JR. Non structural carbohydrate utilization bywheat shaded during grain growth. Agron J 1993;85:844–9.
La Rue and Child. Sensitive fluorometric assay forleghemoglobin. Ann Biochem 1978;92:11–5.
Lauchli A. Salt exclusion: an adaptation of legumes forcrops and pastures under saline conditions. In: RichardC, Gary H, editors. Salinity tolerance in plants. 1984.pp. 171–87.
Levi C, Preiss AS. Amylopectin degradation in peachloroplast extracts. Plant Physiol 1978;61:218–20.
Lopez MF, Torrey JG. Purification and properties oftrehalase in Frankia. Arch Microbiol 1985;143:209–15.
Lutts S, Kinet JM, Bouharmont J. Effect of various saltand mannitol on ion and proline accumulation inrelation to osmotic adjustment in rice (Oryza sativaL.) callus cultures. J Plant Physiol 1996;148:86–95.
Marquez AJ. Lotus japonicus handbook. Dordrecht:Springer; 2005. p. 187–95.
Morell M, Copeland L. Enzymes of sucrose breakdown insoybean nodules: alkaline invertase. Plant Physiol1984;74:1030–4.
Morell M, Copeland L. Sucrose synthase of soybeannodules. Plant Physiol 1985;78:149–54.
Muller J, Xie ZP, Staehelin C, Mellor RB, Boller T,Wiemken A. Trehalose and trehalase in root nodulesfrom various legumes. Physiol Plant 1994;90:86–9.
Muller J, Boller T, Wiemken A. Pools of non-structuralcarbohydrates in soybean root nodules during waterstress. Physiol Plant 1996;98:723–30.
Muller J, Aeschbacher RA, Wingler A, Boller T, WiemkenA. Trehalose and trehalase in Arabidopsis. PlantPhysiol 2001;125:1086–93.
Padilla L, Kramer R, Stephanopoulos G, Agosin E. Over-production of trehalose: heterologous expression ofEscherichia coli trehalose-6-phosphate synthase andtrehalose-6-phosphate phosphatase in Corynebacter-ium glutamicum. Appl Environ Microbiol 2004;70:370–6.
Pozueta-Romero J, Perata P, Akazawa T. Sucrose-starchconversion in heterotrophic tissues of plants. Crit RevPlant Sci 1999;18:179–87.
Rigaud J, Puppo A. Indole-3-acetic acid catabolism bysoybean bacteroids. J Gen Microbiol 1975;88:223–8.
Romero C, Belles JM, Vaya JL, Serrano R, Culianez-MaciaFA. Expression of the yeast trehalose-6-phosphatesynthase in transgenic tobacco plants: pleiotropicphenotypes include drought tolerance. Planta1997;201:293–7.
Salminen O, Streeter J. Enzymes of trehalose metabolismin soybean nodules. Plant Physiol 1986;81:538–41.
Salminen O, Streeter J. Labeling of carbon pools inBradyrhizobium japonicum and Rhizobium legumino-sarum bv. viciae bacteroids following incubationsof intact nodules with 14CO2. Plant Physiol 1992;100:597–604.
Sampedro JG, Uribe S. Trehalose–enzyme interactionsresult in structure stabilization and activity inhibition.The role of viscosity. Mol Cell Biochem 2004;Dec:127–319.
Soussi M, Ocana A, Lluch C. Effect of salt stress ongrowth, photosynthesis and nitrogen fixation in chick-pea (Cicer arietinum L.). J Exp Bot 1998;325:1329–37.
Sprent J. Root nodule anatomy, type of export productand evolutionary origine in some leguminosae. PlantCell Environ 1980;3:35–43.
Stougaard J. Genetics and genomics of root symbiosis.Curr Opin Plant Biol 2001;4:328–35.
Streeter JG. Seasonal distribution of carbohydrates innodules and stem exudate from field grown soya beanplants. Ann Bot 1981;48:441–50.
Streeter JG. Enzymes of sucrose, maltose, and aa-trehalose catabolism in soybean root nodules. Planta1982;155:112–5.
Streeter JG. Accumulation of a a-trehalose by Rhizobiumbacteria and bacteroids. J Bacteriol 1985;164:78–84.
Streeter JG. Carbohydrate, organic acid, and amino acidcomposition of bacteroids and cytosol from soybeannodules. Plant Physiol 1987;85:768–73.
Streeter JG, Strimbu CE. Simultaneous extraction andderivatization of carbohydrates from green planttissues for analysis by gas–liquid chromatography.Anal Biochem 1998;259:253–7.
Vance CP, Boylan KLM, Maxwell CA, Heichel GH, HardmanLL. Transport and partitioning of CO2 fixed by rootnodules of ureide and amide producing legumes. PlantPhysiol 1985;78:774–8.
Wiemken A. Trehalose in yeast, stress protectant ratherthan reserve carbohydrate. Antonie van LeeuwenhoekJ Gen Microbiol 1990;58:209–17.
Witty JF, Minchin FR. Methods for the continuousmeasurement of O2 consumption and H2 productionby nodulated legume root systems. J Exp Bot 1998;49:1041–7.
Zhu JK. Salt and drought stress signal transduction inplants. Annu Rev Plant Biol 2002;53:247–73.
Zrenner R, Salanoubat M, Willmitzer L, Sonnewald U.Evidence of the crucial role of sucrose synthase forsink strength using transgenic potato plants (Solanumtuberosum L.). Plant J 1995;7:97–107.
