Alleviating the inhibitory effect of salinity stress on nod gene expression in Rhizobium tibeticum – fenugreek (Trigonella foenum graecum) symbiosis by isoflavonoids treatment

Abstract

Rhizobia-legume symbiosis depends on molecular dialog, which involves the production of specific plant flavonoid compounds as signal molecules. Rhizobium tibeticum was recovered from the root nodule of fenugreek and identified by sequencing the 16S rRNA gene. The effect of salinity stress on nod gene expression was measured in terms of β-galactosidase activity. R. tibeticum containing Escherichia coli lacZ gene fusions to specific nodulation (nod) genes were used to determine β-galactosidase activity. Combination of hesperetin (7.5 µM) and apigenin (7.5 µM) significantly increased β-galactosidase activity more than the single application of hesperetin or apigenin. Preincubation of R. tibeticum with hesperetin and apigenin combination significantly alleviates the adverse effect of salinity on nod gene expression and therefore, enhances nodulation and nitrogen fixation of fenugreek.

Keywords:

• Rhizobium tibeticum
• isoflavonoids
• salinity
• β-galactosidase
• nodulation

Introduction

Bacteria in the genus Rhizobium invade specific host plants and stimulate the development of highly organized nitrogen-fixing root nodules (Oke & Long 1999). The symbiosis benefits both partners as the prokaryotic partner receives carbohydrate and the symbiotic bacteria provide the plant with nitrogenous compounds. In general, legume plants exude into their rhizosphere complex cocktails of sugars, flavones, or isoflavones, which are perceived as nod gene inducers in Rhizobium. While many details of this relationship remain a mystery, much effort has gone into elucidating the mechanisms governing bacterium-host recognition and the events leading to symbiosis. Several signal molecules, including plant-produced flavonoids, bacterially produced nodulation factors, and exopolysaccharides, are known to function in the molecular conversation between the host and the symbiont (González & Marketon 2003).

Legume species are the crops with the highest potential for biological nitrogen (N₂) fixation already available to be used in the productive systems. Fenugreek (Trigonella foenum graecum) is an annual herb that belongs to the family Leguminosae widely grown in Egypt, Pakistan, India, and Middle Eastern countries (Bukhari et al. 2008). It is an old medicinal plant and has been commonly used as a traditional food and medicine.

Successful symbiotic interactions are complex and require the regulation and function of multiple genes/gene families in both partners (Nandasena et al. 2009). Flavonoids and oxylipin play a crucial role as signal molecules in promoting the formation of nodules by symbiotic bacteria commonly known as rhizobia (Yanagi et al. 2011). Exudates from sprouted seeds can be used as activators of PnodA-lacZ fusion in rhizobia (Maj et al. 2010). It has been documented that some flavonoids are present in the root. The seed extract may be symbiotically ineffective or even inhibit nodulation. Mixtures of flavonoids and other compounds that are exuded by plant roots are thought to act as signals that influence the ability of rhizobia to colonize the roots, survive in the rhizosphere, and affect the competitiveness of rhizobia and symbiotic interactions with legumes (Cooper 2007).

The early interaction between flavonoids and NodD regulatory protein activates nod gene transcription and the synthesis of Nod factor that initiates nodule primordium (Maj et al. 2010; Abd-Alla 2011).

Poor nodulation and N₂ fixation of legumes, which can lead to substantial loss of yield, has been attributed to a range of environmental conditions, including unfavorable soil pH (Coventry & Evans 1989; Lin et al. 2012), high salinity (Zahran & Sprent 1986; Abd-Alla 1992; Echeverria et al. 2013), and heavy metals (Ali et al. 2010). Soil temperature is also an important environmental variable which affects legume nodulation (Abd-Alla & Abdel Wahab 1995; Hayat et al. 2012).

The inhibitory effects of soil stress may be through affecting the signal exchange process between the two partners (Li et al. 2012). According to our knowledge, there is no publication on the effects of signal molecule on the Rhizobium tibeticum – fenugreek symbiosis under saline conditions. Therefore, the current experiments were designed to: (1) determine the stressful effects of salinity on the R. tibeticum – fenugreek symbiosis, hypothesizing that such effects are related to the inhibitory effects of salinity on the signal exchange process between the two partners, and (2) determine whether the addition of the combination of hesperetin and apigenin (a nod gene inducer) to R. tibeticum inocula could overcome the stressful effects of salinity on the R. tibeticum – fenugreek symbiosis.

Materials and methods

Collection and isolation of root nodule bacteria

Fenugreek (T. foenum graecum) plants were collected from different fields at Diurot, Elkossia, Manfalouit, and Assuit, Assuit Governorate, Egypt during the growth season (December and January 2010). The collected samples were kept in the refrigerator until the time of isolation. Roots were washed thoroughly to remove soil. About 10 nodules were collected from each plant. The nodules were severed from the root by cutting the root by about 0.5 cm on each side of the nodule. Collected nodules were washed with sterile water and then surface sterilization was done using 95% ethanol for 5–10 sec and 2.5–3% (v/v) solution of sodium hypochlorite, and soaked for 2 sec and washed repeatedly with sterile water. After surface sterilization, nodules were crushed and then the resulting suspension was streaked onto the yeast extract mannitol agar (YEMA) plate containing Congo Red. A single colony was transferred into YEMA plates. After repeated subculturing, pure culture was obtained from a single colony and preserved in 40% glycerol at −20°C for experimental purpose.

Sequencing of 16S rRNA gene

DNA was extracted from bacterial cultures using an sodium dodecyl sulfate/ cetyl trimethylammonium bromide lysis and phenol/chloroform extraction method (Ausubel et al. 1987). The 16S rRNA gene was PCR-amplified using primer pairs 16S-F 10430 AGAGTTTGATCCTGGCTCAG and 16S-R10430 AAGGAGGTGATTCC AGCC that were used to amplify a near-full length, approximately 1500 bp fragment of 16S rDNA from the isolates. Amplification of 16S rRNA fragments from genomic DNA was carried out in a total reaction volume of 100 µl containing 5 µl of bacteria DNAase template, 5 µl of each forward and reverse primer (10 µM), 50 µl Go TagR Master Mix, 2x Kit (Promega, USA), and 35 µl of nuclease-free water. The reaction was performed using TC-3000G Techne Thermal Cycler (Bibby Scientific Ltd, UK). The reaction conditions were an initial denaturation at 95°C for 3 min, 35 cycles of denaturation at 94°C for 70 sec, annealing at 56°C for 40 sec, and extension at 72°C for 130 sec. A final extension was conducted at 72°C for 370 sec. PCR products were purified (MonoFas DNA Purification Kit I, GL Sciences, Inc. Japan) and quantified photometrically (Evolution 300 UV-Vis Spectrophotometer, USA). Purified PCR products were cycle sequenced in both directions with the same forward and reverse primers using ABI 3730XLApplied Biosystems, DNA Analyzer. The sequence reads were edited and assembled using BioEdit version 7.0.4 (http://www.mbio.ncsu.edu/BioEdit/bioedit.html) and clusstal W version 1.83 (http://clustalw.ddbj.nig.ac.jp/top-e.html). BLASTN searches were done using the NCBI server at http://www.ncbi.nlm.nih.gov/blast/Blast.cgi. Phylogenetic trees derived from 16S rRNA gene sequence was constructed in the context of 16S rRNA gene sequences from 11 different standard bacterial strains obtained from Genbank Rhizobium mesosinicum (043548.1), Mesorhizobium loti LMG6125 (X67229), Rhizobium leguminosarum USDA-2370 (044774), R. tibeticum (JN896365), Rhizobium phaseoli ATCC 14482 (044112), R. tibeticum HM032840.1, R. tibeticum (EU256404), Rhizobium selenitireducens NR (044216.1), Rhizobium daejeonense NR (042851.1), Rhizobium alamii NR (042687.1), and Bacillus subtilis (JQ653051.1).

Construction of the plasmid vector containing PnodA-lacZ fusion

Bacterial strain

E. coli α which contained the expression vector pMP221 was a gift from Dr Anna Skorupska and Dominika Maj, Department of Genetics and Microbiology, Maria Curie-Sklodowska University. R. tibeticum isolated from the nodules on the roots of fenugreek.

Plasmid isolation

DNA plasmid was isolated by using the alkaline lysis method (Bimboim & Doly 1979).

Transformation of pMP221 vector into rhizobial strain by using a freeze–thaw method

Fifty milliliters of YEMA medium was inoculated with R. tibeticum and grown at 28°C with vigorous shaking until they reached the stationary growth phase. Cells were harvested by centrifugation at 12,000g for 10 min at 4°C, and the pellet was resuspended in 1 ml of ice-cold 20 mM CaCl₂ solution. pMP221 vector was transformed into rhizobial strain by using a freeze–thaw method (Vincze & Bowra 2006). Cell suspension was transferred into YEMA medium with tetracycline. The Tc^r colonies were purified by successive isolation and passages in the selective medium.

Conditions of rhizobial culture for bioassay for nod gene-inducing activity

R. tibeticum cells isolated from fenugreek were grown at 28°C on solid YEMA medium (Hooykaas et al. 1977). For stable maintenance of the recombinant plasmids, the medium was supplemented accordingly with streptomycin (400 µg ml⁻¹), chloramphenicol (10 µg ml⁻¹), and tetracycline (2 µg ml⁻¹). Fresh cells grown in YEMA medium at 28°C on a rotary shaker at 150 rpm for 24 h were used for β-galactosidase assay (Miller 1992). Bacterial growth (2 ml) was centrifuged at 6000 rpm at 4°C. Cells were mixed with 1 ml Z buffer and the optical density was measured at 600 nm using a spectrophotometer. Cells were extracted by adding 100 µl chloroform and 50 µl 0.1% SDS. One milliliter aliquot was assayed at 28°C for β-galactosidase using 0.2 ml o-nitrophenyl-galactopyranoside (ONPG; 4 mg/ml). The reaction was stopped by adding 0.5 ml Na₂CO₃ (1 M). The samples were cleared by centrifugation. In the nodC:lacZ assays, the optical density was measured at 420 and 550 nm. Control bacteria that had not been exposed to flavonoids were measured against the standard buffer solution. All the nod gene expression tests were conducted in YEMA medium and repeated for three times.

Determination of the best microbial signal compounds to induce R. tibeticum nod genes

R. tibeticum was grown in the presence of 5 µM of apigenin, daidzein, genistein, hesperetin, or naringenin (Alfa Aesar, USA). β-Galactosidase activity of R. tibeticum was determined after 24 h of induction.

Determination of the optimum concentrations of signal compounds to induce R. tibeticum nod genes

R. tibeticum was grown in the presence of hesperetin at five different concentrations of 0, 5, 10, 15, 20, and 25 µM at 28°C. β-Galactosidase activity was determined after 24 h of induction.

Determination of nod gene induction with combinations of hesperetin and apigenin

Nod gene activities of R. tibeticum grown in the presence of a mixture of hesperetin and apigenin were tested at the molar concentrations of 14/1, 12/3, 10/5, 7.5/7.5, 3/12, 5/10, and 1/14 hesperetin/apigenin. Single application of hesperetin or apigenin at 15 µM was used as controls. β-Galactosidase activity was determined after 24 h of induction.

Effects of NaCl concentrations on nod gene induction

R. tibeticum was induced by 15 µM of hesperetin or mixture of hesperetin and apigenin 7.5/7.5 (molar concentrations) under different salinity stress (0, 25, 50, 75, 100, 200, 300 mM). β-Galactosidase activity was determined after 24 h of induction.

Plant culture and experimental conditions

Seeds were surface sterilized by immersing in 95% (v/v) ethanol for 10 sec, followed by a 4-min treatment with 0.5% (v/v) sodium hypochlorite and then rinsed four times with sterilized water. Seed were soaked in 50 ml of R. tibeticum (10⁷ cells/ml) or preinduced rhizobia by combination of hesperetin and apigenin pretreated R. tibeticum (7.5/7.5 µM) for 4 h. Fifteen seeds were planted per pot containing 3 kg autoclaved mixture of clay and sandy soil (1:1) and at emergence the number of plants was reduced to 10 per pot. Five levels of salinity (0, 50, 75, 100, and 200 mM NaCl) were applied during sowing. Plants were irrigated with tap water as needed until harvesting. The experiment was organized on the greenhouse bench following a randomized complete block design. Plants were harvested 45 days after planting. This time period has provided large enough samples of nodules for analysis.

Analysis

Nitrogenase activity was determined on a detached root system, using the acetylene reduction assay (ARA) of Hardy et al. (1968). Acetylene was obtained from the industrial gas company, Howamdeya, in Egypt. The purity of acetylene is 99% with impurities of methane 1%. Assays were conducted in a closed system. Roots were cut off at coteledonary nods, gently shaken to remove loose soil particles, placed in 100-ml glass bottles, sealed with a rubber septum, 10% volume of air was removed and replaced with an equal volume of acetylene. Bottles were then incubated at 28°C for 1 h. The reaction was terminated using HCl (6 N). Gas samples of 500 µl from each bottle were obtained by means of a syringe and injected into a Thermo Scientific TRACE GC Ultra (Rodano, Milan, Italy) gas chromatograph equipped with manual injector, injector loop, sample splitter, and flame ionization detector (FID) was used. Using the sample loop and splitter Capillary column CP-PoraBOND U fused silica plot 25 m×0.32 mm, df =7 µm connected to the FID. Standard curve for ethylene was developed to determine ARA activity. Afterward, nodules of each individual root were counted and nodule fresh weights were measured. Plant dry matter was determined by drying the plants at 80°C for 48 h.

Statistical analysis

Experimental data were subject to a one-way analysis of variance using a computer program (PC state). Means were compared to test significance between treatments using the least significant difference (LSD) at 0.05% probability (Rao et al. 1985).

Results and discussion

Isolation and identification

Two rhizobial isolates E5 and E11 recovered from root nodules of fenugreek were chosen for further identification using phylogenetic analysis of 16S rRNA gene sequences. A partial 16S rRNA gene sequence of 1400 base pairs of the representative isolate had a sequence with 99% similarity to R. tibeticum (EU256404.1). A phylogenetic tree was constructed from a multiple sequences alignment of 16S rRNA gene sequences (Figure 1).

Figure 1. Phylogenetic tree indicates the phylogentic relationship of the isolated rhizobia. A neighbor-joining tree was calculated using partial 16S rRNA gene sequences and a frequency filter included in the ARB (from Latin arbor, tree) software package (Ludwig et al. 2004). Bacillus subtilis (JQ653051.1) was used as out group.

Transformation of the plasmid vector containing PnodA-lacZ fusion into R. tibeticum

The pMP221 plasmid carrying PnodA-lacZ fusion was introduced into rhizobial strain (Figure 2). NodA is the first gene in nodABC operon preceded by a typical inducible nod promoter containing the conserved nod box (Spaink et al. 1987). The expression of nodA gene is positively regulated by the trans-activator NodD protein that binds to nod box and upon interaction with the flavonoid activates transcription of the nodABC operon (Hong et al. 1987).

Figure 2. Agarose gel profile of plasmid DNA preparations of (a) Escherichia coli α and (b) different transformed rhizobial isolates (2, 3, 4, 6, 7, 8, 10, 11, 12) and not transformed (1, 5, 9).

Nod gene expression test

Hesperetin, apigenin, genistein, naringenin, and daidzein at 5 µM concentrations have the ability to induce nod genes expression. The hesperetin or apigenin application resulted in a significant increase in β-galactosidase activity of R. tibeticum as indicators of nod gene expression. Hesperetin has a more pronounced effect on nod gene expression in terms of β-galactosidase activity (8286 millers units) than apagenin (6873 millers units). It is also noted that the induction of nod gene expression was recorded in the following order: genistein>naringenin>daidzein (Figure 3).

Figure 3. Effect of different signal molecules on nod gene expression of Rhizobium tibeticum. β-Galactosidase activity was determined after 24 h of induction. Means with the same letters are not significantly different at the 0.05 level using an LSD test. Vertical bars are standard error.

The optimum concentrations of exogenous application of best inducers to activate nod genes expression

The optimum concentration of hesperetin for maximum β-galactosidase activity was increased with increasing concentrations of hesperetin. Over the tested range, hesperetin application at 15 µM was the most effective in stimulating nod gene expression (Figure 4). The combination of hesperetin (7.5 µM) and apigenin (7.5 µM) significantly increased with β-galactosidase activity as compared with the single application of hesperetin or apigenin (Figure 5).

Figure 4. Effect of difference concentrations of hesperetin on the induction of nodC-lacZ fusion containing Rhizobium tibeticum. β-Galactosidase activity was determined after 24 h of induction. Means with the same letters are not significantly different at the 0.05 level using an LSD test. Vertical bars are standard error.

Figure 5. Effects of combination of hesperetin (H) and apigenin (A) in different concentration ratios (the number in parentheses was in µM) on nod gene activities of Rhizobium tibeticum after 24 h of induction in YEMA medium. Fifteen µM of hesperetin and apigenin were used as corresponding controls in this test. Means with the same letters are not significantly different at the 0.05 level using an LSD test. Vertical bars are standard error.

It was reported that R. leguminosarum bv.viciae and phaseoli was induced by naringenin, genistein, daidzein, and coumestrol (Zaat et al. 1987; Hungria et al. 1992; Dakora et al. 1993). Moreover, Begum et al. (2001) found that the optimum concentration of luteolin for maximium expression of nod genes of R. leguminosarum was at 20 µM. Peck et al. (2006) reported that luteolin has the ability to activate nod gene expression of Sinorhizobium meliloti. Zhang and Cheng (2006) identified 23 alfalfa root exudates-inducible genes of which 17 were flavonoid (apigenin)-inducible S. meliloti nod genes expression during the early root infection stage of symbiosis. Cooper (2004, 2007) concluded that nod genes from different rhizobia may respond to different sets of flavonoids. However, Guasch-Vidal et al. (2013) reported that high NaCl (300 mM) concentrations induce the nod genes of Rhizobium tropici ciat899 in the absence of flavonoid inducers.

Effect of salinity stress on nod gene expression

Nod gene expression (β-galactosidase activity) of R. tibeticum inhibited significantly as the salinity level increased. Hesperetin–apigenin combination increased β-galactosidase activity of R. tibeticum grown up to 75 mM NaCl compared to hesperetin (Figure 6). Salinity decreased the legume root hair responses to the Nod factor. Salinity can directly restrict legume root growth, affecting the responses of roots to rhizobia and the Nod factor (Tu 1981). Minor effects of osmolarity have been reported on nodD and nodABC expression in Bradyrhizobium japonicum (Wang & Stacey 1990).

Figure 6. Effects of hesperetin (H) and combinations of hesperetin (H) and apigenin (A) in concentration ratios 7.5/7.5 µM on nod gene activities of Rhizobium tibeticum after 24 h of induction in YEMA medium under salinity stress. Means with the same letters among salinity levels and with same superscript numbers among induction by mixture of hesperetin and apigenin or hesperetin alone are not significantly different at the 0.05 level using an LSD test.

Effect of preincubation of R. tibeticum with mixture of hesperetin and apigenin on nodulation and N₂ fixation and growth of fenugreek

The total number of nodules on fenugreek plants inoculated with uninduced R. tibeticum significantly reduced as salinity increased. The inhibitory effect of salt on nodulation of fenugreek appeared at the lowest salt level of 100 mM NaCl and became more inhibitory with the increase in salt concentration (Table 1). The reduction in number was 42, and 100% for 100 and 150 mM NaCl, respectively. The decrease in number of the nodules was accompanied by a decrease in the nodule fresh weight. The decrease in fresh weight of nodules was 53, and 100% for 100 and 150 mM NaCl, respectively.

Table 1. Effect of rhizobial cells preinduced by combinations of hesperetin and apigenin (7.5/7.5 µM) on nodulation and nodule function of fenugreek grown under salinity levels. Each value represents the mean of three replicates±SE.*

NaCl treatment (mM)	Nodules number plant^−1		Nodules fresh weight (g plant^−1)		Acetylene reduction (µmol C2H4 h^−1 plant^−1)
	Uninduced	Induced	Uninduced	Induced	Uninduced	Induced
0	17±0.67a^2	22±0.33a^1	0.051±0.004a^2	0.078±0.004a^1	1.8±0.15a^2	3.3±0.09a^1
50	16±0.58a^2	20±0.33a^1	0.047±0.004a^2	0.073±0.010a^1	2.0±0.20a^2	3.0±0.10a^1
100	10±0.88b^2	16±0.89b^1	0.024±0.001b^2	0.044±0.001b^1	1.2±0.11 b^2	1.7±0.87b^1
150	0c^2	11±0.89c^1	0c^2	0.024±0.0005c^1	0c^2	0.6±0.09c^1
200	0c^2	0d^1	0c^1	0c^1	0c^1	0d^1

* Means with same letter among salinity levels and with same superscript numbers among Rhizobium tibeticum induction (uninduced or induced) are not significantly different at the 0.05% level using an LSD test.

The results obtained for nitrogenase activities showed similar trends to the number and fresh weight of nodules. Nitrogenase (acetylene reduction) activity per plant (total activity) was severely depressed by salinity. The results suggest that the decrement in nitrogenase activity was due to the reducing nodule formation and nodule fresh weights per plant at 100 mM NaCl (Table 2). The inhibition of acetylene reduction by salt stress may be due to a limitation of oxygen diffusion in nodules or due to toxic effects of Na or Cl accumulation (Serraj et al. 1998). Abd-Alla (1992) reported that the depression in specific nitrogenase activity was caused by the salt reducing the protein, leghaemoglobin, and carbohydrate contents of both the cytosol and the bacteroids.

Table 2. Effect of rhizobial cells preinduced by combinations of hesperetin and apigenin (7.5/7.5 µM) on shoot and root dry mass of fenugreek grown under salinity levels. Each value represents the mean of three replicates±SE.*

NaCl treatment (mM)	Shoot dry weight (g plant^−1)			Root dry weight (g plant^−1)
	Cont	Uninduced	Induced	Cont	Uninduced	Induced
0	0.047±4.7×10^−3a^1	0.067±1.2×10^−3a^2	0.075±2.9×10^−3a^3	0.028±1.2×10^−3a^1	0.03±2.3×10^−3a^2	0.040±1.4×10^−3a^3
50	0.044±2.3×10^−3a^1	0.065±1.1×10^−3a^2	0.072±2.6×10^−3a^3	0.030±2.0×10^−3a^1	0.035±2.4×10^−3a^2	0.045±1.7×10^−3a^2
100	0.036±1.4×10^−3b^1	0.058±1.4×10^−3b^2	0.067±1.1×10^−3b^3	0.020±1.1×10^−3b^1	0.027±2.1×10^−3b^2	0.039±2.0×10^−3 b ^2
150	0.034±1.1×10^−3b^1	0.049±1.4×10^−3c^2	0.063±1.7×10^−3b^3	0.011±1.2×10^−3c^1	0.01±0.9×10^−3c^1	0.025±2.7×10^−3c^2
200	0.020±1.8×10^−3c^1	0.025±1.6×10^−3d^1	0.028±4.1×10^−3c^1	0.003±0.6×10^−3d^1	0.004±0.9×10^−3d^1	0.005±1.1×10^−3d^1

* Means with same letter among salinity levels and with same superscript numbers among Rhizobium tibeticum induction (uninduced or induced) are not significantly different at the 0.05% level using an LSD test.

Inoculation of fenugreek by R. tibeticum grown in YEMA medium containing mixture of hesperetin (7.5 µM) and apigenin (7.5 µM) significantly increased the number of nodules and mass (Table 1) of plant grown at 100 mM NaCl as compared to plants inoculated with uninduced Rhizobium. Total nodule weight per plant followed a similar pattern to number of nodule per plant. Nitrogenase activity of plants inoculated with R. tibeticum grown in mixture of hesperetin (7.5 µM) and apigenin (7.5 µM) was significantly increased as compared with plants inoculated with uninduced R. tibeticum at 100 mM NaCl. The induction of R. tibeticum with mixture of hesperetin (7.5 µM) and apigenin (7.5 µM) alleviates the adverse effect of salinity on nodulation and N₂ fixation of plants grown at 150 mM NaCl. As the salinity level increased to 200 mM, the nodule formation is completely inhibited either in plants inoculated with uninduced or induced rhizobia with mixture of hesperetin (7.5 µM) and apigenin.

The inhibitory effect of salt on the dry weight of the shoot and root was apparent at salinity levels of 100 mM NaCI. Salinity levels of 100, 150, and 200 mM caused a 24%, 28%, and 58% decrease in dry weight of shoot, respectively. The declines in nodulation and N₂ fixation in plants were translated into significant reduction in dry matter production. There are strong correlations that exist between pre-activation of rhizobia and nodule formation, and nitrogenase activity (Zhang and Smith 1995; Bandyopadhyay et al. 1996). Activated rhizobia, a combination of hesperetin and apigenin (nod gene inducer), alleviates salt stress on plant growth.

Inoculation of fenugreek with induced R. tibeticum significantly improved the dry matter accumulation as compared with plants inoculated with uninduced R. tibeticum at the salinity level of 150 mM NaCl. These results agree with the results of Yousef and Sprent (1983), Zahran and Sprent (1986), and Elsheikh and Wood (1990). Twenty mM NaCl (2.3 dSm⁻¹) inhibited nodulation of chickpea inoculated with four different Rhizobium strains by 16–20% (Lauter et al. 1981). Hafeez et al. (1988) reported that the nodulation of Vigna radiate was reduced by about half at a salinity level of 5.0 dSm⁻¹ compared to 1.4 dSm⁻¹. They also found that nodulation was completely depressed when salinity was raised to 10.0 dSm⁻¹, regardless of the plant's growth stage. In the experiments reported here, nodulation was recorded at 100 mM NaCl (14.6 dSm⁻¹). The reduction of failure in nodulation at high salinity might be attributed to shrinkage of root hairs (Tu 1981; Zahran & Sprent 1986). The processes of nodule initiation in soybean were reported to be extremely sensitive to NaCl (Abd-Alla et al. 1998). A reduction in nodule number and weight of 50% occurred with 26.6 mM NaCl (3.1 dSm⁻¹) in the rooting medium (Singleton & Bohlool 1984). It was reported that nodulation of faba bean was reduced by 24% at 50 mM NaCl (Abd-Alla 1992). Abd-Alla and Omar (1998) reported that 0.5% NaCl significantly reduced the nodulation of fenugreek and the addition of wheat straw and cellulytic fungi minimized the adverse effect of salinity on nodulation. Miransari and Smith (2007) indicated that under high levels of stress, the plant spent most of its energy inhibiting or alleviating the stress rather than developing a symbiosis with N₂-fixing bacteria. Miransari and Smith (2008) reported that genistein application to rhizobia inoculant improves plant growth through improved nodulation and N₂ fixation in both normal and salt stress conditions.

Salinity may cause alteration of root morphology and structure. In addition, under saline conditions roots are involved in regulating ion uptake, growth, and N₂ fixation (Cordovilla et al. 1994). The effect of salinity stress on reducing plant growth is through osmotic stress and the ion's toxic effects. Evidence exist that the introduction of exogenous nod gene inducers increases nodulation and N₂ fixation of some legume species. Pretreatment of B. japonicum with genistein increased nodulation and N₂ fixation of soybean and common bean (Zhang and Smith 1996; Abd-Alla 1999, 2011); and preinduction of R. leguminosarum with hesperetin and naringenin was found to stimulate nodulation and plant dry matter accumulation of pea and lentil plants (Begum et al. 2001). Flavonoid inducers act in low concentrations and the pre-activation of rhizobia used as inoculants (biofertilizers) in undoubtedly economically justified (Hassan & Mathesius 2012). Preactivation of strains might increase rhizobial competitiveness in the soil environment (Hungria & Philips 1991) Indeed, it has been shown that flavonoid preactivated B. japonicum increased soybean nodule quantity and weight (about 30%), the seasonal level of N₂ fixation (35%) and yields (10–40%) when compared to conventional inoculants (Zhang and Smith 2002). Likewise, field pea and lentil plants displayed increased nodulation and biomass production when inoculated with R. leguminosarum preinduced with hesperetin (Begum et al. 2001). Efforts to overcome the inhibitory effect of salinity on nodulation of fenugreek by preinocubation of R. tibeticum with hesperetin and apaginein were successful. The results shown here support that flavonoid have selective value in plant–microbe interaction under salinity stress. Improving legume inoculation efficiency is extremely important to improve legume production under harsh conditions.

Conclusion

The findings of this experiment indicated that salinity stress significantly inhibited the early process of nod gene expression of R. tibeticum. This inhibitory effect of salinity can be alleviated by growing R. tibeticum in YEMA supplemented with the mixture of hesperetin (7.5 µM) and apigenin (7.5 µM). As far as the authors are aware, this is the first study addressing plant receiving preactivated R. tibeticum with specific flavonoids had better nodulation, N₂ fixation, and dry matter accumulation of fenugreek under salinity stress.

Acknowledgments

We are grateful to Professor Anna Skorupska and Dominika Maj for kindly providing E. coli α which contained the expression vector pMP221. This research was financially supported by the Science and Technology Development Fund (STDF), Ministry of Higher Education and Scientific Research, Egypt, project Grant no. 12 awarded to Professor Dr Mohamed Hemida Abd-Alla. The authors are grateful to the precious comments and careful correction made by anonymous reviewers for further improvements of this manuscript.