Plant growth-promoting native rhizobia isolated from red clover (Trifolium pratense L.) nodules: potential for metal-phytostabilization

Abstract

Background

Red clover (Trifolium pratense L.) is widely cultivated as forage crop. Due to the ability to accumulate potentially toxic elements from the soil, red clover is proposed as a bioindicator plant, but this potential is still being explored. Rhizobium leguminosarum biovar trifolii forms endosymbiosis with red clover which provides nitrogen to the plant, but also produces different plant growth-promoting compounds contributing to alleviation of different stress factors. The aim of this study was to characterize the native rhizobial strains associated with red clover plants and to evaluate in vitro the capacity of nitrogen fixation, plant-growth promotion potential and tolerance to nickel (Ni) as well as the effect of selected strains on elemental composition and growth of red clover in soil with elevated Ni concentration.

Results

More than 60% of tested strains nodulated red clover when re-inoculated in in vitro culture and induced significant increase in height, shoot dry weight and shoot nitrogen content up to 140.53%, 1026.27% and 196.85%, respectively, over the non-inoculated plants. The strains were characterized as Rhizobium sp. Plant growth-promoting traits such as indole-3-acetic acid (22.18–135.32 µg/mL in the presence of 2 g/L L-tryptophan) and 1-aminocyclopropane-1-carboxylate deaminase production and phosphate solubilization were detected in some strains. Inoculation of red clover in soil with increased Ni concentration showed that inoculated treatments had increased shoot dry weight (up to 16.13%) and nitrogen content (up to 16.21% in shoots and 18.62% in roots) compared to non-treated plants. Inoculation increased Ni, Cr and Pb concentrations (up to 52.33, 145.88 and 218.03%, respectively) compared to the control. Generally, red clover plants accumulated more metals in their roots than in shoots, and shoot concentrations were mainly within common range for plants.

Conclusion

Inoculation of red clover with effective rhizobial strains showed potential for plant growth improvement and phytostabilization of Ni in metal-burdened soils.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12866-025-04424-3.

Introduction

Leguminous plants belong to a large family of plants (Fabaceae) with great importance in human and animal food production. Trifolium spp. (clovers) contains at least 250 known, both annual or perennial species, important as fodder plants, originating from the Mediterranean region but grown in the temperate climate worldwide [1].

Red clover (Trifolium pratense L.) can improve soil quality and fertility, thanks to the endosymbiotic relationship with nitrogen-fixing bacteria (rhizobia) which enable it to fix the atmospheric nitrogen inaccessible to the plants [1, 2]. Rhizobia is a group name of heterogenous soil bacteria capable of interacting with particular legume species and forming specialized root structures called nodules, where atmospheric nitrogen is being reduced into ammonia through a process called biological nitrogen fixation (BNF). The average annual nitrogen fixation rate by legumes is around 110 kg N/ha (information available at [3]). Rhizobium leguminosarum bv. trifolii forms symbiotic relationship with Trifolium spp. plants [4]. R. leguminosarum bv. trifolii possess large multipartite genomes, characterized with significant intragenomic diversity and strains from this species contain various specific additional genes besides the „core“ genes common for all strains [5].

Negative environmental aspects of agricultural practices such as intensive use of artificial fertilizers as well as growing only monocultures have been recognized [6]. Legumes help in reducing the need for artificial nitrogen fertilizer application, contributing to more sustainable and eco-friendly agriculture. Therefore, the interest in growing legumes, including clovers, which are frequently sown as grass-clover mix for improving and maintaining the soil fertility, is growing [2].

Red clover gives high yields of protein-rich biomass suitable for hay production and grazing, if planted together with grasses, and is a quality feed for livestock [7]. Unlike alfalfa, it tolerates more acidic and wetter soils [8]. Moreover, it contains various active compounds, such as isoflavonoids with anti-inflammatory [9], antioxidant [10], estrogenic [11], neuroprotective activity [12] etc. with potential interests as pharmaceutical agent.

Trace metals normally exist in soil due to natural processes, usually present in small concentrations, with some exceptions such as serpentine soils. Anthropogenic activities such as industry, transport, agriculture etc. lead to more serious metal pollution [13, 14]. Trace metals can have numerous negative effects on soil microbial communities, physical and chemical properties of the soil, for review see [15]. Trace elements enter food chains from polluted soil, water or air, then find their way to human body where they exert acute and chronic toxic effects [16, 17]. Nickel (Ni) is an essential element in small quantities, but toxic at high concentrations, causing nutrient deficiency in plants, reduction of growth parameters, enzyme activities, and photosynthesis [18]. Currently, Ni is extensively studied due to its significant deposition in sediments across the world [19].

Recent studies indicate possibility of heavy metal resistant PGPRs (Plant Growth Promoting Rhizobacteria) to enhance the yield of agriculturally important crops with minimum or no accumulation of metal in edible part of plants, enabling safe food and feed production [20]. Metal-tolerant rhizobacteria produce many compounds that can alleviate stress in plants, modulate plant’s physiology and provide essential macronutrients to enhance tolerance to metal stress [21]. Some basic PGP traits of rhizobacteria are useful for the plant growth promotion under metal stress including: production of siderophores, phytohormones such as indole-3-acetic acid (IAA) or enzymes such as 1-aminocyclopropane-1-carboxylic acid (ACC) deaminase, nitrogen fixation, etc [20]. Besides that, rhizobacteria including rhizobia can lower bioavailability of metals through biosorption, bioaccumulation, bioprecipitation, forming complexes with metals, transform metals into less toxic forms [20, 22] or increase bioavailability [23, 24]. Data on how R. leguminosarum bv. trifolii affects the growth and metal uptake of red clover in metal-contaminated soil is scarce. Detrimental effects on survival and symbiotic potential were already noticed in white clover rhizobia grown in metal-polluted soil, where they were ineffective and less genetically diverse compared to rhizobia from similar uncontaminated soil [25]. However, other studies suggest the presence of native nodulating Rhizobium isolated from the genus Trifolium, resistant to multiple elements (As, Cd, Pb) [26, 27], indicating that more research in this area is required.

The existing data showed that red clover retains most of the accumulated metals in the roots without significant translocation to the shoots, making it promising candidate for phytostabilization. Leguminous plants are generally non-hyperaccumulating species, therefore phytostabilization is a process to be considered for the remediation of contaminated soils [28]. When cultivated in a soil from a mining area contaminated with very high concentrations of multiple elements (As, Pb, Cd and Zn) red clover accumulated several times higher concentrations of all tested elements inside their roots compared to the shoots [29]. In addition, red clover retained much greater amounts radionuclide (Cs) in the roots compared to the aboveground parts, independent of the availability of the element [30].

Considering red clover as promising plant for the cultivation in agricultural and marginal soils exposed to heavy metals, further studies of symbiosis and N fixation capacity of red clover under metal pollution are significant, as well as studies on autochthonous rhizobial strain diversity, their PGP traits and the ability for improving red clover growth and elemental composition in heavy metal contaminated soils.

The aim of this study was to explore the effects of rhizobial strains on growth parameters and metal uptake of red clover plants in soil containing high Ni concentrations. We hypothesized that rhizobial strains possessing efficient nitrogen fixation and other Plant Growth Promoting (PGP) characteristics could improve red clover growth and affect metal content in plants. Initially, isolates were tested for their capacity of nitrogen fixation in vitro, then the selected strains were tested for their PGP traits. Further, selected strains were used as inoculants in pot experiment in soil with elevated Ni concentrations.

Materials and methods

Isolation and cultivation of rhizobial strains

Isolates used in this study are either a part of the historic collection of the Institute for Soil Science, Belgrade or new isolates from this study (Supplementary Table 1). All strains were cultivated in yeast mannitol broth (YMB) at 28 °C and 150 rpm, or streaked on yeast mannitol agar (YMA) with the addition of Congo red for purity check [31]. The pure strains are kept on yeast mannitol agar slant for a year at 4 °C. The rhizobial strains were isolated from the nodules formed on the roots of red clover plants and represent indigenous population of rhizobia present in particular soil. Nodules were collected and surface sterilized using 96% ethanol for 30 s, then 0.1% HgCl₂ solution for 5 min and in the end thoroughly washed with sterile tap water [32]. Nodules were homogenized in a drop of sterile saline solution and streaked on YMA. Pure strains obtained after 2–3 days of growth at 28 °C, were tested on nodulation efficacy by inoculating red clover plants in vitro in glass tubes. Five nodulating strains, labelled as KD1, KD2a, KD3, KD4a and KD5a were isolated in this study (Supplementary Table 1).

Nodulation test

Nodulation capability of different rhizobial isolates was tested by planting red clover seeds axenically in test tubes containing 30 mL of nitrogen-free Jensen agar medium [32]. Seeds of Trifolium pratense L. cultivar K-38 selected at the Institute for Fodder Crops in Kruševac, Serbia, were used. After 7–10 days, plants were inoculated with 500 µL of bacterial culture prepared by growing bacteria for 72 h in YMB at 28 °C and 150 rpm (~ 10⁹ cells ml⁻¹, and OD600nm = 1.00). Plants were grown for 6 weeks under uniform light and 16/8 h photoperiod, and then height and root length were measured. Plant material was dried in an oven at 70 °C and shoot (SDW) and root dry weight (RDW) were measured. Nitrogen content in the dried shoots was determined using CNS Element Analyzer (Vario EL III).

Nickel tolerance of the isolates

Nickel tolerance was tested on Petri dishes, when raising concentrations of NiSO₄ × 7H₂O (0, 0.1, 0.4, 0.7, 1, 1.2 mM) were added to either YMA containing Congo red or minimal salt medium. Each isolate grown in YMB medium with culture opacity adjusted to 1 at OD600nm, was applied (20 µL) to Petri dishes in triplicates. Petri dishes containing 0 mM NiSO₄ × 7H₂O were used as control. After incubation at 28 °C for three days, the growth of bacterial colonies on Petri dishes with Ni was analysed comparing against control plates without Ni.

IAA production

Indole-3-acetic acid (IAA) production was quantified by growing bacteria in YMB medium supplemented with L-tryptophan (2 mg/mL) for 48–72 h at 28 °C and 150 rpm [33]. After incubation, bacterial cultures were centrifuged (13000 rpm) and supernatant was mixed with Salkowski reagent, containing 0.01 mM of FeCl₃ in 35% HClO₄, in ratio 1:2. The mixture was incubated at room temperature for 25 min in dark, and absorbance at 535 nm was measured. Concentration of the produced IAA was determined using standard curve with known concentrations of pure IAA.

ACC deaminase production

Production of 1-aminocyclopropane-1-carboxylate (ACC) deaminase was tested using Dworkin and Foster (DF) medium [34] supplemented with 3 mM ACC. Isolates were first grown in minimal salt medium for 72 h, then thoroughly rinsed with saline solution 3 times to remove all the medium. After rinsing, cells were resuspended in saline solution and 20 µL was inoculated on DF agar. The appearance of bacterial colonies growth observed after 7 days of incubation at 28 °C was considered positive result.

Siderophore production

Siderophore production was determined using CAS-blue agar according to [35]. Color and relative size of the color change zones of CAS-blue agar were analyzed after 7 days of incubation at 28 °C.

Phosphate solubilization

Inorganic tricalcium phosphate solubilization ability of isolates was tested on Pikovskaya’s agar [36]. The halo zone around the bacterial colonies indicated the ability of phosphate solubilization.

Rhizobial inoculation in Ni or NaCl-supplemented medium

One seed of red clover plant was planted in each test tube containing 30 mL of Jensen N-free medium supplemented with NiSO₄ × 7H₂O (0.005 mM) or NaCl (0.2%). After 7–10 days, plants were inoculated with 500 µL of bacterial culture, prepared by growing bacteria in YMB for 72 h at 28 °C and 150 rpm. For this test, three strains in total were chosen: two of the strains (407a and 449n) were randomly selected among strains with highest %N, and one of the new isolates (KD5a) which caused the highest %N in shoots in nodulation test (Table 1). Plants were grown for 6 weeks, after which shoot and root length and dry weight and nodules number were determined.

Table 1.

Root nodulation status of the 22 collection and 5 isolated strains and shoot nitrogen content (%N) of red clovers grown in N-free Jensen medium in glass test tubes

Strain name	Nodulation status	Shoot %N
402a	Nod-	1.40
402b	Nod-	1.77
404	Nod-	1.37
409a	Nod-	1.89
414	Nod-	1.90*
418a	Nod-	1.43
418aa	Nod-	1.40
420b	Nod-	2.07*
422	Nod-	1.88
446n	Nod-	1.29
407a	Nod+	2.94*
407b	Nod+	2.97*
441	Nod+	2.47*
444	Nod+	2.42*
445	Nod+	2.99*
448	Nod+	2.56*
448n	Nod+	2.60*
449	Nod+	2.46*
449n	Nod+	2.28*
450	Nod+	2.94*
459	Nod+	2.63*
KD1	Nod+	3.10*
KD2a	Nod+	3.62*
KD3	Nod+	3.32*
KD4a	Nod+	3.46*
KD5a	Nod+	3.77*
Control		1.27

*The N means followed by * are significantly different by Tukey HSD test from the value realised by control, non-inoculated plants (p ˂ 0.05), Nod- nodules are not present, Nod+ - nodules are present on the root

Genetic identification of strains

After extraction of complete DNA using protocol from [37] 16 S rRNA gene was amplified using conditions and primers of Supplementary Table 2. The sequencing services were provided by Macrogen Europe (Amsterdam, the Netherlands). The phylogenetic tree was constructed using MEGA 11 software [38].

Soil analysis

To set up the pot experiment, top soil layer (30 cm) Gleysol (Humogley according to national classification) [39], a natural meadow, with the following chemical characteristics were used. The soil was with naturally increased total Ni concentrations (106.7 mg/kg). The initial soil characteristics were: pH (H₂O) 7.05, Ntot 33.1 mg/kg, Corg 437 mg/kg, P 104 mg/kg, K 205 mg/kg, Ca 10.5 g/kg, Mg 650 mg/kg, Coarse sand 1.8%, Fine sand 20.3%, Silt 33.6, Clay 44.3, Textural class is clay. Cation exchange capacity (CEC) was 34.8 cmol/kg. The soil was air-dried, ground and sieved through 2 mm mesh. Soil pH was measured in 1 M KCl (ratio soil: KCl = 1:2.5) or H₂O, available P and K were determined by AL-method [40]. Soil content of C, N and S was determined using the elemental CNS analyser, Vario model EL III (Hanau, Germany). Granulometric composition of the soil was determined using Robinson’s pipette method [41]. Since the natural soil was used for the pot experiment, the total content of various trace elements which can be found in the soil including As, Cd, Co, Cr, Cu, Pb, Mn, Ni and Zn using Thermo iCAP 6300 Duo (ICP-OES) after acid digestion of the soil by HNO₃ [42, 43] was determined, while available forms of these elements were determined using Thermo iCAP 6300 Duo (ICP-OES) after extraction by DTPA [44]. Total and available concentrations of trace elements are presented in Table 5.

Table 5.

Trace elements concentrations in plants inoculated with the four selected rhizobial strains and non-inoculated control

Element mg/kg	Plant part	407b	449n	448n	459	Control	Soil		MAC Soil**
As	Shoot	bdl*	bdl	bdl	bdl	bdl	Total As	5.07	25
	Root	0.18	0.16	0.14	0.15	0.17	Available As	0.03
Cd	Shoot	bdl	bdl	bdl	bdl	bdl	Total Cd	0.78	3
	Root	1.88	2.62	1.99	2.41	2.17	Available Cd	0.13
Co	Shoot	0.53	0.29	0.38	0.50	0.26	Total Co	16.13	n.d
	Root	1.84	2.30	1.65	1.87	1.92	Available Co	0.05
Cr	Shoot	4.77	2.09	3.78	4.69	1.94	Total Cr	86.99	100
	Root	5.15	4.48	4.51	5.01	4.58	Available Cr	0.00
Cu	Shoot	10.39	11.34	9.53	12.72	10.81	Total Cu	37.63	100
	Root	17.83	24.00	19.22	22.77	23.03	Available Cu	3.78
Fe	Shoot	1057.9	491.4	759.4	995.9	412.7	Total Fe	n.d.	n.d.
	Root	1572.9	1390.4	1355.4	1608.4	1339.9	Available Fe	25.37
Mn	Shoot	27.6	26.4	24.9	28.0	26.3	Total Mn	426.0	n.d.
	Root	57.1	96.0	47.3	46.3	72.7	Available Mn	6.66
Ni	Shoot	5.13	3.57	4.14	4.85	3.37	Total Ni	106.72	50
	Root	11.53	11.81	10.36	11.31	11.32	Available Ni	2.37
Pb	Shoot	1.94	0.59	0.99	1.57	0.61	Total Pb	36.29	100
	Root	1.42	1.82	1.20	1.70	1.85	Available Pb	6.64
Zn	Shoot	51.0	42.2	38.9	45.3	47.0	Total Zn	96.06	300
	Root	50.3	74.6	48.0	50.3	82.9	Available Zn	1.00

Results are presented as mg/kg DW for average composite samples (n = 3)

*bdl-below detection limit, **Maximum allowed concentration for agricultural soil [48], n.d. Not determined

Pot experiment

For the pot experiment, gravel was put at the bottom of each pot, then filled with 400 g of coarsely ground soil and 450 g of finely ground soil at the top. Red clover seeds inoculated with either one of four isolates selected based on the height, SDW, RDW, %N and overall appearance of the plants from the nodulation test (407b, 448n, 449n and 459) or uninoculated (control). Bacterial inoculants were prepared by strain cultivation in YMB for 72 h at 28 °C and 150 rpm. Seeds were spread evenly on the surface of the pot then covered with a thin layer of soil. Each of five treatments had three replicates, with ten planted seeds in each pot. Pots were arranged in the completely randomized design and kept in a greenhouse, in semi-controlled conditions, with regular watering, 16/8 h day/night regime, average daily temperature during three months of 24.0 °C, and average humidity of 66.07%. No nutrients or fertilizers were added. After three months of growth plants were harvested, and height and shoot dry weight (SDW) of each plant was measured. Roots were thoroughly washed, spread over dark paper and photographed, and the photos were used for counting the total number of nodules in each pot. Root dry weight (RDW) of all the roots in each pot was measured.

Plant material analysis

Plant material collected from the pot experiment, with roots and shoots separated from each other, was dried in an oven at 65 °C for 72 h, then ground in an analytical mill to obtain fine powder and then analysed as in [45]. Shortly, the powder was used to determine the nitrogen content (%) using CNS analyzer, Vario model EL III (Elementar Analysensysteme GmbH, Hanua, Germany) and As, Cd, Co, Cr, Cu, Fe, Mn, Ni, Pb and Zn concentrations using Thermo iCAP 6300 Duo (ICP-OES), with previous digestion of the ground plant material with 65% HNO₃ and 30% H₂O₂.

Bioconcentration (BCF) and translocation (TF) factor were calculated using the following equations: BCF = trace metal concentration in plant part (root/shoot) (mg/kg DW)/total trace metal concentration in soil (mg/kg DW); TF = trace metal concentration in shoots (mg/kg DW)/trace metal concentration in roots (mg/kg DW).

Statistical analysis

Statistical analyses and visualizations were performed using Python 3 (available at [46]), employing libraries statsmodels, SciPy, seaborn and matplotlib. Analysis of variance (ANOVA) test was used to test the overall differences among the different treatments. Tukey Honest Significant Difference (HSD) test was used as a post-hoc test for pairwise comparison of the differences between group means, with a significance threshold level p < 0.05.

Results

Nodulation and nitrogen fixation potential of strains

The strains used in this study were listed in Supplementary Table 1. From 27 strains tested for nodulation ability, all the 17 strains able to form nodules with red clover significantly increased the plant height compared to the control, with the greatest height (20.06 cm) in plants inoculated with 407b (Fig. 1). Plants inoculated with strains which did not form nodules on red clover roots (nod-), had significantly lower height compared to control, with the lowest height (5.3 cm) observed in plants inoculated with 420b (Fig. 1).

Fig. 1.

Height of red clover plants when inoculated with different strains. Bars represent mean ± standard deviations (n ≥ 4). Grey-Nod-; Green-Nod+; Orange-Control: non-inoculated plants. Asterisks indicate statistically significant difference between the tested experimental group and control at p < 0.05 by Tukey HSD test

Plants with significantly greater heights also had significantly greater SDW compared to control (Fig. 2). Efficient strains improved overall appearance of the plants compared to control (Supplementary Fig. 1) after six weeks of growth. The isolates nodulating red clover were considered efficient in nitrogen fixation according to Pochons criteria [47], showing 2.5 times higher SDW values than the control plants without mineral nitrogen (control).

Fig. 2.

Shoot dry weight of red clover plants when inoculated with different strains. Bars represent mean ± standard deviations (n ≥ 4). Grey-Nod-; Green-Nod+; Orange-Control: non-inoculated plants. Asterisks indicate statistically significant difference between the tested experimental group and control at p < 0.05 by Tukey HSD test

Root dry weight was either significantly increased when inoculated with efficient strains, except for KD2a, KD3 and KD4a, or significantly reduced when inoculated with non-efficient strains, except for 418aa, compared to control plants (Fig. 3).

Fig. 3.

Root dry weight of red clover plants inoculated with different strains. Bars represent mean ± standard deviations (n ≥ 4). Grey-Nod-; Green-Nod+; Orange-Control: non-inoculated plants. Asterisks indicate statistically significant difference between the tested experimental group and control at p < 0.05 by Tukey HSD test

In plants inoculated with non-nodulating strains (Nod-), %N ranged from 1.37 to 2.07%. In plants inoculated with efficiently nodulating strains (Nod+), %N varied between 2.28% in 449n and 3.77% in KD5a (Table 1). Control plants had the lowest %N (1.27%), followed by plants inoculated with 446n, which had very similar results (1.29%).

Plant growth-promoting characteristics of strains

Three out of eight tested isolates produced siderophores, with low production in two Nod- isolates (402b and 446n) and moderate in one Nod + isolate (444) (Table 2). Eight out of 15 tested isolates could solubilize inorganic phosphates, with the greatest solubilization in 418aa, moderate in KD3 and low solubilization in KD1, KD2a, KD4a and KD5a. Five out of 15 tested isolates reached full and 2 isolates had weak growth on medium containing ACC as the only source of carbon, indicating the ACC deaminase production. ACC deaminase gene (acdS) was detected in 2 (KD1 and KD5a) out of 8 tested isolates (Table 2).

Table 2.

Plant growth promoting traits of strains

Strain name	Nodulation status	Siderophore production	Siderophore color	Phosphate solubilization	ACC deaminase activity	ACC deaminase gene
402b	Nod-	+	orange	-	+++	n.d.
418a	Nod-	n.d.	n.d.	-	+++	n.d.
418aa	Nod-	-	-	+++	-	n.d.
446n	Nod-	+	orange	-	+++	n.d.
407a	Nod+	-	-	-	-	n.d.
407b	Nod+	-	-	-	-	-
444	Nod+	++	orange	-	+++	n.d.
448n	Nod+	-	-	+	-	-
449n	Nod+	-	-	-	-	-
459	Nod+	n.d.	n.d.	+	-	-
KD1	Nod+	n.d.	n.d.	+	+	+
KD2a	Nod+	n.d.	n.d.	+	-	-
KD3	Nod+	n.d.	n.d.	++	+++	n.d.
KD4a	Nod+	n.d.	n.d.	+	+	-
KD5a	Nod+	n.d.	n.d.	+	-	+

Characteristic not present, + low production/activity, ++ medium production/activity, +++ strong production/activity, n.d. Not determined

Production of IAA was measured for the isolates used in a pot experiment (407b, 448n, 449n and 459) and the new isolates (KD1, KD2a, KD3, KD4a and KD5a) (Table 3). All the tested strains produced IAA in range from 22.18 to 100.26 µg/mL after 48 h and from 25.24 to 135.32 µg/mL after 72 h of growth in Trp-supplemented medium. The greatest production of IAA was measured in 407b and the lowest was measured in KD3.

Table 3.

IAA production of the five isolated rhizobial strains (KD) and the four collection strains used in the pot experiment

Isolate	IAA (µg/mL) ± SD (48 h)	IAA (µg/mL) ± SD (72 h)
KD1	61.87 ± 1.24	n.d.
KD2a	48.49 ± 8.54	n.d.
KD3	22.18 ± 0.14	25.24 ± 0.09
KD4a	42.08 ± 1.25	45.26 ± 0.79
KD5a	70.32 ± 3.26	n.d.
407b	100.26 ± 7.41	135.32 ± 0.03
449n	28.10 ± 6.35	n.d.
459	26.81 ± 1.59	40.02 ± 0.25
448n	62.04 ± 0.81	81.86 ± 0.73

Mean ± standard deviation (n = 4), n.d. Not determined

Phylogenetic analysis of isolates

Gene sequences of 16S rRNA from our five new isolates and two collection strains were submitted to GenBank (accession numbers PV731403.1, PV731404.1, PV731407.1, PV731406.1, PV731409.1, PV731410.1 and PV731411.1) and together with 16S rRNA sequences of the reference strains obtained from GenBank (accession numbers given in Supplementary Table 3) were used for phylogenetic tree construction using maximum likelihood method and Tamura 3-parameter model (Fig. 4). All the tested strains stood in the same group containing R. leguminosarum, R. indigoferae, R. laguerreae and R. anhuiense reference strains, indicating that 16S rRNA sequence is not sufficient for Rhizobium species determination.

Fig. 4.

Maximum likelihood tree constructed using 16S rRNA gene sequences. Bradyrhizobium japonicum LMG 6138 strain was used as an outgroup. Scale bar indicates number of nucleotide substitutions per site

Rhizobial tolerance to Ni

Strains 459, 407b, 448n could tolerate at least 1.2 mM Ni-supplemented YMA (Supplementary Table 1). Most of the tested strains could tolerate at least 0.7 mM Ni added to YMA medium and one strain (407a) could not grow in the YMA medium supplemented with 0.1 mM Ni.

Plant growth in Ni or NaCl-supplemented medium

When grown in Ni-supplemented Jensen N-free medium, the three tested isolates (407b, 449n and KD5a) significantly increased the height and SDW of red clover plants compared to the control. Number of nodules was also significantly higher in inoculated plants compared to control, which did not form nodules. Root parameters were more variable, but positive effects of inoculation were also observed as increased length or RDW (Table 4).

Table 4.

Growth parameters of red clover plants grown in Ni or NaCl-supplemented Jensen N-free medium

Plant parameter		407b	449n	KD5a	Control
0.005 mM NiSO 4	Plant height (cm)	13.6 ± 3.4 *	12.1 ± 4.0 **	10.0 ± 5.6 *	4.3 ± 1.5
	Root length (cm)	16.78 ± 4.56 **	11.44 ± 5.18 *	10.36 ± 5.37	9.95 ± 4.23
	No of nodules	8.00 ± 3.04 **	8.13 ± 5.06 **	12.50 ± 8.85 **	0
	SDW (mg/plant)	17 ± 6 **	16 ± 10 **	13 ± 10 *	4 ± 3
	RDW (mg/plant)	6 ± 2 **	7 ± 5	6 ± 4 *	3 ± 2
0.2% NaCl	Plant height (cm)	14.60 ± 1.47*	13.49 ± 3.02**	15.00 ± 1.82*	5.52 ± 1.01
	Root length (cm)	20.83 ± 3.66**	16.73 ± 6.02	26.50 ± 5.86**	11.20 ± 3.39
	No of nodules	14.86 ± 5.59**	21.14 ± 8.46**	15.40 ± 8.21**	0
	SDW (mg/plant)	27 ± 10**	25 ± 12**	30 ± 9**	4 ± 2
	RDW (mg/plant)	11 ± 4**	14 ± 5**	11 ± 5**	4 ± 2

Mean ± standard deviation (n = 10). Asterisks indicate statistically significant difference between the tested experimental group and control at * p < 0.05, **p < 0.01

When grown in NaCl-supplemented Jensen N-free medium (0.2% compared to normal 0.01% present in Jensen medium), plants inoculated with the three tested isolates (407b, 449n and KD5a), had significantly greater values for all growth parameters (height, root length, nodules number, SDW, RDW) compared to control (Table 4).

Pot experiment in Ni-burdened soil

Tendency in increase of SDW when inoculated with isolates 449n and 448n by 9.68% and 16.13%, respectively, compared to control was observed (Fig. 5a). This increase was not significant (p > 0.05). Nitrogen content in the shoots showed tendency to be higher in all inoculated plants compared to control (Fig. 5a). The greatest %N (2.94%) was measured in plants treated with 459, followed by 407b (2.90%), 448n (2.65%), 449n (2.60%) and control (2.53%). However, the greatest content of N, when calculated as g of N per plant, was measured in plants inoculated with 448n (9.5 mg/plant). Control plants had lower percentage of N in their roots (2.47%) compared to all inoculated treatments, while the greatest values were detected in 449n (2.93%) followed by 407b (2.75%), 459 (2.65%) and 448n (2.59%). Number of nodules was higher in all inoculated treatments, with significant difference compared to control in plants treated with 459, 449n and 448n (Fig. 5b). Nodules in inoculated plants were located on the higher (closer to the shoots) parts of the roots, and were not present on the lower parts (Fig. 5c).

Fig. 5.

Plant growth parameters under inoculation in soil. a Shoot dry weight and nitrogen content per plant; b Average number of nodules per pot; c Roots of each experimental group (selected photos). Orange arrows point to the positions of the nodules. Bars represent mean ± standard deviations (n = 3). Control: non-inoculated plants. Asterisks indicate statistically significant difference between the tested experimental group and control at p < 0.05 by Tukey HSD test

In the soil used, concentrations of all trace elements were below the maximum allowed concentrations for agricultural soil [48], except for Ni, which was increased more than twice compared to the MAC (Table 5).

Trace element analysis (Table 5) showed that in red clover, Ni concentrations in plants ranged from 3.37 to 5.13 mg/kg DW in the shoots, with the highest values in 407b and lowest in control. In the roots, Ni concentrations were much higher than in shoots ranging from 10.36 to 11.87 mg/kg DW (in treatment 448n and 449n, respectively).

The As and Cd were detected only in the roots of all the treatment groups, while in shoots their levels were below the detection limits. Concentrations of almost all other tested trace elements including Co, Cr, Cu and Mn were also higher in roots in all treatments. The largest difference in concentration between roots and shoots was detected for Co which had from 3.5 to 8.1 times higher concentrations in roots compared to shoots, depending on the treatment. The only exceptions were Pb and Zn in plants treated with 407b, which had higher shoot than root concentrations, and consequently translocation factor (TF) greater than 1. Bioconcentration factor (BCF) values were smaller than 1 for all tested metals in all experimental groups except for Cd in roots, which was greater than 1 in all experimental groups, with the greatest value of 3.353 in the 449n group (Table 6).

Table 6.

Bioconcentration (BCF) and translocation (TF) factors for each metal in plants inoculated with the four selected rhizobial strains and non-inoculated control plants

		407b	449n	448n	459	Control
As	BCF root	0.036	0.032	0.028	0.030	0.034
	BCF shoot	n.d.	n.d.	n.d.	n.d.	n.d.
	TF	n.d.	n.d.	n.d.	n.d.	n.d.
	BCF root	2.404	3.353	2.551	3.090	2.782
Cd	BCF shoot	n.d.	n.d.	n.d.	n.d.	n.d.
	TF	n.d.	n.d.	n.d.	n.d.	n.d.
	BCF root	0.114	0.142	0.102	0.116	0.119
Co	BCF shoot	0.033	0.018	0.024	0.031	0.016
	TF	0.285	0.124	0.231	0.268	0.135
	BCF root	0.059	0.052	0.052	0.058	0.053
Cr	BCF shoot	0.055	0.024	0.043	0.054	0.022
	TF	0.926	0.467	0.839	0.936	0.423
	BCF root	0.474	0.638	0.511	0.605	0.612
Cu	BCF shoot	0.276	0.301	0.253	0.338	0.287
	TF	0.583	0.473	0.496	0.559	0.469
	BCF root	n.d.	n.d.	n.d.	n.d.	n.d.
Fe	BCF shoot	n.d.	n.d.	n.d.	n.d.	n.d.
	TF	0.673	0.353	0.560	0.619	0.308
	BCF root	0.134	0.225	0.111	0.109	0.171
Mn	BCF shoot	0.065	0.062	0.059	0.066	0.062
	TF	0.482	0.275	0.527	0.604	0.362
	BCF root	0.108	0.111	0.097	0.106	0.106
Ni	BCF shoot	0.048	0.033	0.039	0.045	0.032
	TF	0.445	0.302	0.400	0.428	0.298
	BCF root	0.039	0.050	0.033	0.047	0.051
Pb	BCF shoot	0.053	0.016	0.027	0.043	0.017
	TF	1.362	0.325	0.825	0.921	0.332
	BCF root	0.524	0.776	0.500	0.524	0.863
Zn	BCF shoot	0.531	0.439	0.405	0.472	0.489
	TF	1.015	0.565	0.811	0.900	0.567

*n.d. Not determined

Discussion

Characterization of rhizobia associated with red clover root nodules

Legumes play important role in agroecological systems and stand in line with the sustainability principles by increasing production level due to the nitrogen-fixing ability and lowering greenhouse gas emission [49]. Given the environmental, ecological, social and economic benefits, legumes are grown worldwide [50].

Rhizobia-legume symbiosis and nitrogen fixation efficiency depend on numerous abiotic factors present in the soil such as salinity, pH, temperature, moisture as well as mineral composition, for review see [51]. Therefore, the studies on rhizobial native diversity are important to find both nitrogen efficient and stress-tolerant native isolates [52, 53].

In this study, we aimed to evaluate the efficiency of bacteria–plant interactions under conditions of elevated metal content. We selected the most nitrogen-fixing efficient isolates in in vitro conditions and determined the PGP traits of strains including phosphate solubilization, IAA production, ACC deaminase activity and siderophore production. The selected isolates were used as inoculants for a pot experiment in soil naturally containing elevated concentration of Ni, but below the remediation limit.

More than 62.9% of isolates showed the ability to form nodules on the roots of red clover. The selected nodulating strains were identified as Rhizobium sp. The genetic diversity and phylogeny of rhizobia in different regions showed the presence of R. leguminosarum bv. trifolii and vicia in root nodules of red clover [54–56]. The non-nodulating strains from our study could be considered as nodule endophytes, which, according to their cell and colony morphology, are different from Rhizobium genera (personal observation). It is widely known that the bacteria unable to form root nodules were detected within the root nodules of red clover, such as Rhodococcus, Bacillus, Pseudomonas spp. and designated as root nodule endophytes [57, 58].

In our study, strains of Rhizobium with high nitrogen fixation efficiency were observed. All nodulating strains showed at least two times higher or more SDW compared to control indicating their highly efficient N fixation potential. In addition, the nitrogen percentage was from 2.28% up to 3.77% which is similar to previous studies [59]. Previously, R. leguminosarum bv. trifolii strains isolated from red clovers from two climatic zones showed a high diversity in symbiotic efficiency and differences between the populations and among the strains within the individual populations were detected [59]. Other studies also point out the necessity to select strains showing high biological nitrogen fixation potential with a wide range of clover species in order to inoculate seed mixtures in natural or cultivated pastures [60].

The presence of metals in excessive concentrations in the environment is detrimental and even lethal to all the living organisms. It is widely known that trace metals cause disturbances of soil microbial communities by changing community structure [61], diversity and abundance of microbial taxa [62]. Nickel can significantly reduce the number of nodules in alfalfa [63]. Therefore, identification and selection of rhizobial strains capable of forming efficient symbiosis with legumes under specific conditions, such as presence of elevated metal concentrations in soil, is important when developing bacterial inoculants as natural biofertilizers. The R. leguminosarum species is relatively sensitive to heavy metals including Ni (up to 0.5 mM) [64], and in our work the strains could tolerate up to 1.2 mM Ni. Usually, highly tolerant strains are mostly isolated form soils with higher metal concentrations, like serpentine soils in the case of nickel [53].

Plant growth-promoting traits vary among different bacteria and also among the strains of the same species [65, 66]. In general, the PGP characteristics contribute to the alleviation of different environmental stress in plants, including metal stress. Bacteria possessing multiple PGP traits such as IAA production, phosphate and zinc solubilization, siderophore, ACC deaminase, exopolysaccharide and extracellular enzymes production can improve growth and alleviate toxicity caused by metals [67]. IAA is a plant hormone which can improve plant survival in stressful conditions caused by metal stress, but positive cell protective effects, as improvement of the redox status and activity of antioxidant enzymes, were shown to depend on the dose of IAA applied, where low dose was beneficial and high dose was detrimental [68]. According to [69], the highest production of IAA was 5.33 µg/mL for strains isolated from red clover nodules, while our strains had significant IAA production ranging from 22.18 to 135.32 µg/mL. However, the amount of tryptophan added to the medium for bacterial growth was many times higher in our study, 2 g/L, compared to 50 mg/L in [69]. Compared to results for other nodule symbionts e.g. Sinorhizobium meliloti isolates from alfalfa nodules [63], R. leguminosarum bv. trifolii strains had moderate IAA production.

The 1-aminocyclopropane-1-carboxylate (ACC) deaminase is an enzyme produced by rhizospheric and endophytic bacteria which breaks down the ethylene precursor (ACC) into α-ketobutyrate and ammonia, therefore lowers the production of ethylene, which enhances plant tolerance to various stressful conditions including metal stress, drought etc [70, 71]. Transgenic tomato expressing bacterial ACC deaminase gene accumulated higher amounts of metals and improved resistance to metals compared to normal plants [72]. A gene encoding ACC deaminase, acdS, was found in rhizobial species including R. leguminosarum biovars [73]. Treatment of Phaseolus vulgaris L. with ACC deaminase-producing R. leguminosarum bv. phaseoli together with added melatonin, improved salt stress tolerance by regulating the ion homeostasis and enhancing nitrogen fixation and photosynthetic rate [74]. Although our tested strains (449n, 407b and KD5a) did not possess ACC deaminase activity, they significantly improved growth parameters of red clover plants compared to untreated control, when grown in medium with elevated NaCl concentration (20 times higher than in normal Jensen’s medium) as well as in medium supplemented with 0.005 mM NiSO₄, despite these parameters were lower compared to the plants inoculated with the same strains in normal Jensen’s medium.

Phosphate solubilization is considered as a general mechanism of stress alleviation increasing availability of phosphorous [75, 76], while siderophore production may facilitate metal binding and detoxification of heavy metals [77]. In our study, most of the nodulating strains did not produce siderophores while weak to moderate phosphate solubilization potential was detected in 58% of nodulating strains.

The impact of strains on red clover growth and trace metal content in Ni-burdened soil

In our research, in controlled conditions N-free medium with Ni addition, rhizobial strains increased the height and SDW of red clover plants compared to uninoculated plants. Further, in the pot experiment, plants were in good condition under all treatments without any symptoms of Ni or other heavy metal toxicity [18]. Two treatments (449n and 448n) increased SDW compared to the control, with the highest SDW observed in treatment 448n. All applied bacterial inocula increased nodule number, as well as %N in red clover shoots and roots compared to control, with the increase of %N in shoots ranging from 3 to 16%.

In previous experiments, inoculation of Pongamia pinnata with R. pisi and Ochrobacterium pseudogrignonense increased shoot length, root length, and biomass in the presence of Ni [78, 79]. Inoculation of pea with Rhizobium sp. RP5 enhanced plant dry matter, nodule numbers, root N, shoot N, leghemoglobin, seed yield, and grain protein (GP) in Ni-amended soil, compared to non-inoculated plants [80]. In soybean, an increase of seed yield was detected after inoculation with effective strains of Bradyrhizobium spp. in soils with increased Ni concentrations [37]. In red clover grown in strongly acid soil (pH 4.38), inoculation with R. leguminosarum bv. trifolii was proved to significantly increase the shoot dry weight and N, Mg, Fe, Mn, and Cu contents in plants [81].

Rhizobia can help phytoremediation and also mitigate stress in legumes. Different rhizobial strains were tested for their effects on metal uptake by legumes, where both lower [80] and higher [82] accumulation were observed, as compared to non-inoculated plants. The Bradyrhizobium inoculation increased critical toxic and effective concentration of Cu which reduced shoot and root dry weight by 50%, lowered oxidative stress in inoculated plants, and also increased Cu values in plants, which were higher in roots than in shoots, in both white lupin and soybean plants [83]. In another study, the Bradyrhizobium strains decreased the Ni concentrations in soybean seeds [37].

Different studies showed that red clover plants translocate very small amounts of metals to the aerial parts and retain most of the accumulated metals in the roots [28–30], which is useful to stabilize metals in the soil and limit extraction of metals from the soil into the aboveground parts. However, very limited information on how rhizobia affect metal uptake in red clover plants is available.

The effects of our strains on metal concentrations in shoots of red clover plants depended on the strain and metal tested. For example, higher accumulation of Ni, Cr and Co was observed in all inoculated experimental groups compared to control, while for Cu, Mn, Pb and Zn, different treatments either lowered or increased the shoot concentrations. In general, the concentrations of all trace metals in all treatments in shoots (with exception of Cr) were among the normal (sufficient values) for plants according to [84].

Concentrations of Ni we measured in the shoots of plants inoculated with three different strains as well as uninoculated plants (3.37–4.85 mg/kg DW) were within the normal values (0.1–5.1 mg/kg DW) according to [84]. Shoot concentrations of Ni in plants inoculated with 407b were slightly above the normal (5.13 mg/kg DW), but within the tolerable level of Ni (5–10 mg/kg DW), according to [84].

Concentrations of Cr in shoots of red clover plants were above the sufficient or normal levels (0.1–0.5 mg/kg) in mature leaves [84], while in some of our experimental groups, including 407b and 459, it was near the excessive or toxic level (5–30 mg/kg). This could be due to the relatively high values of Cr in soil (86.99 mg/kg) although being below the MAC.

In the study by [85], red clover plants collected from several urban areas in the same city, and the concentrations of Ni and Cr in leaves ranged depending on the type of urban soil from 1.33 to 15.3 mg/kg and from 3.48 to 52.4 mg/kg, respectively, which is higher compared to the concentrations we measured in shoots of red clover, even though, the soil concentrations of these elements were higher in the soil we used for pot experiment compared to the urban soil from [85].

Bioconcentration factor (BCF) is used to assess the plant’s ability to accumulate metals and values larger than 1 indicate potential hyperaccumulation of the metal. It can be calculated for each part (root, stem, leaves) of the plant separately [86]. Values we obtained, were less than 1 for all tested metals in shoots and roots of all experimental groups, except for BCF for Cd in the roots, which was larger than 1 in all experimental groups. Multiple times higher concentrations of Cd in roots compared to the soil concentrations and no Cd detected in shoots, indicates the potential of red clover to accumulate Cd in their roots without its translocation to the shoots.

Concentrations of Cd that we measured in the roots of red clover, ranging from 1.88 to 2.62 mg/kg, were higher than values observed in red clover plants sampled from several urban areas with different soil metal content, ranging from 0.084 to 0.498 mg/kg [85].

While there was a significant correlation between the red clover aboveground tissue concentration, which ranged between 0.044 and 0.073 mg/kg, and the total soil Cd concentration [87], concentration of Cd in aboveground mass of white clover, collected from different locations in New Zealand, was not correlated to the soil total Cd, and the highest concentration was 0.24 mg/kg in the soil containing 0.5 mg/kg Cd [88].

In another study, T. pratense emerged as a potential soil pollution bioindicator, as it had BCF >1 for Zn, Ni and Cd as well as TF >1 for Ni and Zn at various levels of soil contamination [89]. Very high concentrations of all tested metals were measured in this study [89], but the soil was artificially contaminated by adding available nitrate forms of metals, unlike the soil used in our study, which was natural and contains all elements within the allowed range for agricultural soil [48], except Ni which is above the MAC but below the remediation values.

Our results indicate the potential of red clover plants to accumulate metals in the roots with low translocation to the shoots of red clover plants, indicating the potential of red clover for being used in phytostabilization of Ni and other metals in the soil.

Conclusions

This study reveals native nitrogen-fixing and nickel tolerant Rhizobium sp. strains associated with red clover, which possess multiple PGP traits including IAA, ACC deaminase, siderophores and phosphate solubilization ability, indicating that there are abundant functional microbial resources in soil environment for further improvement of red clover plant tolerance in soils with increased trace metals concentrations. In soil with increased concentration of Ni, inoculation with four nitrogen-fixing efficient strains influenced increase in nitrogen content and overall growth compared to uninoculated control plants. The effects on metal uptake by red clover plants depended on the strain, however the applied strains enhanced the levels of Ni, Cr and Co in the shoots as well as Ni in roots of inoculated plants. Most of the metals were stored in the roots with small rate of translocation to the shoots of red clover plants, indicating the stabilization of metals in the root zone without excessive extraction to aerial parts. Since more Ni is accumulated in the roots of inoculated compared to uninoculated plants, we suggest that rhizobial strains could aid in phytostabilization of Ni in the soil. Further studies regarding the use of rhizobial inoculants in metal-burdened soil are necessary, especially in open-field conditions.

Authors’ contributions

MP: Conceptualization, Data curation, Formal analysis, Investigation, Visualization, Writing - original draft; OSS: Conceptualization, Formal analysis, Investigation, Resources, Supervision, Writing – review & editing; STJ, BS, VM, DD, NK: Investigation. All authors proofread the manuscript. All authors read and approved the final manuscript.

Funding

This research was supported by the Ministry of Science, Technological Development and Innovations of the Republic of Serbia, contract No. 451-03-136/2025-03/200011.

Data availability

Sequence data obtained in this study were deposited in the NCBI repository (available at https://www.ncbi.nlm.nih.gov/) under accession numbers: PV731403.1, PV731404.1, PV731407.1, PV731406.1, PV731409.1, PV731410.1 and PV731411.1.

Declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.