Samples of Rhynchosia legume were collected from four different locations in the central and southern semiarid regions of the Province of La Pampa, Argentina (Table 1).

Edaphic characterization was performed based on texture, phosphorus and nitrogen contents among the macronutrients. The Robinson method [10] was used to determine texture. Nitrogen was evaluated using the Kjeldahl method [11], while phosphorus was determined by the Bray-Kurtz N° 1 method [12].

Forage capacity was evaluated based on both the total and aerial biomass by direct weight. Neutral detergent fiber (NDF) and acid detergent fiber (ADF) were determined using the sequential Goering and Van Soest method [13]. Dry matter digestibility (DMD) and metabolizable energy (ME) were calculated using Eqs. (1) and (2):(1) DMD=88.9×(0.779×%ADF) (2) ME=3.6×DMD

Protein content was measured by the Kjeldahl method [11].

Root segments bearing nodules were washed free of soil under running water, placed in a fine-mesh steel holder, and surface sterilized, as previously described with slight modifications [14]. Briefly, the nodules were sterilized by immersion in 70% ethanol for 1 min, followed by vigorous shaking in 5% sodium hypochlorite containing 0.01% Tween 80, and finally washed seven times with sterile distilled water. To verify correct surface sterilization, samples corresponding to the last wash water were seeded in Petri dishes with TY medium. Surface-sterilized root nodules were transferred to empty sterile plastic dishes and cut in half with a flamed scalpel. They were then squashed in 50–150 ml of sterile physiological saline solution, the volume varying in proportion to nodule size. The squashed nodule suspension was plated on TY agar (5 g/L tryptone, 3 g/L yeast extract, and 0.87 g/L CaCl₂·2H₂O). Plates were incubated at 28°C for up to 2 weeks. Each isolate strain was named according to the sampling area in La Pampa, Argentina (T = Toay; V = Victorica; S = Santa Rosa; R = Rucanelo).

Rhynchosia seeds were surface sterilized by immersion in 70% ethanol for 30 s, followed by stirring in 2% sodium hypochlorite for 4 min, and rinsed in seven changes of sterile deionized water. Germinated seedlings were aseptically transferred to pots containing sterile carrier sand-vermiculite 1:2 mixtures, fed from the bottom, and watered regularly with sterilized distilled water containing nitrogen-free Fahraeus solution. Bacteria were inoculated by dispensing a 1 ml suspension of liquid culture in a TY broth medium (10⁸ cells/ml). The assay was maintained in a greenhouse programmed for a 16-h daylight photoperiod at 23°C and 60% constant relative humidity. Plants were examined for nodule formation after 60 days. They were removed from the pots, the root systems were washed with water and the number of nodules per plant was determined. The experiment had a randomized complete block design with four replicates per tested isolate and one seedling per replicate. Three independent assays were performed. Pots were rotated every 4 days to avoid any border effects.

To identify isolates with the ability to form nodules, the nodC gene was amplified using the primers nodCF (5′-TGATYGAYATGGARTAYTGGCT-3′)/nodCR (5′-CGYGACAGCCANTCKCTATTG-3′) [15]. Colony polymerase chain reaction (colony PCR) was performed in a 20 µl reaction mixture containing 0.5 µM of each primer, 1X PCR buffer, 1.5 mM MgCl₂, 0.04 mMdNTPs (Promega), and 1 U of Taq DNA polymerase (TransGen Biotech). To add the DNA, a small amount of bacterial colony was taken directly from the Petri dish using a sterile tip for each colony and placed in a PCR tube containing the master mix. The temperature profile was as follows: 95°C for 3 min; 35 cycles at 95°C for 1 min, 55°C for 1 min, 72°C for 2 min, and finally 72°C for 10 min. PCR amplification was performed using a thermal cycler (Ivema T21). PCR products were separated on 1.5% (w/v) agarose gels by horizontal electrophoresis and visualized using SYBR Safe staining (Invitrogen). The Ensifer meliloti B399 strain was used as the positive control.

Liquid cell cultures were performed under different osmolarity and pH conditions. TY was used as the culture medium. Specific growth rate (μ) was determined in each culture and used as a kinetic parameter to characterize cell growth. The following treatments were used:

-Control treatments: Cultures were grown at pH 7 without NaCl.

-Stress for pH treatments: Different cultures were grown at pH 6, 8, and 9.

-Stress for osmolarity treatments: Cultures were grown under different salinity conditions as described by Abdelmoumen et al. [19], as follows: 50, 100, and 200 mM NaCl.

These cultures were kept under optimal temperature and agitation conditions (constant temperature of 28°C and 200 rpm, respectively) to determine growth under stress conditions.

Genotypic characterization of the 17 selected isolates was performed using BOX-PCR (repetitive sequence based-PCR), as previously reported by Versalovic et al. [20], with the primers BOX-AR1 (5′-CTACGGCAAGGCGACGCTGACG-3′). The fingerprints obtained were analyzed using the GelCompar II program (Applied Maths BV). The degree of similarity between them was determined using Pearson’s correlation coefficient. Dendrograms were constructed from the distance matrix using the unweighted pair group method with the arithmetic mean (UPGMA) algorithm.

Rhynchosia isolates were grouped according to the degree of similarity of their BOX-PCR banding patterns. A 90% cut-off value was considered. The Shannon et al. [21] index was applied to determine the diversity index (H) and estimated using Eq. (3):

(3) H′=−∑ pi log2 pi where pi is the relative abundance of isolates of each OTU.

Data were analyzed using INFOSTAT software [22]. Variables were analyzed with ANOVA, and a comparison of means was performed using Fisher’s protected test (i.e., LSD), with a significance level of 0.05.

Legume samples of the genus Rhynchosia were collected mainly from sandy loam-like soils (Table 2). Their main characteristic was the low amounts of nitrogen (N) and phosphorus (P) detected. Thus, alternative strategies to the use of agrochemicals could be developed to overcome the poor nutritional status of the soil. Moreover, the pH values detected in these soils showed a remarkable alkaline tendency, with calcium carbonate stones on the surface and in the hardened soil layer near the surface. These facts were further considered when establishing the conditions for cell culture at different pH levels.

According to the results shown in Table 3, native Rhynchosia legumes could be well utilized for forage cultivation. It presented remarkably good biomass yields. Under cultivation, the biomass produced can be much higher than that observed (58–161 g/plant), as indicated by Porta Siota et al. [23].

Moreover, its nodulation capacity, determined by the symbiotic relationship established with rhizobacteria, could improve forage values through nitrogen fixation. Among the selected legume sampling areas, Rucanelo and Victorica showed the highest forage performances, as observed in the biomass values. Native legumes collected from Rucanelo had the best protein content.

Forage capacity results were similar to those obtained for standard legume crops, such as alfalfa (16%–20% protein and 58%–65% digestibility values) [24]. In some cases, lower DMD values were recorded for native Rhynchosia, probably due to the higher proportion of stems compared with leaves, which increases fiber values.

The root systems of all field-collected species bore nodules that varied in number and shape. Mature Rhynchosia nodules were small and elongated. They also showed indeterminate growth, consistent with the morphology described by other authors for Rhynchosia schimperi [25]. A total of 54 bacterial strains were isolated from inside the surface-sterilized nodules, from which liquid cultures were developed. Root nodule endophytic bacteria associated with legumes of the genus Rhynchosia have not been studied in detail.

Garau et al. [26] described the nodules of several plants of the papilionoid legume Rhynchosia ferulifolia and reported the presence of species of the genus Burkholderia inside the nodules. The nodulation test performed in this study revealed that the strains isolated from nodules of Rhynchosia were unable to nodulate their original hosts effectively at the experiment times evaluated here. Subsequently, nodC genes were amplified to evaluate whether the isolates had the ability to nodulate. All isolates were negative for the amplification of this gene. In Fig. 1, the agarose gel is shown with the PCR product of five of the isolates. This indicates, therefore, that endophytic bacteria also establish a symbiotic relationship with the host plant as they exhibit growth-promoting properties. This result is consistent with other studies, in which no cultivable nodulating rhizobia was found [9]. The bacteria that cohabit in these Rhynchosia nodules belong to different taxonomic groups, which could be selected from some mechanism generated by the plant or by the rhizobium that initiates the nodule. This allows endophytic colonization based on the ability of the bacteria to cooperate in the difficult task of surviving in hostile environments such as the Pampean semiarid region [27].

Based on the results of the stress conditions, isolates were grouped according to their growth kinetics behavior under the pH and osmolarity levels tested. Four groups were defined for pH stress, as follows:

GROUP 1: Isolates were not affected under the pH stress conditions evaluated: R1, R2, R3, R4, R5, R7, R8, R9, R10, T22, T25, T26, T27, T31, T35, T63, T77, S40, S54, S56, S58, S59, S68, S78, V11, V45, V48, V50, V51, V52, V62, V73, V74, V75, V76 and V77 (Fig. 3a).

GROUP 2: Little or no cell growth at pH 9: T28, T32, T34, T36, T70, S37, S41, S43, V16 and V61 (Fig. 3b).

GROUP 3: Isolates developed a “lag” stage of adaptation to pH 9 after which the levels of cellular concentration of the other conditions were equated: T66, T67, S55, and V60 (Fig. 3c).

GROUP 4: Isolate growth kinetics and/or cell concentration was negatively affected by alkaline pH levels: R6, T65, S39 and V12 (Fig. 3d).

Regarding osmolarity stress conditions, isolates were grouped as follows:

GROUP 1: Isolates were not affected by the different osmotic conditions: R1, R2, R3, R4, R5, R7, R8, R9, R10, T22, T26, T27, T28, T31, T32, T35, T63, T66, T67, T70, S37, S39, S40, S41, S43, S54, S55, S58, S59, S68, S78, V11, V12, V45, V48, V50, V52, V60, V73, V74, V75, V76 and V77 (Fig. 4a).

GROUP 2: Isolates showed little or no cell growth in low and high osmolarity conditions (0 and 200 mM NaCl, respectively): T25 and T77 (Fig. 4b).

GROUP 3: Isolates developed a “lag” stage of adaptation with conditions of high osmolarity after which the cell concentration levels of the other conditions were equated: T36, T65, V16, V51, V61 and V62 (Fig. 4c).

GROUP 4: Isolates in which an increase in osmolarity favored cell growth: R6 and S56 (Fig. 4d).

To represent the behavior of each strain in relation with its response to the stress conditions, an index was calculated by considering the relationship between the specific growth rate (µ) obtained in each treatment and that observed in the control experiment. Analyzing these relationships (Table 5), we found that the isolates whose relationship values were 1 ± 0.05 overcame the corresponding stress condition and reached the same levels of cell concentration as the control treatment.

Of the 54 isolates subjected to different stress conditions, 57.41% were not affected concerning the cell concentration reached and growth kinetics (highlighted in black and marked with * in Table 5). The high tolerance of the strains to these conditions could be attributed to the fact that they were isolated from soils with pH values tending to alkalinity (Table 2). Based on these values, it could be inferred that they were saline soils since they exhibited pH values between approximately 7.5 and 8. Choudhury et al. [40] found variability in the behavior of microorganisms when they were exposed to variations in osmolarity and pH. In agreement with our results, Karagöz et al. [41] found isolates with plant growth-promoting abilities (N₂ fixers, phosphate solubilizers, and siderophore producers) that were tolerant to high concentrations of NaCl in acidic and alkaline rhizospheric soils. The multiplicity and redundancy of homeostatic mechanisms are tags of stress conditions in bacteria [42].

The results show promise for restoring native vegetation on stressed soils and enhancing restoration benefits through plant-microorganism interactions. Öğütçü et al. [43] found that salinity adversely affected chickpea growth parameters, but inoculation with native strains improved its ability to thrive under saline conditions. Therefore, understanding native microbial flora and its adaptability is crucial for optimizing this association. Field experiments with plant growth-promoting rhizobacteria from stressed areas should be conducted [44].

Ventorino et al. [45] found that the reference strains used were less effective and competitive than the natural population and that no native isolate belonged to the reference strain. Nautiyal et al. [46] reported that phosphate solubilization activity decreased gradually with increasing concentration of NaCl, which emphasizes the importance of working with species that tolerate these conditions. Sachdev et al. [47] reported that the production of IAA by native strains under salt stress conditions correlated positively with greater root growth, thus contributing significantly to improving the plant tolerance to salt. Lebrazi et al. [48] also highlighted that IAA-producing rhizobacteria could be harnessed to improve plant growth.

To carry out genetic characterization analysis, 17 microorganisms were selected based on their ability to tolerate the studied stress conditions and considering whether they exhibited two or more PGPR properties. The dendrogram based on BOX fingerprint patterns (Fig. 5) showed that the genetic diversity of isolates occupying Rhynchosia nodules was high (H = 2.08) [3] although the selected isolates shared certain characteristics. Considering a similarity coefficient of 90%, the BOX patterns could be grouped into nine different operational taxonomic units (OTUs). Only 42 out of 17 isolates had the same genetic profile (V73 and V74). These results confirm the expected high level of diversity.

Based on these results, future trials are planned to select isolates that can be used in the formulation of biofertilizers, through inoculation tests of Rhynchosia plants. In this way, the ability to promote plant growth will be evaluated in vivo, by subjecting the plants to the different stress conditions studied in this work. This will allow a strict selection of strains suitable for biotechnological purposes, on which analyses will be carried out for identification, based on the 16S gene. Although these genomic approaches are of great taxonomic importance, the genes identified remain to be phenotypically characterized [1]. Other authors have characterized and grouped isolates from Rhynchosia nodules in the genus Burkholderia [49].