NodC-based evaluation of the occurrence of bacteria in nodules of Pisum sativum isolated on YEM agar

Bacterial strains were isolated from nodules of two varieties of field pea (Pisum sativum L.)—“Tarchalska” and “Klif”—cultivated in the experimental field of Bayer® company located in Modzurów, Silesia, Poland (N50°9'24" E18°7'55"). A randomized system was adopted in the conducted field experiment, and it consisted of four replicates; each plot had a size of 8.4 m². The experimental field soil was an Umbrisol, a slightly degraded chernozem, formed from loess, classified as good wheat complex and the second bonitation class; therefore, the soil conditions were sufficient for the pea needs. The topsoil had a neutral pH (pH in 1 mol/dm³ KCl: 6.28) and a high nutrient content: 191 ppm P₂O₅, 217 ppm K₂O, and 101 ppm Mg. The following pre-sowing doses were applied: 48 kg ha⁻¹ of phosphorus (P₂O₅) and 72 kg ha⁻¹ potassium (K₂O). Ammonium nitrate was applied as a “starting dose” with 20 kg N ha⁻¹. Both varieties were inoculated before sowing with either of the two inoculants containing Rhizobium leguminosarum bv. viceae: Nitragine^™ produced by the BIOFOOD company (Poland) and noncommercial gel inoculant produced by the Polish IUNG. Non-inoculated plants were used as control. The plant seeds were sown in the first week of April 2011 and in the fourth week of March 2012. One hundred and twenty germinable seeds of “Tarchalska” and of 100 germinable seeds of “Klif ” variety were sown per square meter at a row distance of 15 cm and sowing depth of 6 cm. Four types of nodules (large red, large cream, small red, and small cream) were selected for the study and were collected in July 2011 and 2012.

From each plant sampled, six pea nodules of each type (large red, large cream, small red, and small cream) were randomly excised. In total, 24 samples were used for further analyzes in each year. The collected nodules were washed with sterile water and surface sterilized using 1% NaOCl for 2 min and then with 70% ethanol for 1 min and washed with sterile water for four times. After sterilization, each nodule (one nodule per each type in six replicates) was crushed with a sterile glass rod in 1 cm³ of sterile 0.85% NaCl solution and shaked for 15 min in a Petri dish. Afterwards, the bacteria were isolated on yeast extract mannitol (YEM) agar plates. The isolates were grown at 28°C for 5 days and purified by repeated streaking on YEM (Vincent, 1970). Following incubation, the numbers of Rhizobium-like bacteria CFUs per one nodule of each type and each plant–inoculant combination were determined.

Moreover, bacterial isolates were derived from the two Rhizobium-based inoculants used in this study by streaking serial dilutions onto YEM agar plates. A reference strain of R. leguminosarum bv. viceae (DSM 30143) was also used in this study.

The bacterial isolates were cultured on tryptone yeast agar plates at 28°C and genomic DNA was extracted by following the protocol of Genomic Mini kit for genomic bacterial DNA isolation (A&A Biotechnology, Poland).

The species identification of bacterial isolates was based on the sequencing of the 16S rRNA gene. The partial fragment of 16S rRNA gene was amplified using the primers fD1 (5’-AGAGTTTGATCCTGGCTCAG-3’) and rP2 (5’-AGAGTTTGATCCTGGCTCAG-3’) (Weisburg et al., 1991). Polymerase chain reaction (PCR) solution contained 50 ng of DNA template, 12.5 pM of each primer, 2.5 mM of dNTP, 1X PCR buffer, and 2.5 U Taq DNA polymerase in a total volume of 25 μl. The following temperature profile was used for DNA amplification: initial denaturation at 94°C for 5 min followed by 30 cycles of denaturation at 94°C for 30 s, annealing at 62.5^oC for 30 s, and elongation at 72°C for 1 min; 15 cycles of denaturation at 94°C, annealing at 48°C, and elongation at 72°C for 1 min; and final elongation at 72°C for 10 min. PCR amplifications were performed using Veriti 96-Well Thermal Cycler (Applied Biosystems, USA). The PCR products were visualized by 1×TBE electrophoresis in ethidium-bromide-stained, 1% agarose gel and purified using the Gel Out purification kit (A&A Biotechnology, Poland) before sequencing. Sequencing reactions were performed on 24 samples of 16S rRNA amplicons—one sample per each of the combination. The obtained nucleotide sequences were queried against the NCBI GenBank database using BLAST search to determine the most closely related bacterial species.

Five primer pairs were tested for amplification of approximately 930 bp of the 1,300 bp nodC gene: forward primers used were nodCF, nodCFu, nodCF2, nodCF4, or nodCFn, and reverse primer used were nodCI (Laguerre et al, 2001) (Table 1).

PCR solution contained 50 ng of DNA template, 12.5 pM of each primer, 2.5 mM of dNTP, 1X PCR buffer, and 2.5 U Taq DNA polymerase in a total volume of 25 μl. The following temperature profile was used for DNA amplification: initial denaturation at 94°C for 5 min followed by 30 cycles of denaturation at 94°C for 30 s, annealing at 62.5°C for 30 s, and elongation at 72°C for 1 min; 15 cycles of denaturation at 94°C, annealing at 48°C, and elongation at 72°C for 1 min; and final elongation at 72°C for 10 min. PCR amplifications were performed using Veriti 96-Well Thermal Cycler (Applied Biosystems, USA). The PCR products were visualized by 1×TBE electrophoresis in ethidium-bromidestained, 1% agarose gel and purified using the Gel Out purification kit (A&A Biotechnology, Poland) before sequencing. The part of nodC gene of the selected isolates was sequenced and queried against NCBI GenBank database using BLAST search to confirm their species affiliation.

The PCR amplified, 930-bp long, nodC products were subjected to the restriction fragment length polymorphism (RFLP) analysis by digesting with five restriction endonucleases AluI, BamHI, BsuRI, EcoRI, and MspI following the manufacturer’s instructions (Thermo Fisher Scientific-Fermentas, Canada). This procedure was aimed to determine the molecular diversity of rhizobial strains isolated from the analyzed pea plants. Each digestion reaction was conducted individually with one restriction enzyme at a time. After digestion, resulting fragments were visualized by 1×TBE electrophoresis in ethidium-bromide-stained, 1% agarose gel. The GeneRuler^™ 1kb DNA Ladder (Thermo Scientific) DNA size marker was used to determine the size of the bands obtained as a result of the restriction digestion. Similarly, as in the case of the restriction reactions, the results of restriction reactions were assessed in the form of one electrophoresis per one enzyme. Therefore, five different electrophoreses were prepared to analyze the RFLP results.

Comparison of means and least significant difference (LSD) for the CFUs of bacteria colonizing pea nodules was performed using STATISTICA v. 10 software (Stat-Soft Inc., Tulsa, USA).

The numbers of CFUs per one nodule varied significantly (Table 2). The isolates formed colonies that were visible after 5–7 days. All isolates produced colorless to whitish convex colonies with abundant extracellular gum on the surface. Of all examined variants, the highest number of colonies of nodule-colonizing bacteria was observed for the large red nodules of the IUNG-inoculated “Klif” variety of pea (310 CFU/nodule). In general, this variant showed the highest mean number of bacterial colonies per nodule (175.9 CFU/nodule). On the other hand, the smallest number of CFUs per nodule (CFU/nodule) was observed for the uninoculated “Tarchalska” pea plants. The smallest mean number per nodule was observed for the combination of “Tarchalska” plants inoculated with Nitragine^™ (46.0 CFU/nodule). When considering the differences in the numbers of CFUs per nodules, the highest mean abundance was observed for the large red nodules (132.7 CFU/nodule), as it was nearly twice higher than that in the case of small red and nearly three times higher than in large cream nodules, where the mean number of CFUs per nodule was the smallest (53.2 CFU/nodule).

Considering the plant cultivar–inoculant combination, the LSD test showed that the number of bacterial CFUs isolated from nodules of “Klif ” variety combined with IUNG inoculant was significantly higher than the number of colonies isolated from other variants (p<0.05) (Table 2).

In order to verify the identity of isolated strains, the 16S rRNA gene was amplified and sequenced. The experiment involved DNA samples extracted from bacterial isolates derived from all 24 experimental variants and isolated from inoculants as well as from a R. leguminosarum reference strain (DSM 30143). Amplification of the 16S rRNA region produced a single sharp band of the expected size (1,500 bp).

Sequencing of the product revealed, however, that despite careful sterilization of nodules and using a Rhizobium-selective medium sequences of other bacterial species were mostly detected. Sequences of seven samples (apart from R. leguminosarum bv. viceae DSM 30143) could only be analyzed, as the rest of sequencing patterns presumably resulted from multiple templates that were amplified. Among them only one sequence belonged to R. leguminosarum bv. viceae (Table 3). Such result may indicate that the pea nodules as well as the Nitragine^™ inoculant were dominated by bacterial species other than R. leguminosarum.

To eliminate the Rhizobium-negative samples, the isolates were tested for the presence of nodC gene. This test also confirmed the theoretical ability of strains to form nodules with roots. This experiment was conducted on 46 bacterial strains isolated from the YEM agar plates. These were from one to four isolates per one experimental variant and one strain from each inoculant (IUNG and Nitragine) along with the reference strain of R. leguminosarum. The expected 930-bp region of 1,300-bp long nodC gene was observed only in the case of 22 of 46 samples (21 bacterial isolates from nodules and 1 from the IUNG inoculant; Table 4). The remaining isolates, including the reference strain and the bacteria isolated from a commercial Nitragine^™ inoculant, lacked the expected product. The comparison of the obtained nodC sequences with the NCBI showed that all samples resulted in positive identification of R. leguminosarum bv. viceae.

The RFLP analysis using five restriction enzymes (AluI, BamHI, BsuRI, EcoRI, and MspI) provided conclusive results only in the case of BamHI digestion; therefore, the results obtained only for this enzyme are shown (Figure 1). This reaction resulted in the digestion of a 930-bp long PCR product (nodC gene) into two or three fragments of various sizes, revealing two patterns (A and B). In the pattern A, the digestion process gave two fragments, 780 and 150 bp, whereas in the pattern B, the digestion resulted in three fragments, 560, 220, and 150 bp. In some cases, the digestion revealed a mixed profile (A + B), that is, four bands were observed 780, 560, 220, and 150 bp, indicating that in those cases two profiles were combined in the same sample.

Figure 1