In this study, 12 rhizospheric soil samples, four sand samples from the Błędowska Desert, one sample from Dead Sea mud in Israel, and six vegetables root samples were used for isolation of potential PGPB isolates. Rhizospheric samples of cereal and vegetable crops (viz. rye, wheat, maize, parsley, beet root, and carrot) were collected from different cities of northeastern Poland. Samples were collected in sterile plastic bags and stored at 4°C before isolation of microorganisms. One gram of soil from each sample was inoculated into LB medium at 30 ± 1°C for 5 days, then serially diluted and spread onto LB agar medium plates. After incubation, morphologically distinguishable (e.g., shape and color) colonies were transferred to fresh LB agar medium plates. Isolated bacteria were purified and stored in LB media with glycerol at 4°C for further study and at −80°C for collection. The endophytic isolates were obtained by isolation from the inner part of the vegetable root according to the method of Cochard et al. (2022) with some modifications. A detailed information on the isolates and their sample source in this study can be found in Supplementary Table S1.

The overall screening strategy was divided into two stages. The primary and secondary screening were based on different qualitative and quantitative estimation of different plant growth-promoting (PGP) attributes and biocontrol potential.

The primary screening was carried out on the basis of phosphate and potassium solubilization and auxin production. For the screening experiments, each isolate was initially inoculated in LB broth and incubated at 30 ± 1°C under agitating conditions (160 rpm) for 48 h [until the optical density (OD660) reached 0.8 to 1.2] to obtain a fresh liquid culture of bacteria. The screening experiments were performed with three replicates.

The primarily shortlisted isolates were further examined for their other PGP attributes, which includes qualitative estimation of zinc solubilization, and ACC (1-aminocyclopropane-1-carboxylate) deaminase production, and quantitative estimation of ammonia production. Biocontrol potential such as siderophore and hydrogen cyanide (HCN) synthesis and chitinolytic activity were also included in the secondary screening.

Compatibility was checked for selected bacterial isolates using agar diffusion test according Santiago et al. with some modification (Santiago et al., 2017). Freshly grown cultures were streaked vertically onto freshly prepared LB agar medium. Next, the second isolate was streaked outward at an angle of approximately 90° from the resulting colonies of the first isolate and incubated at 30 ± 1°C. After 72 h, the colony lines were observed and the presence of inhibition zones at the crossing points of the paired isolates was noted.

For Molecular identification, the genomic DNA of selected PGPB strains were isolated using Bacterial & Yeast Genomic DNA Purification Kit (EURx) following the manufacturer’s instructions. Bacterial DNA was quantified using the NanoDrop 2000 spectrophotometer (Thermo Fisher Scientific, Wilmington, United States) and by gel electrophoresis in ChemiDoc XRS+ (Bio-Rad, United States). Near full-length 16S rRNA gene sequences were amplified by polymerase chain reaction (PCR) in Eppendorf AG thermal cycler (Eppendorf, Hamburg, Germany) utilizing pair of primers: 27F (AGAGTTTGATCCTGGCTCAG) and 1492R (GGTTACCTTGTTACGACTT) (Haque et al., 2020). Amplification was done under following conditions: 5 min at 94°C, 30 s at 94°C (×30), 45 s at 57°C, elongation for 1.5 min at 72°C, and a final extension for 10 min at 72°C. The DNA quality was determined using NanoDrop 2000 Spectrophotometer and gel electrophoresis. The amplified products were purified using the PCR purification kit (EPPiC Fast reagent kit, A&A Biotechnology, Poland) following the manufacturer’s protocol. The purified PCR products were further sequenced using different 16S rRNA gene sequencing primers: 341 F (5′CCTACGGGNGGCWGCAG3′); 928F (5′TAAAACTYAAAKGAATTGACGGG3′); and 518 R (5′ATTACCGCGGCTGCTGG3′) at Eurofins Genomics, Germany GmbH. The gene sequences were further assembled with the help of BioEdit Sequence Alignment Editor v7.0.5.3 software. To identify the closest taxa, the 16S rRNA gene sequences were compared to sequences from the National Center for Biotechnology Information (NCBI; United States)¹ nucleotide nr/nt (non-redundant nucleotide) database using the BLAST algorithm. The phylogenetic analysis and construction of the phylogenetic tree was performed using MEGA7 for 33 isolates with Neighbor-Joining (NJ) algorithm. The branch quality of all generated trees was evaluated using 1,000 bootstrap replicates. The biochemical profiling of selected PGPB isolates was characterized using HiCarbo kit KB009 (Hi-Media Laboratories, Mumbai, India) following the manufacturer protocol (Pranaw et al., 2020).

Using the appropriate medium, 10 shortlisted PGPB isolates were tested for their ability to withstand various abiotic stresses, including salt, pH, and osmotic stress (drought). After incubation, the development of isolates under various abiotic stress scenarios was compared to that of isolates produced on typical nutrient agar plates at 28 ± 1°C. Results were compared to control plates/flask and categorized as negative (−) for no growth, moderately positive (++) for growth that was more but still less than the control plates/flask, and highly positive (+++) for growth that was comparable to the control plates/flask. Each assay was repeated at least twice, and all assays were done in triplicate.

Unprimed carrot seeds were selected for the seed germination assay and collected from the local market, Poland. The seed surface sterilization was performed following the method of Pandey and Gupta (2020) with minor modifications. In brief, seeds were surface sterilized with 2% sodium hypochlorite solution and 70% ethanol for 2 min and 1 min, respectively. The sterilized seeds were soaked in bacterial suspension (10⁷ CFU/mL) for 2 h, while in case of control, the seeds were soaked in ddH₂O. After soaking, the seeds were removed from the liquid and placed over autoclaved Whatman filter paper inside Petri plate. Before seeds placement, the filter paper was moistened with 5 mL of bacterial suspension, sterilized ddH₂O. Each Petri plate contained 15 seeds in 3 replicates. The plates containing carrot seeds were placed in a growth room with a daily temperature of 20 ± 1°C and incubated for 6 days, respectively. After incubation, the seeds were monitored and the germination index or germination percentage was evaluated using Equation 3 (Fahsi et al., 2021).

Further, relative seed germination with respect to control is also calculated using following formula:

Statistical analyses we performed using R version 1.2.1335 program. One-way analysis of variance was performed to evaluate the effect of bacterial isolates on carrot seed germination. Differences between means at p < 0.05 were evaluated by Tukey HSD test.

One hundred and fifty-six bacterial cultures were evaluated for their ability to solubilize inorganic phosphate and potassium. The isolates obtained were from samples from ecologically very diverse habitats. Among them, 87 strains were isolated from rhizospheric soils of different crops, and 32 and 29 isolates were obtained from samples from the rhizospheric and endophytic parts of vegetables, respectively. Only eight isolates were obtained from environments under abiotic stress conditions. Four and three isolates were obtained from the Błędowska Desert and the Dead Sea, respectively. Nearly 67 and 58% of the isolates showed the ability to solubilize phosphorus and potassium, respectively (Figure 1 and Supplementary Table S1), which is similar to other studies (Setiawati and Mutmainnah, 2016; Ashfaq et al., 2020). Isolate AF3II1 manifested the highest PSI value (3.67), followed by isolates AF1I8 and AF8I4. In the potassium solubilization test, the highest KSI values of 5.25, 3.67, and 3.60 were obtained for isolates AF3II1, AF2II2, and PC5, respectively. High PSI and KSI values were observed for the tested isolates, which is consistent with the results of other studies for other soil bacteria (Shin et al., 2015; Setiawati and Mutmainnah, 2016; Iyer et al., 2017; Sarikhani et al., 2018; Chen et al., 2020; You et al., 2020; Yaghoubi Khanghahi et al., 2021). 17.8 and 46.2% of isolates that exhibited solubilization ability had a value of PSI and KSI, respectively, that was greater than or equal to 2.0. This indicates that some of the strains tested have a high potential to solubilize phosphate and potassium. In addition, half of the isolates showed the ability to solubilize both sources of elements, and the occurrence of these traits is not related. Many of the benefits in agriculture, such as yield increases, have already been described when bacteria with the ability to solubilize phosphate and potassium were used (Liu et al., 2011; Yaghoubi Khanghahi et al., 2019; You et al., 2020).

Auxins production are one of the most important traits of bacteria that promote plant growth. It is estimated that up to 80% of microorganisms in soil communities are capable of synthesizing IAA, although the amount of hormone produced may depend on the microbial isolate (Govind et al., 2015; Maldonado et al., 2020). In the present study, quantification of IAA after 48 and 120 h showed that over 71 and 78%, respectively, of the 156 isolates tested were able to produce this hormone (Supplementary Table S1). The results obtained are in agreement with many data previously presented for different environments (Indah et al., 2017; Ashfaq et al., 2020; Lebrazi et al., 2020). The IAA concentration in the culture medium ranged from 0.12 and 0.26 to 171.18 and 215.29 μg mL⁻¹ after 48 and 120 h, respectively, in this assay. Isolates AF4II5, AF7II3, and PC3 had the highest IAA concentration of 171.18, 147.18, and 144.24 μg mL⁻¹, respectively (after 48 h of incubation). Further incubation up to 120 h resulted in the detection of IAA production in 11 additional isolates compared to the initial number of IAA producers (after 48 h). Isolates AF6I6, AF7I2, and AF6I5 had the highest IAA concentration after 120 h of incubation, 202.82, 210.12, and 215.29 μg mL⁻¹, respectively. Surprisingly, IAA concentration decreased in 27 of the isolates tested in this research after 120 h compared with 48 h. The decrease in IAA content in culture media is consistent with other studies suggesting degradation by microbial oxidase and IAA peroxidase (Raut et al., 2017; Lebrazi et al., 2020).

Based on primary screening, 33 potentially efficient isolates for plant growth promotion with several studied features were selected. In the next step, they were subjected to molecular identification by partial 16S rRNA gene sequencing of a fragment of length 1,478–1,513 bp. The NCBI GenBank accession number of sequences, closely related species, and similarity value are listed in Supplementary Table S2. Phylogenetic tree was also constructed using fragments of 16S rRNA gene sequences (Figure 2). Comparison of obtained sequences with the deposited sequenced showed that 11 genera of bacteria were present. The largest number, i.e., 12 isolates, was assigned to the genus Pseudomonas, among which the most abundant were Pseudomonas fluorescens (AF4I6, AF8I4, AF8II14, EBV2-4), Pseudomonas azotoformans (AF10I1), Pseudomonas brassicacearum (EBV2-5), Pseudomonas koreensis (AF1I7), Pseudomonas poae (EDC5), Pseudomonas protegens (ML8, ML14, ML15), and Pseudomonas putida (AF1I1). The bacteria isolated from the extremely saline environment were recognized as Oceanobacillus iheyensis (AFI) and Staphylococcus xylosus (AFII1, AFII3). The genus Bacillus was represented by isolates from soil (Bacillus cereus AF8II13), sand samples of the Błędowska Desert (Bacillus toyonensis ML12) and from endophytic parts of vegetables, i.e., carrots (B. toyonensis EDC1 and EDC6) and beets (Bacillus pseudomycoides EBV1 and B. toyonensis EBV3). Two isolates of Agrobacterium tumefaciens were isolated from the rhizosphere (A. tumefaciens DC9) and the endophytic part of the carrot root (A. tumefaciens EDC9). One isolate DC5 from carrot samples was identified as Comamonas koreensis. Brevibacterium frigoritolerans was represented by one isolate Brevibacterium frigoritolerans EBV12 from the endophytic part of beet root. Two isolates of Burkholderia ambifaria (B. ambifaria AF8II7, B. ambifaria AF8II10) were obtained from cereal rhizosphere samples. Isolate AF3II1 was classified as Klebsiella aerogenes and BV2/5 as Lysinibacillus sphaericus. Of the isolates tested, four were identified as Serratia, i.e., AF8I1 and AF8I6 are Serratia marcescens and EDC15 and EEPC5 are Serratia plymuthica DC5.

The selected isolates were further characterized based on Zn solubilization, ACC deaminase, and ammonia production. Implementation of bacteria with Zn solubilization activity allows Zn supplementation in available form in soil when needed (Nagaraju et al., 2020). Although zinc can be artificially incorporated into the soil, similar to P and K, it is immediately and almost completely fixed and immobilized, making it inaccessible to plants (Bhatt and Maheshwari, 2019; Sher et al., 2022).

Zinc solubilization ability has been evaluated for several sources of Zn: zinc oxide, zinc carbonate hydroxide basic, and zinc carbonate in plate test. Solubilization of zinc carbonate hydroxide basic and zinc carbonate was demonstrated by 70% of the isolates tested (Supplementary Table S3). O. iheyensis AF1 showed the highest solubilization ability (5.1) for zinc carbonate hydroxide basic, while isolate K. aerogenes AF3II1 revealed the highest solubilization potential for zinc carbonate at 3.33. Only 18.2% of the isolates showed solubilization activity for zinc oxide. The two isolates Serratia marcescens AF8I1 and AF8I6 showed the highest dissolution for zinc oxide with a solubilization index of 3.7.

The data show that maximum size of halo zones was slightly larger for zinc oxide, followed by ZnCO₃ and 2ZnCO₃ × 3Zn(OH)₂, which is consistent with previous reports (Dinesh et al., 2018). It is worth emphasizing that there were relatively few isolates with solubilization ability of all zinc sources used in analysis, which is comparable to the study by Bhatt and Maheshwari (2019).

The ability to convert nitrogen into a form soluble by plants is another attractive trait sought for potential plant growth-promoting microbes (Sharma et al., 2021). In the current study, all of tested isolates showed the ability to convert organic nitrogen from peptone into ammonia (Supplementary Table S3). The concentrations of ammonia produced differed greatly between isolates and time of incubation from 0.345 to 27.20 ppm. A similar trend was demonstrated in bacterial isolates from the rhizosphere of cotton and chick pea (Ahmad et al., 2008; Iyer et al., 2017). Data showed that isolates of P. fluorescens, B. ambifaria, S. xylosus, and P. brassicacearum were characterized by high ammonia production. The highest ammonia production after 24 h of incubation was observed for S. xylosus AF1 (13.5 ± 0.12), after 3 days for P. putida AF1I1 (27.2 ± 1.23 ppm) and for B. ambifaria AF8II10 (21.80 ± 0.29 ppm and 21.90 ± 0.45 ppm) on day 5 and day 7, respectively.

Plants exposed to stressful conditions produce excess ethylene which is detrimental to plant health. Plants develop many adaptive mechanisms such as ACC deaminase activity or to adapt to this new, hostile environment. The enzyme ACC deaminase (aminocyclopropane-1-carboxylic acid) produced by bacteria, enables the degradation of ACC, the main precursor of ethylene (Meena et al., 2020; Glick and Gamalero, 2021). In our study, all 33 selected isolates demonstrated ACC deaminase activity with three different levels. Almost 82% of the isolates showed moderate activity, whereas next 15% grew very rapidly on ACC as nitrogen source medium, indicating high ACC deaminase activity.

Germinating seeds are incessantly exposed to many hazardous factors, including pathogens, resulting in biotic stress (Brahim et al., 2022). The ability to produce siderophores and HCN was demonstrated by 18 and nine out of 33 isolates tested, respectively (Supplementary Table S3). Isolates tested exhibited an orange halo after a maximum of 10 days of incubation on a CAS agar plate were considered to be siderophore-producing bacteria. SPI was ranged from up to 6.17 (for P. putida AF1I1). Fourteen of the 18 isolates positives for siderophores production showed activity after only four days, while the other four isolates showed this ability only after 10 days only. For HCN production, seven isolates showed a very positive result and two demonstrated moderately positive result. Chitinolytic activity was detected in nine of 33 isolates tested with a range of CHI from 1.67 (S. marcescens AF8I6) to 2.5 (S. plymuthica EEPC5 and B. toyonensis EBV3), after 4 days of incubation. Clearing zones in CHDA medium gradually increased in most isolates. Maximum chitinolytic activity (CHI = 3.60) was shown by B. toyonensis EDC6 after 10 days of incubation. The high results obtained with isolates from the genus Bacillus are consistent with other studies (Drewnowska et al., 2020). In contrast to rhizospheric or soil isolates, the prevalence of chitinolytic activity was observed to be greater in endophytic isolates. The presence of these traits is extremely important as they have been shown to contribute to biocontrol and combat pathogenic microbes present in the environment. Encouragingly, nearly 40% of all strains exhibit at least two of the activities tested here, increasing their potential for biocontrol. The biocontrol properties tested are known to have broad activity against many pathogens (Shah et al., 2021). Therefore, selected strains can be successfully used as biocontrol agents, although none of the strains have all the tested properties simultaneously.

Based on the various above mentioned plant growth-promoting assays, 10 of the most potent bacterial isolates (Table 1) were further selected tested for compatibility, environmental adaptability by testing abiotic stress tolerance and determining the ability to use different carbon sources. The bacterial isolates were then tested for their efficiency in enhancing seed germination.

Seed germination of any plant requires optimal temperature and humidity to break exogenous dormancy. Some factors, such as hormones, can affect the rate and uniformity of mature seed germination (Begum et al., 2022). Carrot seeds, like other plants, do not germinate completely, but only 50–85%. From an economic point of view, the loss of seeds is high. Moreover, increasing the germination rate leads to higher yield and lower seed loss (Makhaye et al., 2021). In this study, the ability of 10 selected isolates to promote germination of carrot seed was investigated to select isolates with biostimulatory potential. Soaked carrot seeds were placed on Whatman paper drenched with the same bacterial suspension. The addition of the bacterial suspension was to support the colonization process and increase the number of cells adhering to the seeds. The potential to enhance seed germination was variable among the tested bacteria (Figure 3). Compared to control, S. marcescens AF8I1 significantly improved the relative seed germination by 156.88 ± 2.35%. In addition, carrot seeds inoculated with P. fluorescens AF8I4, P. putida AF1I1, K. aerogenes AF3II1, and B. cereus AF8II13 increased relative seed germination by 125.64 ± 12.1%, 116.8 ± 1.11%, 115.9 ± 2.67%, and 115.28 ± 2.56%, respectively. Furthermore, the results obtained for P. putida AF1I1 and K. aerogenes AF3II1 were not significantly different from each other, as were the results of treatment of B. cereus AF8II13 and P. fluorescens AF8I4. Similar to these studies, Rudolph et al. (2015) also showed that the use of strains identified as P. fluorescens and P. putida species can improve wheat seed germination.

Bacterial properties are well described in terms of plant growth promotion and biocontrol activity, but the importance of these attributes in enhancing seed germination is still poorly understood (Hagaggi and Mohamed, 2020; Mahdi et al., 2020; Pranaw et al., 2020). In this study, seeds responded differently to tested isolates, likely due to the individual properties and competencies of the isolates used (Majeed et al., 2015). The isolates S. marcescens AF8I1, P. putida AF1I1, K. aerogenes AF3II1, P. fluorescens AF8I4, and B. cereus AF8II13 raised seed germination between 15.28 and 56.88%. All of these isolates showed IAA production as well as phosphate and potassium solubilization ability, except for isolate B. cereus AF8II13, which in contrast showed only phosphate solubilization. The best result in this experiment was obtained with P. fluorescens AF8I1, which previously showed the ability to solubilize phosphate and a high activity in solubilizing potassium. In addition, this isolate showed ACC deaminase activity and strong IAA hormone synthesis. Several studies have shown that seed soaking by some IAA producing bacteria has a positive effect on seed germination. Maize seeds soaked in a bacterial suspension of the IAA producer Mixta theicola SAR resulted in a 27% improvement in germination (Hagaggi and Mohamed, 2020). IAA production among these strains was highly variable, i.e., S. marcescens AF8I1 showed high IAA production, while B. cereus AF8II13 did not synthesize this hormone. The remaining isolates showed IAA production ranging from 15.23 to 51.27 μg mL⁻¹. It is worth mentioning that three of the isolates used S. plymuthica EDC15, B. ambifaria AF8II10, and S. plymuthica EEPC5 showed a positive effect on seed germination, i.e., they increased germination by 8.22, 10.22, and 10.32%, respectively. All these isolates showed the ability to produce auxins. However, the result was not statistically significant. In studies by Mahdi et al. (2020) the seed germination rate of quinoa (Chenopodium quinoa) was increased by 116 and 132% after the application of Bacillus licheniformis QA1 and Enterobacter asburiae QF11, respectively. These strains showed high phosphorus solubilization activity and QF11 exhibited overproduction of auxin and QA1 of ammonia and siderophores. Inoculation of tomato seeds with Pseudomonas sp. S3 resulted in a relative germination of 111% compared to the control. This strain was positive for the production of IAA, ACC deaminase, and siderophores and was a phosphate and zinc solubilizer (Pandey and Gupta, 2020). Brahim et al. (2022) demonstrated that Bacillus pumilus MA9 and Virgibacillus halodenitrificans MA14 significantly improved wheat germination rate due to secretion of auxin, solubilization of inorganic phosphate, and ACC deaminase activity. Comparison of these reports with the present data suggests that bacterial processes such as phosphate solubilization and ACC deaminase activity may also be involved in the induction of carrot seed germination.

However, it should be noted that various studies show that only certain strains are able to promote seed germination. For this reason, many attempts have failed. Treatment of Cuscuta campestris seeds with Bacillus sp. had no significant effect on seed germination compared to the control (Saric-Krsmanovic et al., 2017). Although there is evidence that isolates of the genus Pseudomonas sp. can produce hydrocyanic acid gas, P. putida AFI1 used in the experiment did not inhibit seed germination even though it produced HCN (Makhaye et al., 2021). The effect of HCN was demonstrated with isolate ML15, which was could produce HCN, but was unable to synthesis auxins. In addition, it is worth noting that all strains used showed the ability to solubilize Zn from at least one of the sources tested. Unfortunately, there are no data linking this bacterial activity to seed germination. Isolates that showed a positive effect on seed germination were characterized by different biocontrolling properties. Most of them showed only one of the investigated properties. The exception was isolate AF8II13, which showed both chitinolytic activity and the ability to produce siderophores, and AF1I1, which showed strong production of siderophores and HCN. Data from other studies mention the occurrence of features such as siderophores production in isolates that affect seed germination, and there is no evidence of these traits for this process (Pandey and Gupta, 2020). Findings suggest that there is no relationship between the presence of biocontrol abilities and the effect on seed germination. Siderophores chelate iron this help in iron acquisition by the plants, hence the production of these compounds can be extremely helpful in the acquisition of this element by developing plant.