Study on the Identification Methods for Effective Microorganisms in Commercially Available Organic Agriculture Materials

The identification of microorganisms in closely related groups is challenging. The present work focused on the different molecular methodology for the accurate microbial identification in the five commercially available organic agriculture materials enriched with effective microorganisms. From the tested five organic agricultural materials, a total of seven distinct bacterial colonies (A-1, B-1, C-1, D-1, E-1, E-2, and E-3) were isolated and processed for sequential identification utilizing HiCrome™ Bacillus agar, biochemical tests with API CHB50, 16S rRNA gene analysis, random amplified polymorphic DNA (RAPD), and species-specific PCR analysis. All the isolated microorganisms were Gram-positive rods and spore former belonging to Bacillus group and appeared as a differential characteristic feature on HiCrome™ Bacillus agar. All isolates showed high-percentage similarities with the different members of Bacillus species in biochemical testing and 16S rRNA gene analysis. The collective identification results revealed isolates, A-1, B-1, and C-1, close to B. velezensis. Further RAPD-PCR and species-specific PCR discriminated and provided confirmatory evidence for D-1 as B. thuringiensis and E-1, E-2, and E-3 as B. licheniformis, respectively. In addition, presence of B. thuringiensis was also confirmed by toxin crystal protein staining. In conclusion, the species-specific primers could be used as a rapid and accurate identification tool to discriminate closely related Bacillus species such as B. subtilis, B. licheniformis, and B. thuringiensis.


Introduction
In the past few decades, the frequent use of chemical fertilizers for agricultural purposes has increased drastically to meet the food demands of a rapidly increasing population. The excessive use of chemical fertilizers has an adverse impact on the agricultural ecosystem. Therefore, research has focused on alternatives to chemical fertilizers to improve crop productivity without affecting the soil health and agricultural ecosystem. Several microorganisms have proven their potential as suitable candidates for sustainable agricultural development. Such microorganisms have been intensely

Isolation of Bacteria from AMEM
Microorganisms from different commercial AMEM were isolated by serially diluting up to 10 −7 dilutions in 0.2% peptone water. Finally, 100 µL of each dilution was spread over the agar plates with different growth media, i.e., brain heart infusion agar (BD Biosciences, Franklin Lakes, NJ, USA), nutrient agar (BD Biosciences, Franklin Lakes, NJ, USA), and tryptic soy agar (BD Biosciences, Franklin Lakes, NJ, USA). Plates were incubated at 37 • C for 24 h to observe microbial growth. All bacterial isolates were repeatedly streaked on agar plates to obtain single-cell bacterial colonies and processed for identification utilizing different methodologies as depicted in Figure 1.

Isolation of Bacteria from AMEM
Microorganisms from different commercial AMEM were isolated by serially diluting up to 10 -7 dilutions in 0.2% peptone water. Finally, 100 µL of each dilution was spread over the agar plates with different growth media, i.e., brain heart infusion agar (BD Biosciences, Franklin Lakes, NJ, USA), nutrient agar (BD Biosciences, Franklin Lakes, NJ, USA), and tryptic soy agar (BD Biosciences, Franklin Lakes, NJ, USA). Plates were incubated at 37 °C for 24 h to observe microbial growth. All bacterial isolates were repeatedly streaked on agar plates to obtain single-cell bacterial colonies and processed for identification utilizing different methodologies as depicted in Figure 1.

Morphological Characteristics of AMEM Isolates
All the bacterial isolates from AMEM and reference bacterial strains were stored in 25% glycerol stock at −20 °C for long-term use. Bacterial isolates were studied for their morphological characteristics, such as colony color, colony shape, colony elevation, and colony texture [14], as well as microscopic characteristics, such as Gram reaction, shape of bacterial cells, and spore staining.

Differential Properties and Biochemical Identification
The differential properties of all the bacterial isolates from different AMEM were examined in

Morphological Characteristics of AMEM Isolates
All the bacterial isolates from AMEM and reference bacterial strains were stored in 25% glycerol stock at −20 • C for long-term use. Bacterial isolates were studied for their morphological characteristics, such as colony color, colony shape, colony elevation, and colony texture [14], as well as microscopic characteristics, such as Gram reaction, shape of bacterial cells, and spore staining.

Differential Properties and Biochemical Identification
The differential properties of all the bacterial isolates from different AMEM were examined in HiCrome™ Bacillus agar (HiMedia Laboratories, Mumbai, MS, India). For the biochemical analysis, the single purified colonies of the isolates were suspended in normal saline solution (0.85% NaCl) to achieve a turbidity of 2 McFarland. Later, bacterial suspension was added to the different wells of the API 50CHB, and a further process was performed according to the manufacturer's instructions (bioMerieux, Marcy I'Etoile, France). Finally, the results were analyzed using the online software apiwebTM (https://apiweb.biomerieux.com) by submitting negative and positive responses according to the reference color reading table.

16S rRNA Gene Sequencing Analysis
16S rRNA gene sequences of all the bacterial isolates were obtained by availing the commercial facility provided by SolGent Co., Ltd., Daejeon, Republic of Korea. The 16S rRNA gene region of DNA was amplified using 27F (5 -AGAGTTTGATCCTGGCTCAG-3 ) and 1492R (5 -GTTTACC TTGTTACGACTT-3 ) primers (SolGent Co., Ltd., Daejeon, Republic of Korea) using the PCR conditions: initial denaturation at 95 • C for 15 min, 30 cycles of denaturation at 95 • C for 20 s and annealing at 50 • C for 40 s, extension at 72 • C for 1 min 30 s, and final extension of 5 min at 72 • C. Finally, the amplified product was sequenced using BigDye ® Terminator v3.1 cycle sequencing kit (Thermo Fisher Scientific, Waltham, MA, USA) and analyzed on ABI PRISM 3730 XL DNA analyzer (Applied Biosystems, Foster City, CA, USA). The homology of sequences was confirmed by comparing and analyzing the base sequence of 16S rRNA gene using BLASTn.

Crystalline Protein Staining
For the identification of B. thuringiensis, crystal protein staining was performed as the method suggested by USFDA [17]. Briefly, bacterial isolate was grown on nutrient agar medium and incubated at 30 • C for 24 h, followed by 5 days of incubation at room temperature. After incubation, bacterial smear was prepared, air-dried, and fixed by gentle heating. The bacterial smear was flooded with methanol for 30 s, followed by the addition of 0.5% basic fuchsin. Sides were heated gently until steam appeared. Finally, slides were rinsed with water and visualized under the microscope (SOMETECH, Seoul, Republic of Korea) for the presence of tetragonal (diamond-shaped) toxin crystals.

Isolation of Bacteria from AMEM
From five distinctive types of AMEM, different bacterial colonies were isolated using a standard isolation method. Microscopy examination and investigation of colony morphology in solid medium and appearance in liquid medium revealed the presence of single types of bacteria in products AMEM-A, AMEM-B, AMEM-C, and AMEM-D, and that of three distinct types of bacteria in AMEM-E, which were named as A-1, B-1, C-1, and D-1, and E-1, E-2, and E-3, respectively. Gram staining followed by microscopy examination suggested that all the isolates were Gram-positive, rod-shaped bacteria (Table S3).

Differential Properties of the Bacterial Isolates on HiCrome™ Bacillus agar
Most Bacillus spp. are known as the principal biological components used in various organic agriculture materials, owing to their stress management, disease prevention, and plant growth promotion properties [11,18,19]. The AMEM-producing company highlighted the presence of Bacillus species in their products, so as the first step of identification, all the isolates were grown on chromogenic HiCrome™ Bacillus agar, a differential media for Bacillus species [20]. The bacterial isolates A-1, B-1, and C-1 were white, irregular, and wrinkle shaped with yellow/pinkish-colored media pigmentation ( Figure 2), which are the characteristic features of B. subtilis on HiCrome™ Bacillus agar medium [20]. However, B. amyloliquefaciens also displays similar culture characteristics [19]. These results provide the elementary evidence that isolates A-1, B-1, and C-1 may be B. subtilis, as supported by Alippi and Abrahamovich (2019) [20], who demonstrated the growth of 31 different strains of B. subtilis on HiCrome™ Bacillus agar and observed that 70.9% (22/31) of strains emerged as white, irregular, and wrinkle-shaped colonies with yellow-colored media pigmentation. In contrast, the bacterial isolate D-1 appeared as flat, large, and blue-colored colonies (Figure 2), which are the typical features of B. thuringiensis on HiCrome™ Bacillus agar [20], indicating that the isolate D-1 is B. thuringiensis. However, few closely related bacterial species such as B. cereus and B. mycoides also appear with similar characteristics, thus making the proper identification ambiguous.

Biochemical Characterization
The analytical profile index (API) is a method for the rapid identification of microbes up to the species level based on several biochemical tests and database searches (https://apiweb.biomerieux.com). The bacterial isolates A-1, B-1, C-1, D-1, E-1, E-2, and E-3 were tested on API CH50 CHB, which is a biochemical method conventionally used for the rapid identification of Bacillus and related genera based on biochemical examination and fermentation of 49 carbohydrates. API CH50 CHB examination identified isolates A-1, B-1, and C-1 as B. subtilis or B. amyloliquefaciens with a high percentage similarity of 99.4%, 99.7%, and 99.4% and substantial T-index of 0.85, 0.81, and 0.85, respectively (Table S4). The findings of API were similar to those of HiCrome™ Bacillus agar and failed to identify isolates A-1, B-1, and C-1 as a single species (B. subtilis or B. amyloliquefaciens). Furthermore, the API detection kit identified the isolate D-1 as a B. cereus with 96.0% similarity and 0.95 T index (Table S4). Although the API results suggest D-1 as a B. cereus, the literature suggests that B. cereus has colony morphology and biochemical characteristics similar to those of B. thuringiensis [20]. Hence, further examination is required to avoid ambiguities. Further, isolates E-1, E-2, and E-3 exhibited 99.9%, 99.7%, and 99.7%, respectively, similarity to B. liceniformis (Table S4) and showed consistency with the findings of HiCrome™ Bacillus agar, although molecular characterization is required for confirmatory identification.

Molecular Characterization Based on 16S rRNA Gene Analysis
16S rRNA gene sequencing is the most common method for bacterial identification [21] and is regarded as the most acceptable method for identifying genera and species for the isolates that do not fit properly in the biochemical profile of the existing system [21]. The strains with 97% sequence identity between the 16S rRNA genes are recognized as being the same species [12]. The isolated bacteria from AMEM were processed for 16S rRNA gene sequencing, followed by a gene bank database search (Table S5). More than 99% sequence similarity of A-1 (NCBI GenBank accession no. MW020274), B-1(NCBI GenBank accession no. MW020275), and C-1 (NCBI GenBank accession no. MW020276) isolates was observed between B. velezensis and B. subtilis (Table S5). Similarly, isolate D-1 (NCBI GenBank accession no. MW020277) displayed more than 99% sequence similarity with B. The bacterial isolates E-1, E-2, and E-3 appeared as greenish and glistering with irregular colony characteristics with greenish-yellow media pigmentation (Figure 2), a typical characteristic of B. licheniformis on HiCrome™ Bacillus agar [20]. The culture-characteristic-based appearance of E-1, E-2, and E-3 on HiCrome™ Bacillus agar provided preliminary evidence that these isolates may be B. licheniformis. However, Brevibacillus brevis and Brevi. laterosporus also exhibit similar culture characteristics on HiCrome™ Bacillus agar [20], thereby necessitating the use of an additional robust methodology for accurate identification.
Colony characteristics on the HiCrome™ Bacillus agar provided an elementary idea about the identity of the bacterial isolates from AMEM; therefore, further detailed analysis is mandatory to provide correct confirmatory identification of the bacterial isolates.

Biochemical Characterization
The analytical profile index (API) is a method for the rapid identification of microbes up to the species level based on several biochemical tests and database searches (https://apiweb.biomerieux.com). The bacterial isolates A-1, B-1, C-1, D-1, E-1, E-2, and E-3 were tested on API CH50 CHB, which is a biochemical method conventionally used for the rapid identification of Bacillus and related genera based on biochemical examination and fermentation of 49 carbohydrates. API CH50 CHB examination identified isolates A-1, B-1, and C-1 as B. subtilis or B. amyloliquefaciens with a high percentage similarity of 99.4%, 99.7%, and 99.4% and substantial T-index of 0.85, 0.81, and 0.85, respectively (Table S4). The findings of API were similar to those of HiCrome™ Bacillus agar and failed to identify isolates A-1, B-1, and C-1 as a single species (B. subtilis or B. amyloliquefaciens). Furthermore, the API detection kit identified the isolate D-1 as a B. cereus with 96.0% similarity and 0.95 T index (Table S4). Although the API results suggest D-1 as a B. cereus, the literature suggests that B. cereus has colony morphology and biochemical characteristics similar to those of B. thuringiensis [20]. Hence, further examination is required to avoid ambiguities. Further, isolates E-1, E-2, and E-3 exhibited 99.9%, 99.7%, and 99.7%, respectively, similarity to B. liceniformis (Table S4) and showed consistency with the findings of HiCrome™ Bacillus agar, although molecular characterization is required for confirmatory identification.

Molecular Characterization Based on 16S rRNA Gene Analysis
16S rRNA gene sequencing is the most common method for bacterial identification [21] and is regarded as the most acceptable method for identifying genera and species for the isolates that do not fit properly in the biochemical profile of the existing system [21]. The strains with 97% sequence identity between the 16S rRNA genes are recognized as being the same species [12]. The isolated bacteria from AMEM were processed for 16S rRNA gene sequencing, followed by a gene bank database search (Table S5). More than 99% sequence similarity of A-1 (NCBI GenBank accession no. MW020274), B-1(NCBI GenBank accession no. MW020275), and C-1 (NCBI GenBank accession no. MW020276) isolates was observed between B. velezensis and B. subtilis (Table S5). Similarly, isolate D-1 (NCBI GenBank accession no. MW020277) displayed more than 99% sequence similarity with B. proteolytic and B. thuringiensis (Table S5). The isolate E-1 (NCBI GenBank accession no. MW020278) displayed more than 97% sequence similarity with B. hyanisis and B. licheniformis (Table S5), while isolates E-2 (NCBI GenBank accession no. MW020279) and E-3 (NCBI GenBank accession no. MW020280) exhibited more than 98% sequence similarity with B. hyanisis and B. licheniformis (Table S5). The results of 16S rRNA gene analysis were ambiguous and failed to distinguish the isolates as single species. The basic limitation of 16S rRNA gene analysis is its inability to discriminate closely related groups [22]. Unfortunately, a high percentage sequence homology occurred (98.1-99.8%) between the members of Bacillus species, thus resulting in the inability of 16S rRNA gene analysis to properly distinguish between species [12,15]. In other words, it is difficult to differentiate B. subtilis, B. amyloliquefaciens, and B. licheniformis using 16S rRNA gene sequencing, owing to the high sequence similarities. Therefore, another approach is required for the appropriate identification of isolates that are closely related to each other.

RAPD-PCR and Species-Specific PCR Analysis
RAPD-PCR analysis has been found to be effective in generating strain and species-specific DNA amplification profiles that help in species identification [15]. In some cases, RAPD-PCR amplifies a unique species and strain-specific fragment that is further used to develop a specific primer for rapid and accurate identification [15]. In the present study, four random primers were examined for their RAPD pattern against the eight different bacteria of Bacillus species ( Figure S1). Based on the reproducibility and the amplification profile, primer B was selected for further RAPD analysis. The first isolates A-1, B-1, and C-1, which are supposed to be B. subtilis (as claimed by those companies and also suggested by the earlier results), were examined for RAPD-PCR using six reference strains of B. subtilis  (Figure 3a). In all the six reference strains, a strong amplification of~275 bp and~850 bp was observed and could be utilized as a dominant marker for RAPD identification of B. subtilis (Figure 3a). The RAPD patterns of isolates A-1, B-1, and C-1 did not match the RAPD profile of any reference strain, thus disproving the claim that these bacteria were B. subtilis. To further validate the RAPD-PCR results, B. subtilis-specific primers were used. First, the specificity of the species-specific primer was examined utilizing eight closely related Bacillus species (Figure 3b). The amplification of~500 bp DNA fragment only in B. subtilis (lane 2) while no amplification was observed in any of the closely related Bacillus species was indicative of the high specificity of the selected primer for B. subtilis (Figure 3b). These results are in accordance with the findings of Kwon et al. (2009) [15], who tested the same primer (ytcP) in 14 different B. subtilis strains and observed the amplification of a~500 bp DNA fragment. In addition, when the same specific primer was tested on four different strains of B. amyloliquefaciens, five different strains of B. licheniformis, and one strain of B. thuringiensis, it displayed no amplification, confirming the high specificity of the primer towards B. subtilis [15]. Knowing the high specificity of the primer (ytcP) towards B. subtilis, we tested the ytcP primer against the isolated microorganism from AMEM and observed no amplification of specific 500 bp DNA fragment (Figure 3c), confirming that these isolates are not B. subtilis, which contradicts the claims of AMEM-A, AMEM-B and AMEM-C producing companies. One of the major reasons for this mismatch is the reclassification of several Bacillus species as B. velezensis by recent taxonomy analysis [11]. Furthermore, 16S rRNA gene analysis also suggested that isolates A-1, B-1, and C-1 were the closest homologs of B. velezensis; therefore, we further performed RAPD-PCR for isolates A-1, B-1, and C-1 using B. velezensis as a reference strain. The results revealed that the reference strains, B. velezensis KCTC13012 (lane 2) and B. velezensis KCTC13417 (lane 3), and A-1, B-1, and C-1 have similar RAPD patterns (Figure 4). A dominant amplicon of~750 bp, 2000 bp, and~2500 bp was observed in all the tested reference strains and isolates A-1, B-1, and C-1, indicating that these strains were B. velezensis. However, further confirmatory studies are required to establish these strains as B. velezensis. Many recent studies suggested a high similarity between B. subtilis, B. velezensis, and B. amyloliquefaciens [11]. Even some members of B. amyloliquefaciens has been reclassified as B. velezensis [11]. Additionally, our API CH50B identification results suggested isolates A-1, B-1, and C-1 either B. subtilis or B. amyloliquefaciens, therefore, to rule out any ambiguity, we performed RAPD-PCR for isolates A-1, B-1, and C-1 using B. amyloliquefaciens as a reference strains. Next, the isolated D-1 was examined, which was claimed as B. thuringiensis by the AMEM-D company. Our HiCrome™ Bacillus agar results also suggested the isolate D as B. thuringiensis, and similarly, the 16S rRNA gene findings also displayed a high percentage similarity with B. thuringiensis. However, the API results suggested that it was B. cereus. To address this discrepancy, RAPD-PCR of isolate D-1 was performed using six reference strains of B. thuringiensis (Figure 5a). The RAPD-PCR profile suggested that all the reference strains and isolate D-1 had a dominant amplicon of~2500 bp, indicating isolate D-1 as B. thuringiensis (Figure 5a). As a final confirmatory test, B. thuringiensis-specific PCR was performed. First, the specificity of primer (XRE gene) towards B. thuringiensis was determined by performing PCR for different Bacillus species strains. The results revealed the specific amplification of the~246 bp fragment in only B. thuringiensis, suggesting high specificity of the primer towards B. thuringiensis (Figure 5b). Furthermore, the primer was tested on the isolates from AMEM, and the results demonstrated the amplification of a~246 bp fragment in isolate D-1 and confirmed it as B. thuringiensis (Figure 5c). The results are consistent with the findings of Wei et al. (2019) [16], showing the high specificity of XRE primers for identifying B. thuringiensis. Moreover, it has been well-documented that B. thuringiensis selectively produces insecticidal crystal proteins [16,23], which can be utilized as an appropriate marker for proper identification. Herein, we examined the presence of crystal proteins using crystal protein staining and observed the presence of these crystal proteins in isolate D-1 ( Figure 6). Collectively, all the results (HiCrome™ Bacillus agar, RAPD-PCR, species-specific PCR, and crystal protein staining) affirmed isolate D-1 as B. thuringiensis.
Finally, the isolates E-1, E-2, and E-3, which seem to be B. licheniformis by the earlier finding of HiCrome™ Bacillus agar, API CHB50, and 16S rRNA gene analysis, were processed for RAPD-PCR against six reference strains of B. licheniformis (Figure 7a (Figure 7a). Moreover, all the reference strains and E-1, E-2, and E-3 showed B. licheniformis-specific dominant amplification at~1500 bp (Figure 7a). The RAPD profiles of isolates E-1, E-2, and E-3 identified it as B. licheniformis. Further, as a confirmatory examination, a species-specific PCR was performed, which amplified~600 bp fragment only in B. licheniformis and not in the other closely related Bacillus species (Figure 7b). Consistent with this, isolates E-1, E-2, and E-3 amplified the B. licheniformis-specific 600 bp fragment, confirming that the isolates are B. licheniformis (Figure 7c). recent studies suggested a high similarity between B. subtilis, B. velezensis, and B. amyloliquefaciens [11]. Even some members of B. amyloliquefaciens has been reclassified as B. velezensis [11]. Additionally, our API CH50B identification results suggested isolates A-1, B-1, and C-1 either B. subtilis or B. amyloliquefaciens, therefore, to rule out any ambiguity, we performed RAPD-PCR for isolates A-1, B-1, and C-1 using B. amyloliquefaciens as a reference strains. The results revealed that the reference strains B. amyloliquefaciens KCTC1660 (lane 2), B. amyloliquefaciens KCTC1666 (lane 3), B. amyloliquefaciens KCTC3002 (lane 4), and A-1, B-1, and C-1 do not have much similarities in RAPD patterns ( Figure S2), suggesting A-1, B-1, and C-1 do not belong to the strains of B. amyloliquefaciens.   Next, the isolated D-1 was examined, which was claimed as B. thuringiensis by the AMEM-D company. Our HiCrome™ Bacillus agar results also suggested the isolate D as B. thuringiensis, and similarly, the 16S rRNA gene findings also displayed a high percentage similarity with B. thuringiensis. However, the API results suggested that it was B. cereus. To address this discrepancy, RAPD-PCR of isolate D-1 was performed using six reference strains of B. thuringiensis (Figure 5a). The RAPD-PCR profile suggested that all the reference strains and isolate D-1 had a dominant amplicon of ~2500 bp, indicating isolate D-1 as B. thuringiensis (Figure 5a). As a final confirmatory test, B. thuringiensisspecific PCR was performed. First, the specificity of primer (XRE gene) towards B. thuringiensis was determined by performing PCR for different Bacillus species strains. The results revealed the specific amplification of the ~246 bp fragment in only B. thuringiensis, suggesting high specificity of the primer towards B. thuringiensis (Figure 5b). Furthermore, the primer was tested on the isolates from AMEM, and the results demonstrated the amplification of a ~246 bp fragment in isolate D-1 and confirmed it as B. thuringiensis (Figure 5c). The results are consistent with the findings of Wei et al. (2019) [16], showing the high specificity of XRE primers for identifying B. thuringiensis. Moreover, it has been well-documented that B. thuringiensis selectively produces insecticidal crystal proteins [16,23], which can be utilized as an appropriate marker for proper identification. Herein, we examined the presence of crystal proteins using crystal protein staining and observed the presence of these crystal proteins in isolate D-1 ( Figure 6). Collectively, all the results (HiCrome™ Bacillus agar, RAPD-PCR, speciesspecific PCR, and crystal protein staining) affirmed isolate D-1 as B. thuringiensis.    (c) Species-specific PCR using XRE gene primer on microorganisms isolated from commercially available organic agriculture materials enriched with effective microorganism. Lane 1: 0.1 kb size ladder (unit: bp); Lane 2: Bacillus thuringiensis KCTC3452; Lane 3: A-1; Lane 4: B-1; Lane 5: C-1; Lane 6: D-1; Lane 7: E-1; Lane 8: E-2; Lane 9: E-3.  Finally, the isolates E-1, E-2, and E-3, which seem to be B. licheniformis by the earlier finding of HiCrome™ Bacillus agar, API CHB50, and 16S rRNA gene analysis, were processed for RAPD-PCR against six reference strains of B. licheniformis (Figure 7a) (Figure 7a). Moreover, all the reference strains and E-1, E-2, and E-3 showed B. licheniformisspecific dominant amplification at ~1500 bp (Figure 7a). The RAPD profiles of isolates E-1, E-2, and E-3 identified it as B. licheniformis. Further, as a confirmatory examination, a species-specific PCR was performed, which amplified ~600 bp fragment only in B. licheniformis and not in the other closely related Bacillus species (Figure 7b). Consistent with this, isolates E-1, E-2, and E-3 amplified the B. licheniformis-specific ~600 bp fragment, confirming that the isolates are B. licheniformis (Figure 7c).