Genomic Screening of Nitrogen-Fixing Nostocales Cyanobacteria Reveals Predicted Traits for Soil Fertility and Plant Growth Promotion

Abstract

Background: The urgent need for sustainable agricultural drives the search for effective microbial biostimulants. Cyanobacteria of the order Nostocales are promising candidates due to their nitrogen-fixing capabilities and bioactive secondary metabolites. However, the genomic potential of many soil strains from microorganism collection remains largely unexplored. Methods: We performed a targeted genomic screening of five cyanobacterial strains from the All-Russian Collection of Microorganisms (VKM): Nostoc commune VKM Al-35, Nostoc punctiforme VKM Al-37, Nostoc minutum VKM Al-168, Anabaena pirinica VKM Al-153, and Hassallia pseudoramosissima VKM Al-158. The workflow involved WGS, de novo assembly, and comparative metabolic profiling using KEGG, SEED, PLaBAse, antiSMASH, and RhizoSMASH to identify predicted plant growth-promoting (PGP) traits, biosynthetic gene clusters (BGCs), and rhizosphere competence mechanisms. Biosafety was evaluated via Comprehensive Antibiotic Resistance Database (CARD) and in silico toxomics screening. Results: High-quality genome assemblies were obtained for all strains (completeness > 99%). Functional annotation uncovered complete genetic machinery for nitrogen fixation, predicted phosphate mobilization, and phytohormone biosynthesis pathways. Comparative analysis revealed two distinct genomic strategies: a versatile support profile in Nostoc strains (expanded genomes and diverse accessory pathways) and a specialized stimulation profile in Anabaena and Hassallia strains (focused phytohormone pathways). Comprehensive CARD and antiSMASH screenings demonstrated an excellent biosafety profile, confirming the complete absence of regulated cyanotoxin clusters or acquired antibiotic resistance genes of clinical concern. Conclusions: This genome-based bioprospecting serves as a cost-effective pre-selection filter, providing a strong scientific rationale for downstream experimental validation of these strains. The presence of predicted gibberellin biosynthesis pathways and T6SS/T4SS secretion systems in H. pseudoramosissima VKM Al-158 represents a notable genomic feature among soil cyanobacteria. The identified genomic prerequisites suggest that these strains possess strong predictive potential for future development as safe biological resources for sustainable agriculture.

1. Introduction

Modern agriculture faces a fundamental challenge because food production must continuously increase for a growing global population, which is projected to reach 9.6 billion by 2050 [1]. This growth requires a 50–70% increase in crop yields while simultaneously minimizing environmental impact [1,2]. Such expansion is possible either through the use of new agricultural lands or by increasing the productivity of already cultivated ones. However, the intensive application of mineral fertilizers and pesticides, along with intensive tillage and excessive irrigation, has led to serious consequences, including the deterioration of soil fertility, depletion of soil and water resources, and an increase in the cost of agricultural products. According to the Food and Agriculture Organization (FAO), approximately 75% of global soils are degraded to some extent due to erosion, salinization, loss of organic matter, acidification, and disruption of soil microbiome structure [1]. In particular, the excessive use of nitrogen fertilizers contributes to emissions of the potent greenhouse gas nitrous oxide (N₂O). Agricultural soils account for approximately 60% of global anthropogenic N₂O emissions, while intensive management practices trigger soil acidification and the subsequent decline of natural fertility [3]. Moreover, nitrogen use efficiency (NUE) remains notably low in most agricultural systems. Cereal crops such as rice, wheat, and maize typically recover less than 50% of applied nitrogen, while the remainder is lost through gaseous emissions, leaching, and runoff. The environmental and energetic costs associated with synthetic nitrogen fertilizers are increasingly recognized as unsustainable [3]. In this context, developing environmentally safe and cost-effective alternatives or complements to synthetic agrochemicals has become a primary scientific priority. The intensification of sustainable agriculture is essential to address global food supply demands while minimizing the environmental impact of conventional agrochemicals, which are often associated with soil degradation and toxicity [4]. The use of beneficial microbial inoculants as biopesticides and biofertilizers is a sustainable strategy to improve crop yield and disease resistance. These biological approaches offer multiple mechanisms to improve NUE, such as enhancing root architecture, promoting nitrogen fixation, modulating phytohormone signaling, and facilitating nutrient mobilization in the rhizosphere [3]. Plant growth-promoting bacteria (PGPB) colonize plant tissues and support plant health by enhancing nutrient uptake and stress tolerance through nitrogen fixation, phytohormone production, and siderophore synthesis [5]. Among these microorganisms, cyanobacteria deserve special attention as an ancient and extremely diverse group of oxygenic photoautotrophic bacteria that are natural inhabitants of soil ecosystems worldwide [6]. Cyanobacteria are recognized as the evolutionary ancestors of plant chloroplasts and have played a crucial role in the evolution of biodiversity on Earth [7,8]. Their long-standing evolutionary history is the reason for their success in acclimatization and sustenance in diverse habitats [7]. Unlike heterotrophic rhizobacteria, cyanobacteria do not require organic carbon. They act instead as primary producers that initiate soil-forming processes, which makes them ideal ecosystem engineers capable of supporting soil fertility even under adverse conditions [1]. Their functional role is associated with the ability to perform photosynthesis, nitrogen fixation, and the solubilization of phosphorus [9]. Additionally, they produce extracellular polysaccharides (EPS) that enhance soil particle aggregation, possess water-retaining properties, and adsorb nutrients [9]. Current research shows that cyanobacteria, especially representatives of the order Nostocales, possess a unique set of properties important for sustainable agriculture. These are filamentous, heterocytous cyanobacteria capable of biological nitrogen fixation (BNF) in specialized cells called heterocytes [8]. They represent a substantial fraction of nitrogenase-containing microorganisms, highlighting their importance in global nitrogen cycling [10]. Recent evidence indicates that members of the Nostocales order demonstrate a unique capacity to increase in relative abundance along the soil–root continuum, specifically targeting the rhizoplane and endosphere [11]. In agroecosystems, biological soil crusts provide nitrogen fixation rates ranging from 4.9 to 8.1 kg N ha⁻¹ year⁻¹ in perennial cropping systems [12]. Nitrogen fixation by cyanobacteria can lead to a minimum of 20–25% savings of nitrogenous fertilizers, besides a 10–12% increase in yields [13]. Beyond nitrogen fixation, Nostocales are known for their multifaceted PGP traits, including nutrient mobilization and the secretion of biostimulatory metabolites [14,15]. These cyanobacteria are known to produce more than 700–800 diverse bioactive secondary metabolites, including alkaloids, phytohormones, vitamins, and siderophores [13,16]. Furthermore, they produce specialized secondary metabolites such as scytonemin and mycosporine-like amino acids (MAAs), which provide an advantage for colonizing UV-exposed soil environments [17]. Despite this enormous potential, commercial bioproducts based on soil cyanobacteria are extremely rare on the market. Genomic studies indicate that terrestrial cyanobacteria of the order Nostocales generally possess larger genomes and an expanded repertoire of genes related to regulation and stress response compared to their aquatic relatives [18]. These findings highlight both the high adaptive capacity and the vast, yet underexplored, biotechnological potential of these microorganisms. A scientific contradiction remains evident because the clear demand for novel eco-friendly technologies stands in contrast to the largely untapped gene pool of soil cyanobacteria preserved in culture collections. Whole-genome sequencing (WGS) and functional annotation allow for the comprehensive characterization of agronomically valuable traits, making the bioprospecting of collection strains a critical first step in identifying candidates with optimal agronomic potential.

Therefore, the aim of this study was a targeted genomic screening of five cyanobacterial strains of the order Nostocales from the All-Russian Collection of Microorganisms (VKM) to assess their potential for developing bioproducts aimed at enhancing soil fertility and stimulating plant growth.

2. Materials and Methods

2.1. Cyanobacterial Strains and Culture Conditions

Five cyanobacterial strains of the order Nostocales were obtained from the All-Russian Collection of Microorganisms (VKM, Pushchino, Russia): VKM Al-35 (Nostoc commune), VKM Al-37 (Nostoc punctiforme), VKM Al-153 (Anabaena pirinica), VKM Al-158 (Hassallia pseudoramosissima), and VKM Al-168 (Nostoc minutum). The strains were originally isolated from various soil types in southern Russia: VKM Al-35 and VKM Al-37 from solonetz soil (Volgograd region), VKM Al-168 from chestnut soils (Volgograd region), VKM Al-153 from meadow soil (Volgograd region), and VKM Al-158 from brown semi-desert soil (Republic of Kalmykia).

All strains were cultivated in liquid and agar-solidified BG-11 medium [19]. Cultures were maintained at 23–25 °C under LED illumination with a photosynthetic photon flux density of 60–75 μmol·m⁻²·s⁻¹ and a 12 h photoperiod. Morphological features were documented using a Leica DM750 light microscope equipped with a Flexacam C3 digital camera (Leica Microsystems, Wetzlar, Germany). Captured microphotographs were used to illustrate the studied objects.

2.2. DNA Extraction, Sequencing, and Quality Control

DNA library preparation, including enzymatic fragmentation, adapter ligation, and amplification, was performed using the VAHTS Universal Plus DNA Library Prep Kit for Illumina V2 (Vazyme, Nanjing, China) according to the manufacturer’s protocol. Library quality was verified using a Qsep400 system with the High Sensitivity Cartridge Kit (BiOptic, New Taipei City, Taiwan) and a Qubit 4 Fluorometer with the Spectra Q HS kit (Raissol Bio, Moscow, Russia). WGS (PE150 mode) was conducted on a SURFSeq5000 platform (GeneMind, Shenzhen, China) at the Medgen Company (Moscow, Russia). Raw read quality was assessed using FastQC v0.11.9, including evaluation of Q-score distribution, GC content, adapter contamination, and duplication levels. Quality trimming and adapter removal were performed with Trimmomatic v0.39 using the SURFseq-specific adapter sequences provided by Medgen and the following parameters: ILLUMINACLIP:SURFseq_adapters.fa:2:30:10, AVGQUAL:25, and MINLEN:75. To account for the non-axenic nature of the cyanobacterial cultures, high-throughput WGS was performed in a deep metagenomic mode. For each unicyanobacterial culture, sequencing yielded between 17.99 and 23.61 million paired-end reads (PE150), corresponding to a substantial data volume of 5.40 to 7.08 Gb of raw data per strain. Given the estimated genome sizes of the target cyanobacteria (6.3–9.4 Mbp), this deep dataset provided an exceptional effective sequencing coverage depth exceeding 400× to 700× for the target cyanobacterial genomes after bioinformatic filtering. This highly redundant coverage depth ensured robust de novo metagenomic assembly and high-accuracy functional profiling.

2.3. Genome Assembly, Quality Assessment, and Decontamination

To eliminate potential host-contaminant signals arising from the non-axenic nature of the cyanobacterial cultures, a rigorous bioinformatic decontamination and filtering pipeline was implemented. De novo metagenomic assembly of the deep sequencing reads was executed using SPAdes v3.15.5 [20] with the --meta option for metagenomic assembly and automatic k-mer selection (--k auto). Contigs shorter than 1000 bp were removed using seqkit v2.9.0. To ensure taxonomic purity and strictly isolate the target cyanobacterial genomes from associated satellite bacteria, automated genome binning was conducted using MaxBin2 v2.2.7 [21]. Potential non-cyanobacterial contamination within the bins was further identified and systematically removed using Kraken2 [22] against the comprehensive PlusPFP database. To ensure maximum purity, a positive selection workflow was implemented using the KrakenTools suite (github.com). Specifically, the extract_kraken_reads.py script was utilized to exclusively retain contigs assigned to the phylum Cyanobacteriota (NCBI TaxID: 1117) and its descendants, according to the following command:

python extract_kraken_reads.py -k sample.kraken.output -s sample.fasta -t 1117 --include-children -o sample_clean.fasta

As a result, all non-target sequences (including other heterotrophic bacterial phyla, viruses, eukaryotes, and unclassified contigs) were systematically excluded from the bins. This post-assembly filtering strategy was preferred over pre-assembly read trimming because Kraken2 classification accuracy decreases on short primary read fragments; thus, pre-assembly filtering could introduce false-positive classifications, leading to the accidental loss of genuine cyanobacterial data and degrading the de novo assembly quality. Finally, genome completeness and contamination levels of the refined cyanobacterial bins were rigorously verified using CheckM2 v1.1.0 [23]. Only high-quality bins explicitly assigned to the target cyanobacterial taxa, demonstrating high completeness (>98%) and minimal contamination (<4.5%), were retained for downstream functional profiling and BGC annotations. General assembly metrics (N50, L50, total genome size, and contig count) were evaluated using QUAST v5.2.0 [24].

2.4. Phylogenomic Analysis

The phylogenetic position and taxonomic identity of the studied strains were established using a complex approach. Pairwise ANI values were calculated using FastANI v. 1.33 [25] against the genomes of the most closely related type strains, with a 95% threshold applied for species delimitation. Further genomic delimitation was performed using the online Type (Strain) Genome Server (TYGS) (https://tygs.dsmz.de/) (accessed on 15 February 2026) [26]. User genomes were compared against the most closely related type strain genomes available in the TYGS database. Digital DNA-DNA hybridization (dDDH) values were calculated using the Genome-to-Genome Distance Calculator (GGDC 3.0) with the recommended formula d4 [27]. Results were evaluated against the standard dDDH thresholds of 70% for species and 79% for subspecies. Phylogenetic relationships were illustrated using two complementary trees: a genome-scale tree based on Genome BLAST Distance Phylogeny (GBDP) distances and a 16S rRNA gene sequence tree, both inferred via FastME 2.1.6 with 100 pseudo-bootstrap replicates [28]. Taxonomic conclusions were further supported by G+C content analysis [29].

2.5. Functional Annotation

Functional genome annotation was performed using two independent pipelines. In the first approach, open reading frames (ORFs) were predicted with Prokka v1.14.5 [30], and functional identification was conducted via Hidden Markov Model (HMM) profiles using KofamScan v1.3.0 based on the KEGG database [31]. Genomic mapping and pathway reconstruction were performed within the KEGG PATHWAY framework [32]. In the second approach, genomes were annotated using the RASTtk algorithm on the RAST online server (https://rast.nmpdr.org/) [33,34]. Functional categorization followed the SEED subsystems classification [35].

2.6. In Silico Toxomics and Biosafety Framework

To rigorously assess the toxicological biosafety profile of the five sequenced Nostocales strains, a targeted in silico toxomics framework was integrated into the pipeline. All predicted BGCs and genomic regions identified via antiSMASH v8.0 were systematically cross-referenced against reference entries in the Minimum Information about a Biosynthetic Gene Cluster (MIBiG) database. The screening specifically targeted reference genomic profiles of major regulated cyanotoxins, including cyclic peptide hepatotoxins (microcystins and nodularins) and potent neurotoxins (anatoxins, saxitoxins, and cylindrospermopsins). Homology levels, core synthesis genes (e.g., mcy, nda, ana, and sxt complexes), and chemical structure predictions were evaluated to confirm either the presence or complete absence of toxin-producing genetic machinery across the assembled genomes.

2.7. Prediction of Plant Growth-Promoting Genes and Visualization (PLaBAse)

PGP genes were identified and quantified using PLaBAse [36] at functional levels 3 and 4. To visualize the PGP genetic repertoire across the five strains, a balloon plot was generated based on level 3 categories using the ggplot2 package in R (v. 4.5.2). Additionally, circular genome maps were created with Proksee [37], incorporating custom BED tracks with the coordinates of PLaBAse-classified genes.

2.8. Analysis of Defensive and Mobile Genetic Elements

Antibiotic resistance genes were screened using the CARD RGI workflow in Proksee online platform (https://proksee.ca/) (accessed on 24 April 2026) with the strict cut-off mode [38]. CRISPR-Cas systems were identified using CRISPRCasFinder [39]. Mobile genetic elements were identified via mobile-OG [40] using curated HMMs profile of mobile element proteins.

2.9. Rhizosphere Colonization and Biosynthetic Potential

Rhizosphere colonization potential was assessed using RhizoSMASH (https://rhizosmash.bioinformatics.nl/) (accessed on 6 April 2026) [41] to predict rhizosphere-competent metabolic gene clusters (rCGCs), including those for chemotaxis and biofilm formation. Secondary metabolite BGCs were identified via antiSMASH v8.0 [42] with extra features enabled (KnownClusterBlast, SubClusterBlast, and MIBiG comparison).

2.10. Data Availability

The raw sequencing reads and assembled genome sequences generated in this study have been deposited in the NCBI GenBank database under the BioProject accession number PRJNA1347428. All data will be made publicly available upon publication.

3. Results

3.1. Morphological Characteristics of the Studied Strains

Microscopic examination of the five selected cyanobacterial strains revealed morphological features typical of the order Nostocales, including the presence of heterocytes and akinetes (Figure 1). Detailed morphometric metrics, including cell, heterocyte, and akinete dimensions for each strain, have been compiled and moved to Supplementary Table S1 to maintain focus on the genomic data.

3.2. Genome Assembly and Quality Assessment

WGS and de novo assembly yielded high-quality draft genomes for all five strains. Quality control of raw reads using FastQC demonstrated exceptionally high quality with median Q > 36 for all samples, corresponding to an error rate < 0.025% (less than 1 error per 4000 bases). The 10th percentile was ≥30 even at the 3′-end, and quality distribution remained stable along the reads. Read length after trimming was 75–100 bp. Genome assembly statistics and quality assessment using CheckM2 are summarized in

Table 1. Genome assembly statistics and quality assessment for the studied Nostocales strains.

Strain	Genome Size (Mbp)	Number of Contigs	N50 (kbp)	Completeness (CheckM2, %)	Contamination (CheckM2, %)	GC Content (%)
VKM Al-35	8.79	175	122.6	98.29	4.5	42
VKM Al-37	8.37	224	92.1	99.99	0.5	42
VKM Al-153	6.32	83	126.9	99.95	0.5	40
VKM Al-158	7.41	325	46.9	99.86	0.4	42
VKM Al-168	9.36	311	77.9	100	0.6	42

. The total genome size ranged from 6.32 Mbp (VKM Al-153) to 9.36 Mbp (VKM Al-168). Assembly continuity was high, with N50 values reaching 126.9 kbp for VKM Al-153 and 122.6 kbp for VKM Al-35. Four of the five strains (VKM Al-37, Al-153, Al-158, and Al-168) demonstrated excellent genome completeness exceeding 99.8% with contamination levels below 1%. Strain VKM Al-35 showed good completeness (98.29%) and a contamination level of 4.5%, which, while slightly elevated, still falls within the ≤5% threshold recommended for high-quality draft genomes [43]. These metrics confirm that the assembled genomes are of sufficient quality for reliable phylogenomic and comparative analysis.

3.3. Taxonomic Identification and Phylogenomic Analysis

The taxonomic position of the five studied strains was established using a complex approach, integrating 16S rRNA gene phylogeny, WGS phylogenomics, and genomic distance metrics (dDDH and ANI). Phylogenetic reconstruction based on 16S rRNA sequences (Figure 2) and WGS data (Figure 3) confirmed that all strains belong to the order Nostocales, distributed among the genera Nostoc (VKM Al-35, Al-37, and Al-168), Anabaena (VKM Al-153), and Hassallia (VKM Al-158).

Notably, some topological discrepancies were observed between the two trees. While the 16S rRNA phylogeny provided a consistent family-level framework, the WGS-based tree offered significantly higher resolution and stronger bootstrap support (up to 100%) for terminal branches. These differences highlight the limitations of single-gene markers in cyanobacterial taxonomy and underscore the necessity of WGS for precise species delimitation. The mismatch between 16S rRNA and phylogenomic results, most evident in VKM Al-158, highlights that 16S-based identification often fails to capture the true evolutionary diversity of Nostocales. Such taxonomic uncertainty is largely driven by the current underrepresentation of terrestrial cyanobacterial genomes in NCBI/GTDB databases.

The TYGS identification routine and genomic distance analysis unequivocally identified all five strains as potential novel species. dDDH (formula d4) values between our strains and their closest type strains ranged from 41.3% to 58.3% (

Table 2. Genomic distance metrics (dDDH and ANI) between the studied strains and their closest type strains.

Strain	Closest Type Strain (TYGS)	dDDH (d4, %)	ANI (%)	G+C Diff. (%)
VKM Al-35	Nostoc sp. 210A	51.1	93.00	0.04
VKM Al-37	Nostoc desertorum CM1-VF14	42.6	90.35	0.42
VKM Al-153	Anabaena cylindrica PCC 7122	55.4	93.65	0.04
VKM Al-158	Nostoc desertorum CM1-VF14	58.3	94.47	0.01
VKM Al-168	Nostoc punctiforme NIES-2108	41.3	90.22	0.00

), remaining well below the 70% species-delimitation threshold. Similarly, ANI values calculated using FastANI ranged from 90.22% to 94.47% (Table 2), further confirming their unique taxonomic status, as all values were below the standard 95% threshold for species delimitation.

Although the genomic distance metrics (dDDH < 70%, ANI < 95%) suggest that all five strains represent potential new species, in this study we maintain their current taxonomic names as deposited in the All-Russian Collection of Microorganisms (VKM). This allows for better consistency with previous data while highlighting the need for a future detailed taxonomic revision of these lineages, ideally including the description of novel species (and potentially novel genera) based on additional morphological, ecological, and genomic evidence.

3.4. Functional Annotation and Metabolic Potential

Functional annotation of the five genomes was performed using two complementary pipelines: Prokka/KofamScan integrated with the KEGG database and RASTtk based on the SEED subsystems classification. Prokka-based gene prediction revealed a substantial coding potential across all strains, with the total number of predicted genes ranging from 5701 (VKM Al-153) to 8157 (VKM Al-168). A substantial proportion of the proteome in each strain was successfully assigned to specific enzymatic functions (

Table 3. Gene prediction statistics for the studied Nostocales strains based on Prokka annotation.

Strain	Total CDS	Predicted Proteins	Proteins with Functional Assignment by KofamScan/KEGG *	Genes with EC Numbers
VKM Al-35	7680	7583	2995	1367
VKM Al-37	7361	7268	2950	1345
VKM Al-153	5701	5701	5650	2337
VKM Al-158	6613	6613	6544	2614
VKM Al-168	8157	8157	8063	3011

CDS, Coding DNA Sequences; EC, Enzyme Commission; * proteins excluding hypothetical proteins.

), providing a robust dataset for investigating metabolic pathways related to nutrient cycling and plant growth promotion.

3.4.1. Functional Profile Based on KEGG Orthology

Analysis of KEGG ortholog (KO) distribution revealed the presence of genes across major functional categories (

Table 4. KEGG functional category distribution for the studied Nostocales strains.

KEGG Category	VKM Al-35	VKM Al-37	VKM Al-153	VKM Al-158	VKM Al-168
Cell growth and death	96	84	64	88	99
Cellular community—prokaryotes	220	215	150	197	218
Environmental Information Processing
Membrane transport	211	199	147	197	222
Signal transduction	345	340	214	276	356
Metabolism
Amino acid metabolism	333	333	246	310	336
Biosynthesis of other secondary metabolites	120	110	61	105	115
Carbohydrate metabolism	400	397	311	361	387
Energy metabolism	317	310	287	281	305
Metabolism of cofactors and vitamins	307	326	259	296	315
Metabolism of terpenoids and polyketides	110	104	71	92	113
Unclassified
Unidentified genes	2739	2580	1835	2474	2912

). All five strains showed high representation of genes involved in core metabolic processes:

• Carbohydrate metabolism: 293–417 genes, reflecting the photoautotrophic lifestyle and capacity for polysaccharide production.
• Energy metabolism: 281–337 genes, including photosynthesis and oxidative phosphorylation pathways.
• Amino acid metabolism: 236–341 genes, indicating robust biosynthetic capabilities.
• Metabolism of cofactors and vitamins: 244–327 genes, including pathways for B-vitamin synthesis.
• Biosynthesis of other secondary metabolites: 61–120 genes, suggesting diverse bioactive compound production.

Notably, H. pseudoramosissima VKM Al-158 stood out by possessing a complete gibberellin (GA) biosynthesis pathway (KEGG map00904). This includes the core enzymes for ent-kaurene skeleton formation, specifically ent-copalyl diphosphate synthase (K04122) and ent-kaurene synthase (K05282), as well as the oxidation steps mediated by ent-kaurene oxidase (K04123) and ent-kaureenoic acid oxidase (K04124). The identification of GA14 synthase (K20666), gibberellin 12-oxidase (K04125), and gibberellin 7-oxidase (K14084) further confirms its unique genetic capacity for GA production among the five strains.

3.4.2. Functional Profile Based on SEED Subsystems

RAST annotation with SEED subsystems classification provided complementary functional insights (

Table 5. SEED subsystems functional category distribution for the studied Nostocales strains.

SEED Subsystem Category	VKM Al-35	VKM Al-37	VKM Al-153	VKM Al-158	VKM Al-168
Amino acids and derivatives	154	143	139	140	160
Carbohydrates	150	142	132	146	146
Cell wall and capsule	25	25	21	21	24
Cofactors, vitamins, prosthetic groups	144	145	144	136	154
DNA metabolism	49	49	50	50	49
Fatty acids, lipids, isoprenoids	29	31	43	54	31
Membrane transport	36	37	24	32	32
Nitrogen metabolism	8	7	8	8	8
Phosphorus metabolism	21	17	21	21	17
Respiration	50	50	50	46	46
Stress response	45	42	41	41	43
Secondary metabolism	12	16	8	12	12
Unidentified genes	5735	5422	4944	6819	8409

). Key observations include:

• Amino acids and derivatives: 139–160 genes across all strains, with VKM Al-168 showing the highest representation (160 genes).
• Carbohydrates: 132–150 genes, confirming robust carbon metabolism capacity.
• Stress response: 41–46 genes, indicating well-developed adaptation mechanisms to environmental fluctuations.
• Nitrogen metabolism: 7–10 genes, including the complete nif gene cluster for nitrogen fixation.
• Phosphorus metabolism: 17–22 genes, with complete phosphate mobilization systems.
• Secondary metabolism: 8–16 genes, with higher representation in VKM Al-37 and VKM Al-158.

3.5. In Silico Toxomics and Ecological Profiling

To rigorously evaluate the biosafety level and ecological features of the five Nostocales strains, a comprehensive in silico screening of all antiSMASH 8.0 outputs was performed against the MIBiG database. Crucially, the analysis confirmed the complete absence of BGCs or core synthesis genes for major regulated cyanotoxins, including hepatotoxins (microcystins, mcy complex; nodularins, nda complex) and neurotoxins (anatoxin-a, ana complex; saxitoxins, sxt complex; and cylindrospermopsins), across all five assembled genomes. Instead, high-confidence BGCs were strictly associated with structurally essential or benign terrestrial metabolites, such as heterocyst glycolipids involved in nitrogenase oxygen protection (100% similarity in VKM Al-35, Al-37, Al-153, Al-158, and Al-168) and geosmin (100% similarity in VKM Al-35, Al-37, Al-158, and Al-168). Furthermore, the screening revealed specific genetic prerequisites for interspecies competition and biocontrol. These include anti-fungal lipopeptide hassallidin C/D in H. pseudoramosissima VKM Al-158 (Region 33.1), protease-inhibiting anabaenopeptins (Region 13.1, 100% similarity) and nostopeptolide A2 (Region 36.1, 100% similarity) in N. minutum VKM Al-168, and iron-chelating anachelin siderophores in N. punctiforme VKM Al-37. Photoprotective mycosporine-like amino acids (shinorine) were also conserved across the Nostoc strains. These findings guarantee an excellent toxicological safety profile while highlighting the strains’ genomic capacity for rhizosphere competence and competitive survival.

3.6. Plant Growth-Promoting Potential

The comparative screening of the five strains was performed using the PLaBAse pipeline, focusing on the distribution of genes across functional categories at the PGPT Level 3 (Figure 4).

The analysis revealed a high density of genetic determinants associated with plant–microbe interactions, with the most prominent abundance observed in categories related to Colonizing Plant System, Competitive Exclusion, and Biofertilization. The functional landscape, visualized in the balloon plot, highlights the robust potential of all strains to enhance soil fertility and establish stable associations within the rhizosphere. While the core PGP profile is conserved across the collection, descriptive variations in gene copy numbers reflect the individual ecological strategies of the strains. For instance, the larger genomes of VKM Al-37 and VKM Al-168 show the highest genetic investment in colonization and stress control/biocontrol systems, whereas the more streamlined genome of VKM Al-153 maintains a specialized set of essential PGP functions.

3.6.1. Nostoc Commune VKM Al-35

The genomic architecture of VKM Al-35 reveals a comprehensive PGP profile, with functional genes organized into five distinct tracks visualized on the circular genome map (Figure 5).

In Track I (Nitrogen Acquisition), the strain possesses a complete diazotrophic apparatus, including the core nif-cluster (nifHDKBENVWZ) and supporting maturation genes (nifSU, hupUV, hypA–F), alongside high-affinity nitrate/nitrite assimilation systems (narB, nirA, nasDEF), urea utilization clusters (ureA–G, urtA–E), and ammonium transporters (amtB), highlighting its potential for soil nitrogen enrichment. Track II (Phosphate Solubilization) includes the pyrroloquinoline quinone (PQQ) biosynthesis pathway (pqqBCL), genes for organic acid production (ackA, pta, gcd), and a robust phosphate-related repertoire featuring acid phosphatases (phoA, phoD), phytase (phy), and the high-affinity pstABCS transport system. The iron sequestration potential in Track III (Iron Acquisition) is supported by biosynthetic pathways for various siderophores, including bacillibactin-like (dhbF) and enterobactin-related (entA) compounds, as well as multiple transport systems (feoA, efeU, exbBD, feuAB) and the ferric uptake regulator (fur). Track IV (Phytohormone Production) provides genomic evidence for direct plant growth stimulation through the tryptophan-dependent indole-3-acetic acid (IAA) biosynthesis pathway (trpA–G, ipdC, nitA), cytokinin modification genes (miaABE, log), and GABA metabolism (gabD, aldH). Finally, Track V (Sulfur Assimilation) encompasses the full assimilatory sulfate reduction pathway (cysCHKS) and specialized systems for the utilization of organic sulfur sources, such as sulfonates (ssuABCDE) and taurine (tauD), suggesting high metabolic flexibility in nutrient-limited environments.

3.6.2. Nostoc Punctiforme VKM Al-37

The genomic PGP-profile of VKM Al-37 is characterized by high genetic redundancy and a specialized set of traits for plant–microbe interactions (Figure 6).

In Track I (Nitrogen Acquisition), the strain features a multi-copy diazotrophic apparatus, including three copies of nifH and nifE and four copies of nifS, suggesting a robust nitrogen fixation potential. This is complemented by an extensive nitrogen utilization repertoire, including nitrate assimilation (narB, nirA), urea metabolism (ureA–G), and allantoin degradation pathways (allB, hpxB). Track II (Phosphate Solubilization) reveals a complete PQQ-dependent pathway (pqqBCL) and an exceptionally high number of phosphate transport and regulation genes, including 11 copies of phoP, 6 copies of pstS, and 4 copies of phoR, indicating a highly sensitive phosphate homeostasis system. The iron acquisition potential in Track III (Iron Acquisition) is notable for the presence of pseudomonine and bacillibactin-related clusters, as well as multiple copies of protective proteins such as dps (5 copies) and cotB4 (5 copies). Unlike VKM Al-35, this strain possesses a dedicated Track IV (Plant Vitamin Production), with complete biosynthetic pathways for vitamins B1 (thiamine), B2 (riboflavin), B9 (folate), B12 (cobalamin), E (tocopherol), and K (menaquinone), which are essential for both bacterial fitness and host plant development. Furthermore, Track V (Colonizing Plant System) highlights its superior competitive potential in the rhizosphere, featuring a complex chemotaxis system (cheABRWY), numerous Type IV pili components (pil/pix genes), and a massive expansion of adhesin genes, including 17 copies of the autotransporter adhesin ata, as well as a rich set of genes for exopolysaccharide (exo) and biofilm production.

3.6.3. Anabaena Pirinica VKM Al-153

The genomic PGP-profile of VKM Al-153 is characterized by a pronounced genomic investment in stress mitigation and complex phytohormone metabolism (Figure 7).

In Track I (Nitrogen Acquisition), the strain possesses a nearly complete nif-cluster with four copies of nifH, alongside nitrate assimilation systems (narBK, nirA) and ammonium metabolism genes (glnA, gdhA). Track II (Phosphate Solubilization) is notably reinforced, featuring six copies of the pqqL accessory gene and an extensive regulatory system, including 11 copies of phoP and 6 copies of pstS. The iron acquisition potential in Track III (Iron Acquisition) remains high, with a diverse set of siderophore-related genes (dhbF, entA, cotB) and protective proteins (dps, bfr). Track IV (Phytohormone Production) reveals a highly redundant tryptophan-dependent IAA biosynthesis pathway (trpA–E in multiple copies, ipdC, nitA) and additional genes for cytokinin modification (miaA–E, log) and polyamine biosynthesis (speA–G, potA–D), which are known to enhance plant root architecture and stress signaling. A distinctive feature of this strain is Track V (Neutralizing Abiotic Stress), which contains a massive array of genes for ROS protection (e.g., 11 copies of gst, 5 copies of bcp and trxA), heat shock response (dnaKJ, clpB–X, hsp20), and osmotic/salt stress tolerance. The latter includes 11 copies of betA (choline dehydrogenase) for glycine betaine production, as well as pathways for trehalose (treYZ) and proline (proABC) biosynthesis, suggesting superior resilience in fluctuating soil environments.

3.6.4. Hassallia Pseudoramosissima VKM Al-158

The genomic PGP-profile of VKM Al-158 stands out for its balanced metabolic capacity for nutrient mobilization and plant–microbe signaling (Figure 8).

In Track I (Nitrogen Acquisition), the strain possesses a complete nif-operon, notably including nifJ (pyruvate-ferredoxin oxidoreductase), which may support nitrogen fixation under specific physiological conditions, alongside nitrate (narBK, nirA) and ammonium (glnA, gltB) metabolism pathways. Track II (Phosphate Solubilization) represents an “extreme” P-solubilization potential, featuring a complete PQQ-cluster (pqqBCDE), multiple copies of pqqL (6), and an exhaustive set of genes for the production of varied organic acids (e.g., ackA, pta, gcd, ldhA, maeA), which is further supported by acid phosphatases (phoAD) and phytases (phy). Track III (Iron Acquisition) is highly developed, showing potential for the biosynthesis of diverse siderophores such as bacillibactin, pyoverdine, and enterobactin and reinforced by high genetic redundancy in transport (feuA, feoA) and protective (dps, sirA) proteins. Furthermore, Track IV (Plant Signalling Volatiles) emphasizes the strain’s biocontrol potential, featuring genes for the biosynthesis of growth-promoting volatiles (VOCs) such as acetoin and 2,3-butanediol (budB, budC, acoR), as well as Non-Ribosomal Peptide Synthetase (NRPS) clusters (srfAA, bacF, phzF) involved in antibiotic production and defense. A distinguishing feature of this strain is Track V (Potassium Solubilization), where the production of organic acids is coupled with specialized potassium transport systems (kdpABCE, trkA, trkG, kch), suggesting a strong capability to mobilize mineral K⁺ and maintain potassium homeostasis in the rhizosphere.

3.6.5. Nostoc Minutum VKM Al-168

The genome of VKM Al-168 represents the most functionally diverse and redundant PGP-profile among the studied strains, reflecting its large genome size and high ecological plasticity (Figure 9).

In Track I (Nitrogen Acquisition), the strain acts as an absolute leader in diazotrophic potential, featuring triple copies of nifH and nifE, alongside a well-developed nitrogenase protection system (hup/hyp clusters) and a comprehensive nitrogen assimilation suite (narBK, nirA, ureA–G, amtB). Track II (Phosphate Solubilization) exhibits maximum P-mobilization potential, characterized by a complete PQQ-cluster (pqqBCDEL), a highly expanded pho-regulatory system (14 copies of phoP, 5 copies of phoR), and multiple copies of phosphatases (phoD, 6 copies) and phytases. The iron acquisition apparatus in Track III (Iron Acquisition) is equally extreme, possessing biosynthetic pathways for three different siderophore types and a massive expansion of heme-binding proteins (hxuB, 6 copies) and transport systems (feuA, feoA, exbBD). Track IV (Phytohormone Production) provides evidence for multi-pathway auxin biosynthesis (complete trp-operon, ipdC, nitA), supplemented by cytokinin modification, GABA metabolism, and a robust polyamine biosynthesis system (speA–G, potA–D). A defining characteristic of VKM Al-168 is Track V (Competitive Exclusion), which contains a sophisticated defense and biocontrol arsenal. This includes a vast set of NRPS/PKS clusters for antibiotic production (e.g., 10 copies of srfAA), an extensive multi-drug resistance system (acrA, mexL, emrB), and a remarkable expansion of toxin–antitoxin systems (including 11 copies of vapC and 7 copies of parE), as well as diverse secretion systems (T1SS, T4SS, T6SS). These traits suggest that VKM Al-168 is not only a potent biofertilizer but also a highly resilient competitor capable of active rhizosphere colonization and pathogen suppression.

3.7. Defensive and Mobile Genetic Elements

3.7.1. Antibiotic Resistance Screening and Biosafety Assessment

The biosafety profile of the five Nostocales strains was evaluated by screening their genomes against the CARD database. No «Perfect» or «Strict» hits were identified, confirming the absence of acquired antibiotic resistance genes of clinical concern. All detected sequences were classified as «Loose» hits, exhibiting low amino acid identity (41.8–65.6%) to reference determinants (

Table 6. Summary of distal antimicrobial resistance gene (ARG) homologs identified via CARD RGI.

Strain	Resistance Gene	Product	Resistance Mechanism	Identity (%)	Coverage (%)	ARO Term
VKM Al-35	adeF (3 copies)	RND-type efflux pump	antibiotic efflux	42.6–43.2	97–100	3000777
VKM Al-37	adeF (3 copies)	RND efflux pump	efflux	42.6–44.5	97–100	3000777
	FosA8	fosfomycin thiol transferase	inactivation	43.4	100	3007371
	RSC1-1	class A beta-lactamase	inactivation	65.6	100	3009041
VKM Al-153	adeF	RND-type efflux pump	antibiotic efflux	43.3	99–100	3000777
	FosA8	fosfomycin thiol transferase	antibiotic inactivation	43.4	100	3007371
VKM Al-158	adeF	RND-type efflux pump	antibiotic efflux	42.8	99–100	3000777
	FosA8	fosfomycin thiol transferase	antibiotic inactivation	51.1	100	3007371
VKM Al-168	adeF (2 copies)	RND-type efflux pump	antibiotic efflux	41.8–43.4	99–100	3000777

). These include distal homologs of the RND-type efflux pump adeF, fosfomycin thiol transferase FosA8, and a class A beta-lactamase RSC1-1. Such distant homologs likely represent endogenous chromosomal elements involved in basal metabolism and cellular homeostasis rather than functional resistance. These findings support the biosafety of the studied strains for large-scale environmental and agricultural applications.

3.7.2. CRISPR-Cas Systems

The analysis of adaptive immunity systems using CRISPRCasFinder confirmed the presence of potentially functional CRISPR-Cas arrays in all five Nostocales strains (

Table 7. CRISPR-Cas systems identified in the studied Nostocales strains.

Strain	Total CRISPR Arrays	CRISPR Arrays (Level 1)	Cas Types Detected
VKM Al-35	44	34	I-A, I-D
VKM Al-37	38	32	I-D
VKM Al-153	27	19	I-A, I-D, III-A, III-U
VKM Al-158	28	22	I-D
VKM Al-168	47	29	I-A, I-D

). The total number of identified CRISPR arrays ranged from 27 in Anabaena pirinica VKM Al-153 to 47 in Nostoc minutum VKM Al-168. High-confidence arrays (evidence level 1), which are most likely to be active, constituted the majority of the defensive repertoire, with the highest count observed in VKM Al-35 (34 arrays). All strains possessed cas genes, though their architectural complexity varied substantially across genera. The Nostoc strains primarily harbored Type I-A and Type I-D systems, with VKM Al-37 possessing only the Type I-D system. Hassallia pseudoramosissima VKM Al-158 also exhibited a streamlined Type I-D profile. In contrast, VKM Al-153 demonstrated the greatest defensive diversity, carrying four distinct system types: I-A, I-D, III-A, and III-U. The robust presence of these adaptive immunity systems indicates an enhanced ability to resist bacteriophage infections and foreign DNA integration, a trait that is highly advantageous for the stability of cyanobacterial populations in complex soil ecosystems.

3.7.3. Mobile Genetic Elements

The analysis via mobileOG-db revealed substantial variation in the mobile genetic element (MGE) content across the five strains, highlighting notable genomic plasticity (

Table 8. MGEs in the studied Nostocales strains.

Strain	Total	Int/Exc	Stab/Def	Phage	Rep/Rec	Transfer
VKM Al-35	279	174	48	23	18	10
VKM Al-37	274	168	47	24	19	10
VKM Al-153	146	62	39	19	13	10
VKM Al-158	261	122	58	31	31	19
VKM Al-168	304	151	68	35	34	16

Note. Int/Exc—integration/excision. Stab/Def—stability/transfer/defense. Phage—phage-associated genes. Rep/Rec—replication/recombination/repair. Transfer—conjugation-associated genes.

). VKM Al-168 harbored the highest total number of mobile elements (304), while VKM Al-153 possessed the most streamlined mobilome (146 elements). The integration/excision category was predominant in all strains, with the highest count observed in VKM Al-35 (174 elements). Phage-associated genes were identified in all strains (ranging from 19 to 35 elements) and included chaperones (clpB, dnaK, groL/S), proteases (clpP, ftsH), and structural proteins (qmcA, nusA), indicating the presence of prophage sequences integrated into the genomes. The replication/recombination/repair category was most abundant in VKM Al-158 (31) and VKM Al-168 (34), potentially reflecting specialized DNA maintenance strategies in semi-desert and chestnut soil environments. The presence of transfer genes, including conjugation-associated traD, pilT, and comM, suggests a varying potential for horizontal gene transfer (HGT) across the collection. Notably, the high density of mobile genetic elements (MGEs) in the larger genomes (VKM Al-168 and VKM Al-35) correlates with the high genetic redundancy of PGP traits described above, suggesting that mobile elements have played a crucial role in the expansion of their functional repertoire.

3.8. Secondary Metabolites and Rhizosphere Colonization Factors

3.8.1. Biosynthetic Gene Clusters (antiSMASH)

Genome mining via antiSMASH 8.0 revealed an extensive and diverse biosynthetic potential across the five sequenced strains, with the total number of BGCs ranging from 15 in VKM Al-153 to 23 in VKM Al-168 (

Table 9. BGCs identified in the studied Nostocales strains.

Strain	Total BGCs	Common BGC Types						Unique/Rare BGCs
		NRPS	T1PKS	Terpenes	Lanthi (V/II)	hglE-KS/T1PKS	Phosphonates
VKM Al-35	17	5	2	5	2/1	1	1	NRPS-like (1), mycosporines (1), RRE-containing (1), spliceotides (1), NRP-metallophores (1)
VKM Al-37	22	6	3	4	2/3	1	1	Microviridins (1), azole-containing RiPP (1), RiPP-like (1), mycosporines (1), NRPS-like (1)
VKM Al-153	15	2	2	3	2/0	2	0	CDPS (1), azole-containing RiPP (1), RRE-containing (1), resorcinol + hglE-KS (1)
VKM Al-158	20	4	1	5	1/1	1	1	Darobactin/triceptides (1), betalactones (1), microviridins (1), mycosporines (1), spliceotides + RRE-containing (1)
VKM Al-168	23	9	4	5	2/2	1	1	T3PKS (1), microviridins (3), hybrid NRPS + microviridin (1), mycosporines (1), NRPS-like (1)

Abbreviations: NRPS—Nonribosomal peptide synthetase, T1PKS—Type I polyketide synthase, T3PKS—Type III polyketide synthase, hglE-KS/T1PKS—Hybrid clusters for heterocyst glycolipid synthesis, RiPP—Ribosomally synthesized and post-translationally modified peptide, CDPS—Cyclodipeptide synthase, RRE—Recognition Element-containing. BGC types follow the antiSMASH 8.0 glossary.

).

The identified BGCs represent a broad chemical repertoire, including NRPS, polyketides (PKS), and various ribosomally synthesized and post-translationally modified peptides (RiPPs), highlighting the high adaptive and competitive capacity of these soil-dwelling Nostocales. NRPS was the most abundant cluster type, particularly in the Nostoc strains. High-confidence clusters for potent bioactive compounds were identified, such as anabaenopeptins, nostopeptolides, and vioprolides (in VKM Al-37 and Al-168), as well as genus-specific hassallidins in Hassallia pseudoramosissima VKM Al-158. These metabolites are well-known for their antifungal and antimicrobial properties, suggesting a strong potential for these strains to act as biocontrol agents against soil-borne pathogens. The diversity of RiPP clusters is a notable feature of the collection, encompassing lanthipeptides (classes II and V), microviridins, and azole-containing peptides. VKM Al-168 exhibited the most complex RiPP profile with multiple microviridin clusters, while VKM Al-153 uniquely harbored a cyclodipeptide synthase (CDPS) cluster. Furthermore, the presence of mycosporine-like amino acid clusters (identified in all strains except VKM Al-153) and geosmin biosynthesis machinery (in all Nostoc and Hassallia strains) underscores their specialized adaptation to terrestrial habitats. All strains possessed highly conserved hglE-KS/T1PKS hybrid clusters for heterocyst glycolipid synthesis, confirming their optimized physiological state for nitrogen fixation. Additionally, the detection of rare darobactin (in VKM Al-158), kolossin, and phosphonate clusters further distinguishes these studied species as promising sources of unique natural products. Overall, the vast and varied biosynthetic spectrum of these strains provides a robust genetic basis for their survival in diverse soil ecosystems and their beneficial interactions with host plants through the production of siderophores, antibiotics, and signaling molecules.

3.8.2. Rhizosphere Competence and Colonization Clusters (RhizoSMASH)

The potential of the five Nostocales strains to colonize the plant rhizosphere and utilize root exudates was further characterized using RhizoSMASH. The distribution and diversity of rCGCs are summarized in the comparative heatmap (Figure 10).

The analysis confirmed that all strains possess a core set of clusters for the assimilation of major nitrogenous components of root exudates, including L-proline, glutamine (glutamine synthetase and glutaminase), and glutamate (glutamate dehydrogenase). Notably, VKM Al-37, Al-153, Al-158, and VKM Al-168 carry multiple regions for glutamine metabolism, with two distinct glutaminase clusters identified in each of these strains. Specific catabolic adaptations were observed across the collection (Figure 10). Pathways for the degradation of aromatic compounds via catechol meta-cleavage were identified in VKM Al-37 (Region 1.1), Al-153 (Region 25.1), and Al-168 (Region 5.1). In contrast, VKM Al-158 is uniquely equipped with a catechol ortho-cleavage pathway (Region 199.1), suggesting a specialized niche for degrading lignin-derived aromatics. Carbohydrate utilization potential also varied: mannitol catabolism clusters were found in VKM Al-37, Al-158, and Al-168, while VKM Al-35 stood out for its unique genomic focus on molybdopterin-dependent oxidoreductases (two ARH molybdopterin regions: 11.1 and 43.1) and alpha-diglucoside utilization. Trehalose trehalase, involved in the breakdown of the stress-protective sugar trehalose, was present in all strains except the Anabaena strain (VKM Al-153). These diverse rCGC profiles, visualized in the heatmap, indicate that while all strains are rhizosphere-competent, they have evolved distinct metabolic strategies to thrive in the nutrient-rich environment of the plant root zone.

4. Discussion

4.1. Genomic Bioprospecting as a Rational Strategy for Strain Selection

The transition from traditional polyphasic taxonomy to genome-based bioprospecting marks a paradigm shift in cyanobacterial biotechnology. While previous studies in agroecosystems primarily relied on 16S rRNA sequencing and RAPD fingerprinting [44], our approach provides a profoundly deeper functional characterization. By integrating WGS with specialized pipelines such as PLaBAse and antiSMASH, we identified agronomically critical traits, including predicted gibberellin biosynthesis and secretion systems, that remain invisible to amplicon-based methods. This screening strategy facilitates the tentative prediction of PGP potential a priori, helping screen out non-viable candidates before committing to resource-intensive initial phenotypic evaluations [2,45]. This approach aligns with the understanding that terrestrial Nostocales generally possess larger genomes and expanded gene repertoires related to regulation, transport, and stress response compared to aquatic lineages [18], reflecting their adaptation to fluctuating environments. The studied strains, originating from challenging soil habitats in southern Russia, demonstrate robust genetic machinery for environmental adaptation. Ultimately, this genome-based approach offers a rational functional blueprint for targeted strain evaluation, opening a new frontier for sustainable agriculture by enabling the deciphering of complex interactions even before field application [45,46].

4.2. Taxonomic Diversity, Phylogenomic Discordance, and Ecological Adaptation

The five strains analyzed represent three distinct lineages within the order Nostocales: Nostoc, Anabaena, and Hassallia. The studied taxonomic diversity is particularly valuable for agricultural applications, as different lineages offer complementary functional traits and survival strategies. Nostoc is globally recognized as a dominant component of biological soil crusts in arid regions. Our strains, originating from solonetz and chestnut soils, exhibit a genomic repertoire deeply shaped by water deficit and salinity. This is evidenced by the presence of stress-response genes for trehalose biosynthesis and osmoprotectant uptake, consistent with adaptations reported for other terrestrial cyanobacteria [18]. The most taxonomically intriguing case is H. pseudoramosissima VKM Al-158, which exemplifies the observed discordance between 16S rRNA markers and whole-genome data. While the 16S rRNA tree firmly positions it within the Hassallia clade, the TYGS phylogenomic analysis identifies N. desertorum as its closest type strain. This mismatch underscores a typical phylogenomic discordance within the order Nostocales driven by historical DNA marker limitations, where widespread paraphyly and polyphyly have been recently revealed [47,48,49]. The discrepancy is further exacerbated by the scarcity of high-quality reference genomes for understudied terrestrial genera like Hassallia. While a full polyphasic taxonomic revision of this lineage lies outside the biotechnological scope of this screening study, we maintain the officially verified nomenclature from the VKM collection to ensure reproducibility. Furthermore, the genomic potential of A. pirinica VKM Al-153 reveals predicted prerequisites for plant–microbe interactions. Unlike many aquatic Anabaena species, it possesses a complete set of phytohormone biosynthesis genes, suggesting an evolutionarily refined capability for rhizospheric association. Such host recognition mechanisms are often associated with an expanded repertoire of signal transduction genes, reported as a specific feature in symbiotic Nostocales by Bustos-Diaz et al. [50]. The identified genome sizes and complete functional clusters are consistent with the general pattern where versatile diazotrophs with broad ecological niches tend to possess larger genomes and more complex genetic machinery [10].

4.3. Metabolic Strategies: Universal Support Versus Specialized Stimulation

A key conceptual finding of this study is the functional divergence among the studied Nostocales, which we define as two distinct metabolic strategies: “universal support” and “specialized stimulation”. This differentiation is rooted in the genomic architectures of the strains and reflects their predicted ecological roles in the soil environment. The “Universal Support” strategy is exemplified by the Nostoc strains. These lineages focus on broad metabolic capabilities and robust environmental resilience supported by an expanded functional repertoire [13]. Their genetic capacity for extensive carbohydrate metabolism and complex exopolysaccharide (EPS) biosynthesis highlights their potential to act as ecosystem engineers. In terrestrial habitats, Nostoc species function as primary colonizers that stabilize soil surfaces, enhance water infiltration, and accumulate organic matter, thereby establishing a baseline foundation for soil health and fertility [1]. In contrast, the “Specialized Stimulation” strategy, represented by A. pirinica VKM Al-153 and H. pseudoramosissima VKM Al-158, prioritizes high-impact PGP traits over broad metabolic versatility. A notable genomic feature is their capacity for direct phytohormone production. Unlike many cyanobacteria that rely on precursor pathways, VKM Al-158 possesses a complete blueprint for direct gibberellin synthesis—a rare mechanism for direct plant–microbe interaction. Furthermore, the presence of specialized T6SS and T4SS secretion systems in Hassallia likely mediates sophisticated host recognition and interbacterial competition within the rhizosphere microbiome. This functional divergence suggests that instead of single-strain inoculants, designing synergistic mixed consortia represents the most effective path forward. Pairing the broad structural and environmental resilience of Nostoc isolates with the high-impact hormonal edge of specialized stimulation profiles offers a theoretically optimized path to enhance both soil conditioning and crop productivity [13].

4.4. Defensive and Competitive Potential: CRISPR-Cas, MGEs, and Secondary Metabolomes

Beyond their PGP traits, the five Nostocales strains possess sophisticated genetic systems for defense and intermicrobial competition essential for survival in complex soil environments. The comprehensive CARD screening demonstrated a highly favorable biosafety profile, with no acquired resistance genes of clinical concern detected. The identified loose-hit determinants likely represent ancestral chromosomal elements involved in basal metabolic dynamics and cellular homeostasis rather than providing acquired, functional antibiotic resistance [51,52]. The absence of MGEs in close proximity to these sequences further confirms that these strains pose no risk of disseminating resistance determinants in agroecosystems. This intrinsic safety is complemented by robust adaptive immunity via diverse CRISPR-Cas systems found across the strains. The high density of CRISPR arrays suggests an enhanced level of protection against phage contamination—a major bottleneck in industrial cultivation. Notably, the greater diversity of system types (including Type III) in specific strains provides a broader defense spectrum against foreign nucleic acids. Our MGE analysis revealed a striking functional trade-off in genomic plasticity that correlates with these immune repertoires: while some strains rely on a stable, streamlined genome protected by expanded CRISPR-Cas diversity, others exhibit elevated MGE loads, facilitating rapid adaptation to fluctuating soil conditions. The competitive edge of these strains is further reinforced by their biosynthetic versatility. Our antiSMASH analysis revealed an extensive secondary metabolic potential, including NRPS and PKS clusters responsible for the predicted synthesis of specialized antifungal lipopeptides and protease inhibitors, aligning with profiles characteristic of the Nostocales order [53]. Combined with high genomic plasticity and metabolic versatility, these defensive and competitive traits position our strains as resilient and safe biological candidates for sustainable agriculture, capable of maintaining functionality within the rhizosphere microbiome.

4.5. Rhizosphere Competence and Colonization Strategies

The ability to effectively colonize the plant rhizosphere is a prerequisite for any successful microbial inoculant. Using RhizoSMASH, we identified a predicted core metabolic capacity for the assimilation of key nitrogenous root exudates across all strains. However, distinct catabolic specializations suggest that these lineages have evolved to occupy complementary nutritional niches. For instance, the identification of diverging aromatic compound degradation pathways (meta-versus ortho-cleavage) indicates variations in the capacity to utilize lignin-derived aromatics, which is particularly advantageous in soils with high plant residue turnover. Furthermore, the varying distribution of carbohydrate catabolism systems across the genera underscores the metabolic complementarity of these lineages, supporting their use in designed consortia to maximize colonization efficiency under fluctuating edaphic conditions. These genomic predictions are conceptually supported by functional literature evidence from close relatives. For example, N. punctiforme PCC 73102 establishes stable endophytic associations with rice, colonizing root tissues and exhibiting substantial nitrogenase activity even under nitrogen-limiting conditions [54]. This is consistent with evidence demonstrating that despite the diversity of the surrounding soil, members of the Nostocales order are specifically recruited by host plants to become dominant taxa within the root endosphere [55]. Our analysis confirms that these strains encode complete pathways for nitrogen fixation, phytohormone synthesis, and glycoside hydrolases potentially involved in plant cell wall penetration via the apoplastic route, suggesting that they possess the genetic prerequisites for rhizospheric and intracellular colonization [54]. While endophytic potential is well-documented for Nostoc, our findings open new perspectives for understudied lineages like Hassallia. The presence of specialized Type VI and Type IV secretion systems, combined with unique aromatic degradation pathways, suggests an active and competitive colonization strategy. The practical efficacy of such associative cyanobacteria in the rhizosphere can substantially enhance soil nutrient availability and crop yield [56]. The ability of these Nostocales to form stable biofilms, essential for long-term survival in the rhizosphere [54], further enhances their biotechnological value. As suggested in the literature, the development of “biofilmed biofertilizers” using such rhizosphere-competent strains could profoundly improve nutrient uptake efficiency, positioning our strains as prime predictive candidates for advanced agricultural formulations [57].

4.6. Implications for Soil Health and Sustainable Agriculture

The genomic repertoire of the five studied strains provides a robust scientific basis for their evaluation as multifunctional biofertilizers. By integrating nutrient mobilization, soil conditioning, and stress alleviation, these Nostocales offer a potential “three-in-one” solution for sustainable crop production, embodying the “sustainable biofactory” model [58]. All five strains harbor complete functional gene clusters, confirming their capacity for biological nitrogen fixation, which is critical to mitigate the environmental costs of synthetic N-fertilizers, such as greenhouse gas emissions and nitrate leaching into groundwater. Our findings align with conserved baseline patterns across these genera [10]. Given that cyanobacterial inoculants can theoretically substitute a substantial portion of synthetic nitrogen requirements under optimized conditions [59], these strains could notably lower the carbon footprint of cereal production, as supported by general yield trends reported for associative cyanobacteria in field trials [60,61]. Furthermore, as global phosphate rock reserves dwindle [9], the predicted ability of these strains to solubilize and mobilize phosphorus through encoded phosphatases and phytases becomes a strategic asset. By unlocking phosphorus already present in the soil but unavailable to plants, these cyanobacteria can promote a more circular and self-sustaining nutrient cycle, potentially reducing dependence on non-renewable mineral phosphate inputs. The presence of EPS biosynthesis pathways, particularly in the Nostoc strains, also highlights their potential capability to act as ecosystem engineers. In degraded or semi-arid soils, such as the solonetz and chestnut soils from which these strains were isolated, cyanobacterial EPS can increase soil polysaccharide content and improve aggregate stability [62]. This structural improvement, combined with the identified genomic prerequisites for osmoregulation and antioxidant defense, enhances the theoretical potential for plant survival under the escalating threats of climate change, including drought and soil salinization [8].

4.7. Comparison with Previously Reported Cyanobacterial Biostimulants

The strains characterized in this study compare favorably with previously reported cyanobacterial biostimulants regarding their genomic blueprints, offering distinct theoretical advantages for terrestrial applications. As reviewed by Zahra et al. [63], the majority of current commercial cyanobacteria-based products rely on Arthrospira (Spirulina) strains, which are primarily aquatic and lack heterocytes, thereby substantially limiting their nitrogen-fixing capacity in soil environments. In contrast, all five strains in our study are indigenous soil inhabitants equipped with complete nitrogen fixation machinery, making them intrinsically better suited for sustainable agriculture. The predicted phytohormone production capacity of our strains suggests a notable metabolic potential; while auxin production is common in many soil lineages, the presence of predicted pathways for direct gibberellin synthesis and specialized secretion systems represents a rare genomic finding among terrestrial cyanobacteria. However, it is critical to emphasize that these genomic predictions represent biological potential rather than guaranteed phenotypic expression under field conditions. While the literature provides promising benchmarks for the magnitude of effects from strains with identical genetic pathways [61,64], direct extrapolation to our collection strains remains purely theoretical until comprehensive in vivo trials are conducted. Furthermore, while the identification of complementary metabolic strategies provides a rational genomic foundation for designing synergistic mixed consortia, the practical implementation of complex inoculants faces substantial ecological hurdles. As highlighted by the antiSMASH screening, these strains possess active biosynthetic gene clusters for competitive secondary metabolites, such as antifungal lipopeptides and protease inhibitors. Under natural rhizosphere conditions, intense interspecies competition or unpredictable secondary metabolic cross-talk could emerge. Such competitive interactions might lead to the mutual inhibition of the co-inoculants or even exert unexpected negative pressures on the host plant. Therefore, rigorous, step-by-step in vitro compatibility testing and controlled greenhouse bioassays are mandatory prerequisites before any polyfunctional consortia can be safely recommended for agricultural application.

4.8. Limitations of the Study and Future Research Directions

Despite the identified biotechnological potential, several limitations must be acknowledged to provide a balanced assessment of the current findings. The primary limitation is that our functional characterization relies on in silico genomic predictions. While the presence of genetic pathways indicates metabolic capacity, actual phenotypic expression depends on environmental stimuli and regulatory networks. Future studies must quantify parameters such as auxin levels, nitrogenase activity, and exopolysaccharide yield under controlled conditions, alongside metabolomic validation via GC-MS or LC-MS/MS to confirm specific compound synthesis [65]. However, this computational focus was a deliberate strategy. Comprehensive phenotypic testing and field trials are highly labor-intensive and financially expensive; therefore, this genome-based screening serves as a critical, cost-effective pre-selection filter to rationally screen out non-viable candidates and minimize R&D costs before committing to extensive “wet-lab” experiments. Similarly, our analysis of CRISPR-Cas systems and mobile genetic elements represents a static genomic snapshot. The presence of multiple defense arrays does not guarantee simultaneous activity, as some loci may be degenerate remnants or inactive under standard conditions. Experimental phage challenge assays and transcriptomic profiling remain necessary to confirm industrial reliability against viral contamination. Furthermore, maintaining these strains as non-axenic unicyanobacterial cultures introduces potential cross-kingdom interactions with satellite bacteria. These companion microorganisms may either synergistically enhance cyanobacterial performance through nutrient cycling or interfere with product standardization. Deciphering these relationships through targeted “synthetic community” reconstruction is a critical future step for developing predictable commercial products [2]. Finally, transitioning from laboratory predictions to field efficacy remains a major hurdle. Associative diazotrophs often face intense competition from the native soil microbiome for root attachment sites [3]. Additionally, high mineral nitrogen content in agricultural soils can potentially repress nitrogenase expression, meaning these biofertilizers are expected to be most effective under moderate nitrogen limitation [66].

5. Conclusions

This genome-based bioprospecting study provides a comprehensive characterization of five terrestrial cyanobacterial strains from the order Nostocales and demonstrates their substantial predictive potential as eco-friendly agents for soil fertility enhancement and plant growth promotion. The analysis revealed two complementary, genome-encoded metabolic strategies that provide a rational blueprint for targeted strain and consortia selection: the “universal support” profile characteristic of Nostoc strains, which act as potential ecosystem engineers via polysaccharide production and broad resilience, and the “specialized stimulation” profile found in Anabaena and Hassallia strains, which are optimized for directed hormonal and functional plant interactions. Notable specific findings include the discovery of a rare direct gibberellin biosynthesis pathway, predicted here for the genus Hassallia for the first time.

In a broader context, this work underscores the high value of integrating whole-genome sequencing and specialized computational pipelines with classical culture screening to establish a cost-effective, predictive pre-selection filter for agricultural biotechnology. The characterized strains represent promising biological candidates with a well-defined genomic foundation, serving as prime targets for the downstream development of multifunctional biofertilizers capable of contributing to sustainable agricultural systems.