Genomic insights into Gordonia diversity: proposal of three novel species (Gordonia. altitudinis sp. nov., Gordonia. ligustrum sp. nov., and Gordonia. pistacia sp. nov.) with distinct environmental and biomedical traits

Abstract

Background

Rare actinomycetes, particularly Gordoni spp., are emerging as critical sources of bioactive metabolites and opportunistic pathogens.

Results

In this study, we isolated three novel Gordonia strains from soil samples and characterized their taxonomic status using a polyphasic taxonomic approach. Phylogenetic analysis of 16S rRNA genes and whole-genome comparisons indicated that strains CPCC 205333 ^T, CPCC 205515 ^T, and CPCC 206044 ^T represent three distinct novel species. The overall genome relatedness indices of average nucleotide identity (ANI) and digital DNA-DNA hybridization (dDDH) between these studied strains and their related type strains of the genus Gordonia were all below the established thresholds for species delineation, confirming the classification of these three as novel species, for which we propose the names Gordonia altitudinis sp. nov., Gordonia ligustrum sp. nov., and Gordonia pistacia sp. nov., respectively. Functional annotation revealed their ecological versatility, with Gordonia spp. contributing significantly to soil microbiome functionality through plant growth-promoting traits (e.g., nitrogen fixation, siderophore production) and biosynthetic gene clusters (BGCs), while also harboring virulence factors. Pan-genomic analysis of 225 Gordonia strains delineated an open gene pool (α = 0.82; 22% fluidity), reflecting adaptive plasticity. Core genomes were enriched in conserved metabolic pathways, whereas accessory and strain-specific genes showed niche-driven functional diversification, suggesting ecological specialization.

Conclusion

These findings expand the genomic and functional understanding of Gordonia, highlighting its dual role in environmental resilience and pathogenicity, with potential applications in biotechnology and microbiome engineering.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12866-025-04362-0.

Background

Rare rhizo-Actinomycetes are a non-Streptomycetes group of Actinomycetes abundant in the rhizosphere of plants. These microorganisms have the capacity to stimulate plant growth in a number of direct and indirect ways, including mineral nutrients, regulating phytohormone levels, and biocontrol of phytopathogen [1, 2]. Several genomic analyses reveal conserved gene clusters encoding chitinases, β−1,3-glucanases, and laccases that synergistically degrade phytopathogen cell walls while releasing lignin-derived aromatic metabolites [3]. Notably, genera like Micromonospora and Actinoplanes exhibit lignocellulose-degrading enzyme systems (e.g., dye-decolorizing peroxidases, aryl-alcohol oxidases). This metabolic versatility is further evidenced by hydrocarbon degradation pathways (alkB hydroxylases) and xenobiotic detoxification networks (cytochrome P450 monooxygenases) in Nocardia and Rhodococcus [4, 5].

The genus Gordonia, a member of rare Actinobacteria, stands out for its functions of biotransformation, biodegradation, and active substance synthesis. Gordonia (originally ‘Gordona’), belonging to the family Gordoniaceae, was initially proposed by Tsukamura in 1971 [6], with Gordonia bronchialis as the type species. At the time of writing, there are 55 species with validly published and correct name belonging to the genus Gordonia (data from LPSN, https://lpsn.dsmz.de/genus/gordonia, accessed on May 27, 2025). Members of the genus Gordonia are Gram-positive, non-motile that may appear as short-rods, cocci or filamentous forms in wastewater treatment environments. They are distinguished by a chemotaxonomic profile including A1γ-type peptidoglycan with long‑chain mycolic acids, meso-diaminopimelic acid, glycolated muramic acid, and arabinogalactan cell wall polysaccharides [7]. Initially, Gordonia were isolated from patients with pulmonary disease [8]. Nowadays, novel species of this genus have been found in various environmental sources, such as soil [9, 10], foul water [11], wastewater [12, 13], oil-producing well [14], and plant-associated source, such as rice stem [15]. Some members of the genus Gordonia have attracted much attention due to their ability to degrade toxic environmental pollutants such as benzothiophene and hydrocarbons, and to produce diverse bioactive compounds [7, 13, 16–22].

In this study, three Gordonia strains were isolated and identified from the soil samples collected from various regions of Yunnan Province, China. Utilizing a polyphasic approach involving phenotypic and phylogenetic analyses, alongside whole-genome comparisons, we have conclusively verified that these strains indeed belong to three novel species within the genus Gordonia. Furthermore, we investigated the genetic basis of plant growth promotion and human infection across Gordonia strains and functional diversity based on the pangenome of Gordonia strains.

Materials and methods

Strain isolation

Strains CPCC 205333 ^T (original number CS487^T), and CPCC 205515 ^T (original number I21A-00135 ^T), and CPCC 206044 ^T (original number A305^T) were isolated from ecologically distinct soil environments in Yunnan Province, southwestern China. Specifically, strain CPCC 205333 ^T was isolated from a montane soil in the Cangshan Mountain. Strain CPCC 205515 ^T and CPCC 206044 ^T was recovered from the rhizosphere soil of Pistacia weinmanniifolia in Honghe Hani and Yi autonomous, and Ligustrum lucidum in Chuxiong Yi autonomous, respectively. Soils were suspended in PBS buffer (1 g of soil was suspended in 9 mL of 1X PBS buffer), and serially diluted to a concentration of 10^–4 and plated on chitin agar medium (2 g L⁻¹ chitin, 0.7 g L⁻¹ K₂HPO₄, 0.3 g L⁻¹ KH₂PO₄, 0.5 g L⁻¹ MgSO₄·7H₂O, 0.1 g L⁻¹ FeSO₄, 0.001 g L⁻¹ ZnSO₄, 0.001 g L⁻¹ MnCl₂ and 15 g L⁻¹ agar) with nystatin (50 mg L⁻¹), novobiocin (30 mg L⁻¹), and nalidixic acid (30 mg L⁻¹). After culture at 28℃ for 3 weeks, bacterial colonies were purified according to the morphology of the colonies. The purified cultures were maintained as glycerol suspensions (20%, v/v) at −80℃.

Phylogenetic analysis based on 16S rRNA gene

The purified strains were incubated in ISP2 media at 28 ℃ for 7 days for DNA extraction. The genomic DNA of newly isolated strains was extracted using a genomic DNA extraction kit (Tiangen, China). The 16S rRNA genes were amplified by PCR using forward primer 27 F (5′-AGA GTT TGA TCC TGG CT-3′) and reverse primer 1492R (5′-GGT TAC CTT GTT ACG ACT T-3′), according to the method previously described by Li et al. [23]. The sequences were quality-checked and assembled using Seqman program [24]. The EzBioCloud (https://www.ezbiocloud.net/) platform was used to analysis the 16S rRNA gene sequence similarities to closely related species [25]. Moreover, top hits were selected as closest neighbours to determine the precise taxonomic positions for these novel strains. The phylogenetic trees using neighbor-joining (NJ) [26], maximum-likelihood (ML) [27] and Maximum parsimony (MP) [28] methods were conducted on the MEGA software package (version 7.0), with a bootstrap value of 100 resamplings [29].

Genomic sequencing, assembly and annotation

Three novel Gordonia isolates were incubated under the same growth conditions with 16S rRNA gene sequencing. The total DNA of the were extracted using the QIAamp DNA MiniKit (Qiagen) according to the manufacturer’s protocol. Whole-genome sequencing was performed at the Beijing Genomics Institute (Beijing, China) using the Illumina HiSeq 4000 platform (Illumina, San Diego, CA, USA), producing 2 × 150 bp paired-end reads. After adaptors and low-quality bases were trimmed using Fastp v1.0.1 tool, the pair-end reads were assembled by SPAdes v3.10 [30] with default settings. The quality of assemblies was evaluated using CheckM2 [31]. All assemblies passed the quality control threshold (Completeness > 90% and Contamination < 5%).

Genome based identification and phylogenetic analysis

The overall genome relatedness indices (OGRIs) were used to determine pairwise genome similarity for species delineation. The average nucleotide identity (ANI) and digital DNA-DNA hybridization (dDDH) were calculated using the online ANI calculator [32] and Genome-to-Genome Distance Calculator (https://ggdc.dsmz.de/ggdc.php/), respectively. Furthermore, a phylogenomic analysis was performed to decipher the evolutionary relationships among related species. The single-copy orthologroups (SOG, one sequence per genome) were inferred by OrthoFinder v2.0.0 [33] based on the proteomes annotated by Prokka v1.1.4 [34]. For each SOG, a multiple sequence alignment was implemented using ClustalO [35], and the ambiguous sites were trimmed by trimAI [36] with the option "automated1". A maximum-likelihood species tree was constructed based on a concatenation of these alignments using IQ-tree [37] with 1000 ultrafast bootstrap replicates, and the best model was selected by ModelFinder [38].

Morphological and physiological characterization

Bacterial suspensions were adjusted to an optical density at 600 nm of 0.5, and 100 µL of this suspension was used to inoculate for each phenotypic test. Growth on various culture media was tested by using International Streptomyces Project 2 (ISP2) agar, peptone yeast glucose (PYG) medium, tryptone soy agar (TSA; Difco), Luria–Bertani agar (LB; Difco), Reasoner’s 2 A agar (R2A; Difco), glucose yeast malt agar (GYM: 4.0 g L⁻¹ yeast extract, 10.0 g L⁻¹ malt extract, 4.0 g L⁻¹ dextrose, 2.0 g L⁻¹ calcium carbonate, 20.0 g L⁻¹ agar) and nutrient agar (NA; Difco) for up to 7 days at 28 ℃. The growth temperature was tested at 4, 10, 15, 20, 25, 28, 30, 37, 40, 42, and 45 ℃ using TSA medium. NaCl tolerance was determined in tryptone soy broth (TSB) supplemented with 0, 1, 3, 5, 7, and 10% (w/v) NaCl. The pH tolerance was tested in TSB incubated at 28 ◦C. The pH was adjusted to pH 4 to 13 (1 pH unit interval incrementing) with the following buffer systems: citric acid/sodium citrate (0.1 M, pH 4.0–5.0), KH₂PO₄/NaOH (0.1 M, pH 6.0–8.0), Na₂CO₃/NaHCO₃ (0.1 M, pH 9.0–10.0), Na₂HPO₄/NaOH (0.05 M/0.1 M, pH 11.0), and KCl/NaOH (0.2 M, pH 12.0–13.0). The Gram reaction was determined by the standard Gram-stain method and observed using light microscopy (Zeiss Axio Scope, A1 Vario). To observe cell motility, the strain was grown on ISP2 agar at 28 °C for 4 days, flooded with 0.1 M potassium phosphate buffer (pH 7) at room temperature for 10–30 min before observation with a Light microscope. Catalase activity was determined by bubble production in a solution of 3% (v/v) hydrogen peroxide and oxidase activity was tested using API oxidase reagent (bioMérieux) according to the manufacturer’s instructions. Metabolic characters were tested with Biolog GEN III MicroPlate, API 50CH, and API ZYM test kits (bioMérieux) according to the manufacturer’s instructions after incubation at 28℃ for 48–96 h. Other physiological tests, includingd H₂S production and the hydrolysis of gelatin, cellulose, and starch, were examined according to previously described procedures [39].

Chemotaxonomic analysis

Biomass for chemotaxonomic characterization of the three strains was collected from 3-day-old cultures growing in shake flasks on a rotary shaker (150 r.p.m.) using TSB medium at 28℃. Polar lipids were extracted and examined by two-dimensional TLC and identified according to the method described by Minnikin et al. [40]. The menaquinones were extracted [41] and analyzed by High Performance Liquid Chromatography (HPLC) [42]. Cellular fatty acids were extracted and analyzed using the Sherlock Microbial Identification System (MIDI) with the MIDI Sherlock Version 6.0 and ACTIN1 database [43].

Pangenome analysis of Gordonia strains

To provide a comprehensive pangenome framework for assessing functional gene content variation and evolutionary trends, we conducted a genome dataset covering 303 Gordonia strains, which included 3 strains from this analysis and additional genomes from the NCBI assembly database. All infant-associated genome assemblies were subjected to a contamination check via CheckM2, and contaminated (contamination > 5%) and/or incomplete (completeness < 90%) genome assemblies were excluded from further analysis. Genomes with an average nucleotide identity (ANI) > 99.95% and alignment fraction > 50% were considered redundant; one representative genome from each redundant group was retained. The remaining genomes constituted a representative Gordonia genome dataset (n = 225).

Pangenome of Gordonia genomes based on Prokka annotation of coding sequences (general feature format) was next generated using Orthofinder with default options. Rarefaction curves and Heaps’ law ɑ values were calculated using the R package micropan [44]. Maximum-likelihood tree was constructed on a concatenation of 259 single copy alignments using IQ-TREE v.2.0.5 with ultra-fast bootstrap replicates -B 1000 and automatic evolutionary model selection option -m TEST (best fit model for the trees was determined as LG + F + I + G4 according to the bayesian information criterion).

The amino acid sequences of these genes were aligned against Cluster of Orthologous Groups (COG) and Kyoto Encyclopedia of Genes and Genomes (KEGG) databases to obtain their corresponding annotations using eggnog Mapper [45] with “–id 30 –subject_cov 40 –query_cov 40 –evalue 0.001”. Virulence factor (VF) annotations were conducted with DIAMOND (v.2.0.14) against the VFDB database with an e-value < 1e⁻⁵, subject and query coverage > 40%. PGPf_finder pipeline was used to search and compare genomes according to their plant growth-promoting traits (PGPT) [46]. Genome mining of biosynthetic gene clusters (BGCs) was performed using antiSMASH 7.0 [47].

Results and discussion

Isolation and identification of Gordonia strains

Strains CPCC 205333 ^T, CPCC 205515 ^T, and CPCC 206044 ^T were isolated from three soil samples in Yunnan, China. Full-length 16S rRNA gene sequences of these strains were obtained and subjected to comparative analysis using multiple alignment tools and phylogenetic tree construction methods. The highest 16S rRNA sequence identities for isolates CPCC 205333 ^T and CPCC 206044 ^T were with the type strains of the species G. effusa DSM 44810 ^T (98.21%) and G. rhizosphera JCM 10426 ^T (98.68%), respectively. CPCC 205515 ^T shared highest 16S rRNA sequence similarity with G. mangrovi HNM0687^T and G. bronchialis DSM 43247 ^T (both of 98.48%). The observed similarity values fall substantially below the generally accepted 98.7% cutoff criterion for species delineation [48, 49]. In the phylogenetic tree using neighbor-joining analysis of 16S rRNA gene sequences, the isolates constituted distinct monophyletic clusters compared to their phylogenetic relatives in the genus Gordonia (Fig. 1). Notably, congruent topological structures recovered from ML and MP trees consistently demonstrated substantial evolutionary distances from species with validly published and correct name (Fig. 1). Moreover, CPCC 205333 ^T and CPCC 206044 ^T clustered closely with G. effusa DSM 44810 ^T and G. rhizosphera JCM 10426 ^T, respectively. In contrast, CPCC 205515ᵀ did not cluster with G. bronchialis DSM 43247ᵀ but instead grouped as a sister lineage to G. mangrovi HNM0687ᵀ. Consequently, these three strains (G. effusa DSM 44810 ^T, G. rhizosphera JCM 10426 ^T, and G. mangrovi HNM0687^T) were selected as reference strains and integrated genomic analyses were subsequently conducted to clarify the exact taxonomic status of three novel strains.

Fig. 1.

Neighbour-joining phylogenetic tree based on 16S rRNA gene sequences showing the relationships between strain CPCC 205333 ^T, CPCC 205515 ^T and CPCC 206044 ^T with other representatives of the order Mycobacteriales. Numbers at nodes represent bootstrap percentages (> 50%) based on 100 replicates. Filled circles indicate that the corresponding nodes were also recovered in phylogenetic trees generated using Maximum Likelihood and Maximum Parsimony methods. Corynebacterium accolens ATCC 49725 ^T (GenBank accession no. AJ439346) was used as an outgroup. Bar, 0.008 substitutions per nucleotide position

Comparative genome and phylogenomic analysis for three novel strains

The whole-genome sequences of isolates CPCC 205333 ^T, CPCC 205515 ^T and CPCC 206044 ^T have been deposited in GenBank under accession numbers JBCLVI000000000, JBCLVJ000000000, JBCLVL000000000, respectively. The genomes sizes of these novel strains range from 4.5–6.2 Mbp with G + C content of 62.4–66.5%. Detailed genomic characteristics including coding sequences and rRNA operon counts are presented in Supplementary Table S1.

The OGRIs values, including ANI and dDDH, were calculated between the strains with their closely related type strains. The dDDH estimates between the novel strains CPCC 205333 ^T, CPCC 205515 ^T, and CPCC 206044 ^T and their phylogenetically neighbors were 19.3–21.9%, which was lowerthan 70%, the threshold for species delineation [50]. Corresponding ANI values calculated through OrthoANI (73.1–82.8%) were consistently below the 95% species boundary threshold [50].

To establish precise phylogenetic relationships of strains CPCC 205333 ^T, CPCC 205515 ^T and CPCC 206044 ^T, a curated dataset comprising 49 relatives of validated genomes was determined and subjected to phylogenomic analysis. The resulting phylogenomic tree exhibited strong topological congruence with 16S rRNA gene phylogeny while resolving previously ambiguous branching patterns with 100% bootstrap support at key nodes (Fig. 2). Strain CPCC 205333 ^T and G. effusa NBRC 100432 ^T formed a clade inside the phylogenomic tree, and it was distant from the other members of Gordonia. In addition, CPCC 205333 ^T was placed independently of the other Gordonia species and formed a monophyletic clade with a bootstrap value of 100%. Similarly, strain CPCC 206044 ^T also exhibited a distinct phylogenetic placement, forming its own monophyletic clade.

Fig. 2.

The phylogenetic tree based on the genome sequences of CPCC 205333 ^T, CPCC 205515 ^T and CPCC 206044 ^T with other representatives of the order Mycobacteriales. The type strain of Corynebacterium accolens was used as an outgroup. The bold text represents novel species in this study. The best-fit substitution model was LG + F + I + G4 according to Bayesian Information Criterion (BIC). Bootstrap values that equal to 100% are not shown. The scale bar denotes 0.3 substitutions per nucleotide position

Consequently, phylogenetic analysis and genomic relatedness indicated that strains CPCC 205333 ^T, CPCC 205515 ^T and CPCC 206044 ^T should be classified as three novel species of the genus Gordonia.

Phenotypic characteristics and chemotaxonomy

Based on the criterion of polyphasic taxonomy, three candidates of novel species were investigated for phenotypic characterization. Both strains displayed Gram-stain-positive attribute, lacked motility and gliding capabilities, exhibited aerobic behaviour, and featured a rod-shaped morphology. Notably, they were catalase positive and oxidase negative. The phenotypic attributes that distinguished these three strains from their most closely related species (CPCC 205333ᵀ vs. G. effusa DSM 44810ᵀ; CPCC 205515ᵀ vs. G. mangrovi HNM0687ᵀ; and CPCC 206044ᵀ vs. G. rhizosphera JCM 10426ᵀ) are summarized in Table 1. Strain CPCC 205333ᵀ could not utilize D-fructose, nor grow at and above 7% NaCl, pH 9.0, or 37 °C, distinguishing it from G. effusa DSM 44810ᵀ. CPCC 205515 ^T could be differentiated from G. mangrovi HNM0687^T by its acid production form D-mannitol, D-sucrose and D-trehalose, while it did not utilize α-D-Glucose, D-Mannose, D-Fructose, D-Mannitol D-Arabitol, and myo-Inositol. Strains CPCC 206044 ^T could grow at 4 °C and utilized D-Mannose, D-Galactose, D-Arabitol and myo-Inositol, while G. rhizosphera JCM 10426 ^T did not. The formal proposal of these new species is in the Species Descriptions section.

Table 1.

Physiological characteristics of three novel Gordonia strains and their closely related type strains

Characteristics	1	2	3	4	5*	6
Growth Temperature (℃)	4–28	10–42	4–42	4–37	20–40	20–37
Growth pH	6.0–8.0	5.0–11.0	6.0–11.0	6.0–9.0	5.0–10.0	6.0–12.0
NaCl tolerance (%)	5	10	5	7	8	10
Utilization of
α-D-Glucose	-	-	+	-	+	+
D-Mannose	-	-	+	-	+	-
D-Fructose	-	-	+	+	+	+
D-Galactose	-	-	+	-	-	-
D-Sorbitol	-	-	+	-	-	-
D-Mannitol	-	-	+	-	+	+
D-Arabitol	-	-	+	-	+	-
myo-Inositol	-	-	+	-	+	-
Activities of
Alkaline phosphatase	+	+	+	+	ND	-
Esterase(C4)	-	+	+	-	ND	-
Lipase (C14)	+	+	+	+	ND	-
Valine arylamidase	+	+	+	+	ND	-
Cystine arylamidase	+	+	+	+	ND	-
Trypsin	+	+	-	+	ND	+
α-chymotrypsin	+	+	-	+	ND	+
α-galactosidase	+	+	-	+	ND	-
β-glucosidase	+	+	-	+	ND	+
Acid production from
glycerol	-	+	+	-	+	-
D-glucose	-	+	+	-	+	-
D-fructose	+	+	+	+	+	-
D-mannose	-	+	+	-	+	-
inositol	-	+	+	-	+	-
D-mannitol	+	+	+	+	-	-
D-sucrose	-	+	+	-	-	-
D-trehalose	-	+	+	-	-	-
Major fatty acids (> 10%)	C16:0, C18:0 10-methyl, C18:1ω9c, Summed Feature 4	C16:0, C18:0 10-methyl, C18:1ω9c, Summed Feature 4	C16:0, C18:0 10-methyl, C18:1ω9c, Summed Feature 3	C16:0, C18:0 10-methyl, C18:1ω9c, Summed Feature 4	C16:0, C18:0, C18:0 10-methyl, Summed Feature 3	C16:0, C17:0, C18:0 10-methyl, C18:1ω9c, Summed Feature 4

Strains: 1, CPCC 205333 ^T; 2, CPCC 205515 ^T; 3, CPCC 206044 ^T; 4, Gordonia effusa DSM 44810 ^T; 5, Gordonia mangrovi HNM0687^T; 6, Gordonia rhizosphera JCM 10426 ^T. +, positive; -, negative activities; ND, not determined. Summed Feature 4: C_15:0 iso 2-OH/trans-C_16:1ω9c, Summed Feature 3: C_16:1ω7c/C_16:1ω6c

^* Data taken from Xie et al., 2020 [51]

The chemotaxonomic characteristics of the isolates were consistent with those typically observed in the genus Gordonia. All isolates exclusively contained MK-9 (H₂) as the predominant respiratory quinone. The polar Lipid profiles of strains CPCC 205333 ^T, CPCC 205515 ^T and CPCC 206044 ^T were dominated by diphosphatidylglycerol, phosphatidylglycerol, phosphatidylethanolamine, and phosphatidylinositol mannoside (Supplementary Fig. 1). Differentially, phosphatidylglycerol was detected in the G. mangrovi HNM0687^T. Moreover, the primary fatty acid constituents (> 10%) identified in these strains were C_16:0, C_18:0 10-methyl, and C_18:1ω9c. While C_17:0 was detected as a major fatty acid (> 10%) exclusively in G. rhizosphera JCM 10426 ^T, while it was either absent or present only in trace amounts (< 2%) in the other strains. Detailed cellular fatty acid compositions (%) based on FAME analysis for these isolates, along with their closely related type strains, are presented in Supplementary Table S2.

Phylogenetic and functional genome analysis of Gordonia spp.

Here, a comprehensive genomic investigation was conducted on 225 Gordonia strains representing diverse ecological niches, including aquatic (14 strains), wastewater (28 strains), soil (75 strains), sediment (21 strains), host-associated habitats (54 strains), and others (33 strains) (Supplementary Table S3). The genome sizes of 255 strains spanned 3.19–7.33 Mb, and protein-coding gene counts ranged from 3,008 to 6,655 per strain. Notably, genome sizes differed significantly between habitats (Kruskal–Wallis test, P value = 1e⁻⁵), with soil‑derived and general aquatic Gordonia strains exhibiting larger genomes than those originating from wastewater treatment systems and host‑associated niches (Fig. 3A and Supplementary Figure S2). Core genome-based phylogenetic analysis of 225 Gordonia strains from different habitats was conducted to examine their evolutionary relationships. Type strains of validly published Gordonia species sharing ≥ 95% ANI, forming strongly supported monophyletic clades (bootstrap support = 100%) across the maximum likelihood phylogeny reconstructed from concatenated single-copy orthologues (Fig. 3A). However, beyond these species-level groupings, strains isolated from similar environments were distributed across distinct phylogenetic branches, suggesting phylogenetic distance between strains was not related to their ecological origin or genome size.

Fig. 3.

Genomic features and phylogenomics of Gordonia strains. A The phylogenetic tree based on the genome sequences of 225 Gordonia strains. Red filled circle on the node represent type strains. Color strip and bar outside the tree are depicted by isolated sources and genomic size, respectively. Bootstrap values that are equal to 100% are not shown. The scale bar denotes 0.08 substitutions per amino acid position. B Gene copies or numbers associated with virulence factors, plant growth promotion traits, and biosynthetic gene clusters across Gordonia strains

As an emerging dual-functional microorganism with significant roles in both soil ecosystems and clinical infections, we conduct a subsystems analysis for Gordina strains, including VFs, PGPTs, and BGCs. VF profiling analysis identified an average of 77.61 ± 5.28 VFs per strain, with 100% of isolates encoding conserved modules for immune modulation, adhesion and stress survival VFs (Fig. 3B). These findings were consistent with previous case reports of clinical infections caused by Gordonia spp. [52–54]. In the PGPT analysis, over 40 functional traits assigned to 7 categories associated with plant growth promotion were identified across the Gordonia strains (Fig. 3B). Notably, A large number of genes related to biofertilization (phosphate solubilization, nitrogen and iron acquisition), bioremediation (xenobiotics biodegradation, heavy metal detoxification), plant colonization (plant derived substrate usage and bacterial fitness), and stress response (neutralizing abiotic stress) were consistently enriched in genomes of all Gordonia strains. Moreover, using antiSMASH, we identified a total of 4,024 BGCs in the 225 genomes (Fig. 3B). The predominant BGC types in our dataset are annotated as: nonribosomal peptide synthetase (NRPS) (1,596 BGCs), terpene (447 BGCs), polyketide synthase (PKS) (282 BGCs), ribosomally synthesized and post-translationally modified peptide (RiPP) (267 BGCs).

Pangenome analyses of Gordonia

The 225 Gordonia strains had a pangenome size of 17,170 genes. Based on this dataset, the core genome was composed of 1,393 genes (genes shared by more than 99% strains), the accessory genome was composed of 15,547 genes, and the unique genome (strain-specific genes) was composed of 230 genes (Supplementary Fig. S3). To assess how pan-genome size and core-genome content scale with the number of Gordonia strains, we generated rarefaction curves for all 225 genomes and fitted them to a Heap’s law model (Fig. 4). The fitted exponent (α = 0.82) is < 1, indicating an open pan-genome whereby each newly added genome contributes novel genes and the curve fails to reach saturation. Moreover, the pangenome fluidity value was , indicating that Gordonia genomes differ 22% on average. By and large, moderate openness and fluidity of the pangenome were typically associated with Gordonia inhabiting multiple environments and showing evidence of the exchange of genetic material [55].

Fig. 4.

The pan-genome of genus Gordonia. A Curves depicted the correlation between genome numbers and pan (blue) and core (yellow) genome sizes. The alpha exponent of Heap’s Law was used to infer whether a pangenome is open or closed. Thus, if α (alpha) < = 1, the pan-genome is open. In contrast, α > 1 represents a closed pan-genome. This coefficient was computed using the heaps function of the micropan R package (See Methods). B Pangenome structure of genus Gordonia

To investigate the functional divergence in three different pangenome sets (Core, accessory, and unique genomes), we compared the frequency of genes assigned to KEGG pathways and COG functional categories using eggNOG-mapper, and Fisher’s exact tests were conducted between each frequency of KEGG pathway and COG category within each pangenome set. As excepted, COG J (translation, ribosomal structure and biogenesis) was significantly enriched in the core genome (adjusted P < 0.05 & Odd ratio > 1) (Fig. 5A). The core genes were also significantly assigned (p adjust < 0.05) to metabolics (F: nucleotide transport and metabolism and H: coenzyme transport and metabolism) and cellular processes and signaling (D: Cell cycle control, cell division, chromosome partitioning and O: Posttranslational modification, protein turnover, chaperones). However, with the exception of COG S (Function unknown), no COGs were found with either frequent significant enrichment or odd ratio > 1 in the unique and accessory genomes (Supplementary Table S4).

Fig. 5.

Functional enrichments by COG functional category and KEGG pathway in the core and accessory genomes. The distribution of odds ratios is shown. Only COGs and KEGG pathways showing positive enrichment (Fisher’s exact test, adjusted P under Bonferroni correction < 0.05) and with mean odd ratios > 1 are shown. COG “S: Function unknown” is not shown

In the similar analysis of KEGG pathways, core genome enriched various metabolism pathways (Odd Ratio > 1 and adjust P value < 0.05), including central and secondary metabolites pathway (Fig. 5B). Additionally, two translation-related pathways (ribosome and aminoacyl − tRNA biosynthesis) were also detected to significantly enrich in core genomes, consistent with the J COG previously found enriched in core genomes. In contrast, no pathways were found to be significantly enriched in a majority of accessory or unique genomes (Supplementary Table S5). Overall, this functional analysis suggests that the core genomes of Gordonia spp. were likely enriched for metabolic and translational functions, while non-core genes might draw from a wider variety of relatively niche functions.

Species descriptions

Description of Gordoniaaltitudinis sp. nov.

Gordonia altitudinis (al.ti.tu'di.nis. L. gen. fem. n. altitudinis, of a high place).

Cells are Gram-stain-positive, non-motile, aerobic and rod-shaped. Colonies on TSA are off-white, marginal irregular, opaque and rough, approximately 1 mm in diameter after growing for 3 days at 28℃ and pH 7.0. Growth occurs between 4℃ and 28℃ with optimum growth at 28℃. The pH range for growth is pH 6.0–8.0 (optimally at pH 7.0). NaCl is not required for growth, but growth occurs in the presence of 0–5% NaCl (w/v). Positive for catalase activity and hydrolysis of gelatin, but negative for oxidase reaction, H₂S production, hydrolysis of starch and nitrate reduction. Positive for alkaline phosphatase, lipase (C14), leucine aminopeptidase, valine aminopeptidase, cystine aminopeptidase, trypsin, α-chymotrypsin, α-galactosidase, β-galactosidase, β-glucuronidase, α-glucosidase, β-glucosidase, N-acetyl-β-glucosaminidase, α-mannosidase and α-fucosidase. Acetic acid, D-glucuronic acid, glucuronamide, α-keto-glutaric acid and pectin were utilized as sole source of carbon. Acid production from carbohydrates are D-fructose, D-mannitol, starch, and 2-ketogluconate. The major respiratory quinone is MK-9 (H₂). The primary polar lipid includes diphosphatidylglycerol, phosphatidylglycerol, phosphatidylethanolamine, phosphatidylinositol mannoside. The major fatty acids are C_16:0, summed feature 4 (iso-C_15:0 2-OH/trans-C_16:1ω9c), C_18:0 10-methyl, C_18:1ω9c, C_14:0 and C_19:0.

The type strain, CPCC 205333 ^T (= CS487^T = KCTC 49415 ^T), was isolated from a soil sample collected from Cangshan Mountain, Yunnan province, China. The genomic DNA G + C content of the type strain is 62.4% and genome size is 4.5 Mbp. The 16S rRNA gene and whole genome sequence of strain CPCC 205333 ^T are publicly available through the accession numbers PP657641 and JBCLVI000000000, respectively.

Description of Gordonia ligustrum sp. nov.

Gordonia Ligustrum (li.gus'trum. L. gen. fem. n. ligustrum, of Ligustrum, referring to the botanical genus name of Ligustrum lucidum, from the rhizosphere of which the strain was isolated).

Cells are Gram-stain-positive, non-motile, aerobic and rod-shaped. Colonies on TSA are orange, convex, and opaque after growing for 3 days at 28℃ and pH 7.0. Growth occurs between 10℃ and 42℃ with optimum growth at 15–37℃. The pH range for growth is pH 5.0–11.0 (optimally at pH 7.0–8.0). NaCl is not required for growth, but growth occurs in the presence of 0–10% NaCl (w/v). Positive for catalase activity and nitrate reduction, but negative for oxidase reaction, H₂S production, hydrolysis of starch and hydrolysis of gelatin. Positive for alkaline phosphatase, esterase (C4), esterase lipase (C8), lipase (C14), leucine aminopeptidase, valine aminopeptidase, cystine aminopeptidase, trypsin, α-chymotrypsin, acid phosphatase, naphthol-AS-B1-phosphohydrolase, α-galactosidase, β-galactosidase, β-glucuronidase, α-glucosidase and β-glucosidase. Acetic acid, D-aspartic acid, N-acetyl neuraminic acid,, propionic acid can be used as the sole carbon. Acid production from carbohydrates are amygdalin, arbutin, D-arabinose, D-fructose, D-galactose, D-glucose, D-mannitol, D-mannose, D-ribose, D-sorbitol, D-trehalose, D-xylose, erythritol, glycerol, inositol, methyl α-D-glucopyranoside, N-acetylglucosamine, salicin, xylitol. The respiratory quinone is MK-9 (H₂). The cellular polar Lipid system includes diphosphatidylglycerol, phosphatidylglycerol, phosphatidylethanolamine, phosphatidylinositol, and phosphatidylinositol mannoside.. The major fatty acids are summed feature 4 (iso-C_15:0 2-OH/trans-C_16:1ω9c), C_16:0, C_18:010-methyl, C_18:1ω9c and C_14:0.

The type strain, CPCC 205515 ^T (= I21A-00135 ^T = KCTC 59069 ^T), was isolated from a rhizosphere soil sample of Ligustrum lucidum in Chuxiong Yi autonomous prefecture, Yunnan province, China. The genomic DNA G + C content of the type strain is 66.5% and genome size is 6.2 Mbp. The 16S rRNA gene and whole genome sequence of strain CPCC 205515 ^T are publicly available through the accession numbers PP657644 and JBCLVJ000000000, respectively.

Description of Gordonia pistacia sp. nov.

Gordonia pistacia (pis.ta'cia. L. gen. fem. n. pistacia, of Pistacia, referring to the botanical genus name of Pistacia weinmannifolia, from the rhizosphere of which the strain was isolated).

Cells are Gram-stain-positive, non-motile, aerobic and rod-shaped. Colonies on TSA are cream-coloured, convex, smooth and opaque after growing for 3 days at 28℃ and pH 7.0. Growth occurs between 4℃ and 42℃ with optimum growth at 20–37℃. The pH range for growth is pH 6.0–11.0 (optimally at pH 7.0–8.0). NaCl is not required for growth, but growth occurs in the presence of 0–5% NaCl (w/v). Positive for catalase activity and nitrate reduction, but negative for oxidase reaction, H₂S production, hydrolysis of starch and hydrolysis of gelatin. Positive for alkaline phosphatase, esterase (C4), esterase lipase (C8), lipase (C14), leucine aminopeptidase, valine aminopeptidase, cystine aminopeptidase, acid phosphatase, naphthol-AS-B1-phosphohydrolase and α-glucosidase. D-Arabitol, D-fructose, D-galactose, D-glucose, D-mannitol, D-mannose, D-sorbitol, and myo-Inositol can be used as the sole carbon sources for energy and growth. Acid produced from 2-ketogluconate, D-fructose, and D-mannitol. The respiratory quinone is MK-9 (H₂). The cellular polar lipid system includes diphosphatidylglycerol, phosphatidylglycerol, phosphatidylethanolamine and phosphatidylinositol mannoside. The major fatty acids are C_16:0, summed feature 3 (C_16:1 ω7c/C_16:1ω6c), C_18:1ω9c, C_18:0 10-methyl and C_14:0.

The type strain, CPCC 206044 ^T (= A305^T = KCTC 59071 ^T) was isolated from a rhizosphere soil sample of Pistacia weinmanniifolia in Honghe Hani and Yi autonomous prefecture, Yunnan province, China. The genomic DNA G + C content of the type strain is 66.5% and genome size is 5.8 Mbp. The 16S rRNA gene and whole genome sequence of strain CPCC 206044 ^T are publicly available through the accession numbers PP657647 and JBCLVL000000000, respectively.

Conclusions

In this work, we isolated and characterised three novel Gordonia strains from the soil samples. Based on physiological, biochemical, and taxogenomics characteristics, strains CPCC 205333 ^T, CPCC 205515 ^T, and CPCC 206044 ^T represent three novel species of the genus Gordonia for which the names Gordonia altitudinis sp. nov., Gordonia ligustrum sp. nov., and Gordonia pistacia sp. nov. were proposed. Functional genome analysis revealed that Gordonia spp. have genetically conserved dual functional capacities: beneficial to plant and pathogens to human. They harbored extensive plant growth-promoting (PGP) traits specializing in nutrient acquisition (nitrogen fixation, potassium solubilization, and iron uptake) and abiotic stress tolerance, while simultaneously having virulence-associated modules for host immune modulation and biofilm-mediated adhesion. Moreover, the open pangenome architecture (α = 0.82) with moderate fluidity (0.22 ± 0.09) reflects widely ecological adaptation for Gordonia. Functional enrichments suggested core genomes of Gordonia eriched essential translational and metabolic pathways, while the accessory and unique genome provided genomic plasticity for niche-specific adaptations, as evidenced by the absence of conserved functional signatures in non-core compartments.

Competing interests

The authors declare no competing interests.

Abbreviations

NJ

Neighbor-joining

ML

Maximum-likelihood,

MP

Maximum parsimony

MEGA

Molecular Evolutionary Genetics Analysis

OGRIs

Overall genome relatedness indices

ANI

Average nucleotide identity

dDDH

Digital DNA-DNA hybridization

SOG

Single-copy rthologroups

ISP2

International Streptomyces Project 2 agar

PYG

Peptone yeast glucose

TSA

Tryptone soy agar

TSB

Tryptone soy broth

LB

Luria–Bertani

R2A

Reasoner’s 2A agar

GYM

Glucose yeast malt agar

NA

Nutrient agar

HPLC

High Performance Liquid Chromatography

MIDI

Microbial Identification System

COG

Cluster of Orthologous Groups

KEGG

Kyoto Encyclopedia of Genes and Genomes

VF

Virulence factor

PGPT

Plant growth-promoting traits

BGCs

Biosynthetic gene clusters

NRPS

Nonribosomal peptide synthetase

PKS

Polyketide synthase

RiPP

Ribosomally synthesized and post-translationally modified peptide

Authors’ contributions

C-JL carried out the experiments and data analysis; JZ, JS and L-YY collected the environmental samples and carried out partial data analysis; H-HC and Y-QZ conceived the research, analysed the data and prepared the manuscript. All authors contributed to the article and approved the submitted version.

Funding

This research was supported by National Natural Science Foundation of China (32170021), CAMS Innovation Fund for Medical Sciences (CIFMS, 2021-I2M-1–055) and the National Infrastructure of Microbial Resources (NIMR-2024–4).

Data availability

The 16S rRNA gene and the whole genomic sequences of strains CPCC 205333 T, CPCC 205515 T and CPCC 206044 T have been deposited in the NCBI Assembly and NCBI GenBank database, respectively. The accession numbers of 16S rRNA gene and the whole genome sequence of strain CPCC 205333 T are PP657641 and JBCLVI000000000, respectively; the accession numbers of 16S rRNA gene and the whole genome sequence of strain CPCC 205515 T are PP657644 and JBCLVJ000000000, respectively; the accession numbers of 16S rRNA gene and the whole genome sequence of strain CPCC 206044 T are PP657647 and JBCLVL000000000, respectively.

Declarations

Ethics approval and consent to participate

This research did not contain any studies with animals performed by any of the authors.