Genome-based taxonomy of Burkholderia sensu lato: Distinguishing closely related species

Abstract

The taxonomy of Burkholderia sensu lato (s.l.) has been revisited using genome-based tools, which have helped differentiate closely related species. Many species from this group are indistinguishable through phenotypic traits and 16S rRNA gene sequence analysis. Furthermore, they also exhibit whole-genome Average Nucleotide Identity (ANI) values in the twilight zone for species circumscription (95-96%), which may impair their correct classification. In this work, we provided an updated Burkholderia s.l. taxonomy focusing on closely related species and give other recommendations for those developing genome-based taxonomy studies. We showed that a combination of ANI and digital DNA-DNA hybridization (dDDH) applying the universal cutoff values of 95% and 70%, respectively, successfully discriminates Burkholderia s.l. species. Using genome metrics with this pragmatic criterion, we demonstrated that i) Paraburkholderia insulsa should be considered a later heterotypic synonym of Paraburkholderia fungorum; ii) Paraburkholderia steynii differs from P. terrae by harboring symbiotic genes; iii) some Paraburkholderia are indeed different species based on dDDH values, albeit sharing ANI values close to 95%; iv) some Burkholderia s.l. indeed represent new species from the genomic viewpoint; iv) some genome sequences should be evaluated with care due to quality concerns.

Keywords:
Average Nucleotide Identity; Burkholderia; Paraburkholderia; Digital DNA-DNA hybridization; heterotypic synonyms

Introduction

In 1973, Palleroni et al. (1973) carried out ribosomal ribonucleic acid (rRNA)-DNA hybridization studies that indicated that the Pseudomonas genus was composed of five RNA homology groups (I-V). In 1992, Yabuuchi et al. (1992) proposed the creation of the genus Burkholderia for the RNA homology group II based on the 16S rRNA gene sequence, DNA-DNA hybridization (DDH) values, phenotypic characteristics, and composition of cellular lipids and fatty acids. Twelve years later, with as little as 34 validly described species, Burkholderia already exhibited a complex taxonomy. Due to resolution limitations of the 16S rRNA gene sequence analysis, Payne et al. (2005) developed a recA gene-based identification for the genus. Further multilocus sequence analysis (MLSA) indicated the presence of at least two distinct lineages within the genus (Estrada-de Los Santos et al., 2013), which corroborated several previous works based on 16S rRNA gene sequences (Gyaneshwar et al., 2011). Applying phylogenomics and analyzing 42 conserved molecular markers of sequence insertions or deletions (CSIs), Sawana et al. (2014) confirmed the presence of at least two different clades within Burkholderia. These authors proposed the division of the genus and the creation of the genus Paraburkholderia.

The increasing availability of genomic data allowed the recognition of differential CSIs and the reclassification of some Paraburkholderia species to the new genus Caballeronia (Dobritsa and Samadpour, 2016). In addition, these authors provided the emended description of several species. In all these studies, Burkholderia andropogonis consistently formed a distinct clade. Through a combination of phylogeny based on 30 conserved genes and genome metrics, the evolutionary distance of B. andropogonis led to the creation of the monotypic genus Robbsia (Lopes-Santos et al., 2017). More recently, phylogenomics associated with Average Nucleotide Identity (ANI) calculations showed the necessity of additional divisions of the genus and the creation of the new genera Mycetohabitans and Trinickia (Estrada-de Los Santos et al., 2018). Moreover, polyphasic approaches led to the proposal of the novel Pararobbsia genus to accommodate two species, Pararobbsia silviterrae and Pararobbsia alpina comb. nov. (Lin et al., 2020). Therefore, Burkholderia sensu lato (s.l.) is currently composed of Burkholderia sensu strictu (s.s.), Paraburkholderia, Caballeronia, Mycetohabitans, Trinickia, Robbsia, and Pararobbsia species (Bach et al., 2022b).

Correctly identifying an isolate is helpful for many research areas since this information provides insights into the biotechnological potential, biosafety, clinical outcomes, ecological roles, and evolutionary origin of features. Accurate species assignment is especially critical for members of the Burkholderia cepacia complex (Bcc), which includes strains competent in producing a myriad of bioactive compounds with biotechnological applications but that may also cause worrisome lung infections in patients with cystic fibrosis (Bach et al., 2022a). Some pathogenic Bcc species exhibit higher patient-to-patient transmissibility and the disease caused by different species have distinct clinical outcomes (Pope et al., 2010). Another example within Burkholderia s.s. species are Burkholderia mallei and Burkholderia pseudomallei, which cause the zoonotic diseases glanders and melioidosis, respectively. While glanders is primarily a horse disease, melioidosis may affect humans and other animals (Godoy et al., 2003). Furthermore, distinct rhizobial Paraburkholderia species can establish a beneficial symbiotic relationship and form root nodules for nitrogen fixation in different legume species (Mavima et al., 2022). The species mentioned above are hardly distinguished by the evaluation of 16S rRNA or recA gene sequences, highlighting the importance of using genome metrics to differentiate closely related species.

Since 1987 the ad hoc committee of the International Committee for Systematic Bacteriology (current International Committee on Systematics of Prokaryotes, ICSP) agrees that the DNA sequence should be the reference standard to determine taxonomy (Wayne et al., 1987). At that time, the recommended procedure for defining a species were two measures of genetic relatedness: the change in the melting temperature (ΔTm) of heteroduplex DNA and the extent of DDH. Two strains should belong to the same species if presenting both 70% or more DDH relatedness and 5 ºC or less of ∆Tm. However, these procedures show significant technical drawbacks being prone to giving imprecise results (Goris et al., 2007; Sant’Anna et al., 2019). The genome metrics ANI and digital DDH (dDDH) are surrogates for the ΔTm and DDH, using the thresholds of 95 and 70%, respectively (Goris et al., 2007; Meier-Kolthoff et al., 2013; Chun et al., 2018), providing more reliable and portable taxonomic results in substitution to wet-lab genomic relatedness comparisons.

Another advantage of using genome metrics in taxonomy is adopting a universal cutoff value for species delimitation that defines an objective criterion for species circumscription and standardizes communication among the scientific community of different areas. However, different thresholds are suggested and might be accepted to delineate species that show diagnostic distinct phenotypic traits as is the case of closely related Paraburkholderia species (Mullins and Mahenthiralingam, 2021). Moreover, genome metrics are increasingly being used to split known species into novel ones (Velez et al., 2023) and detect the presence of synonyms, which are species described with different names that belong to the same species (Madhaiyan et al., 2022). These peculiarities and updates complicate the correct species assignment for those unfamiliar with the taxonomy of the group. In previous work, we studied the pangenome and provided a genome-based taxonomy of Burkholderia s.l. species (Bach et al., 2022b). In this work, we provided an updated Burkholderia s.l. taxonomy focusing on closely related species to search for synonyms and give other recommendations for those developing genome-based taxonomy studies.

Material and Methods

Burkholderiales genome sequences were obtained from the RefSeq NCBI database in August 2022 and subjected to a cluster analysis (^{Table S1}) to scan for synonyms. Briefly, the genome metrics were calculated with FastANI (Jain et al., 2018) and clustered with ProKlust (Volpiano et al., 2021), a graph-based approach for downstream analysis of large identity matrices. Genomes that formed clusters with ANI ≥ 95% were selected for further analysis.

Reference genome sequences of Paraburkholderia and selected Burkholderia and Caballeronia were downloaded from the RefSeq database up to March 2023 and quality was checked using CheckM (Parks et al., 2015) (^{Table S2}). Genome metrics were calculated using FastANI (Jain et al., 2018), JSpecies (ANIb and ANIm), and Genome-to-Genome Distance Calculator (GGDC) web tools at http://jspecies.ribohost.com/jspeciesws/ and http://ggdc.dsmz.de/home.php, respectively. Two genomes were considered belonging to the same species if both metrics showed results above the thresholds recommended for species delineation. Phylogenomics was performed as described previously (Bach et al., 2022b). Briefly, genomes were annotated with Prokka (Seemann, 2014) and single-copy orthologous proteins were obtained by the intersection of results provided by three clustering algorithms implemented in the GET-HOMOLOGUES tool using default parameters (Contreras-Moreira and Vinuesa, 2013). The phylogeny was reconstructed following the GET-PHYLOMARKERS pipeline, using the maximum likelihood approach and estimating the best tree through IQ-TREE (Vinuesa et al., 2018).

Results and Discussion

Our previous pangenome and phylogenomic study of Burkholderia s.l. has already indicated the presence of many new species and synonyms within the group (Bach et al., 2022b). Our work corroborated previous results regarding Bcc members (Jin et al., 2020) and agreed with a concomitant study of 4,000 Burkholderia s.l. genome assemblies (Mullins and Mahenthiralingam, 2021). To formally describe a new bacterial species according to the rules of the ICSP, the type-strain should be deposited in two international culture collections (Garrity et al., 2015). Thus, there is a gap between finding a potentially new species in a genome dataset and formally describing it. After the effective description, the new species name should be validated by the ICSP. All not validly published species should be mentioned within quotation marks, following the List of Prokaryotic names with Standing in Nomenclature (LPSN). In this work, we provided an updated taxonomy of Burkholderia s.l. and used genome metrics to evaluate whether recently described novel species could be validly accepted and if closely related species should be considered synonyms.

Many genome metrics are available for species delineation (Sant’Anna et al., 2019), especially variations of ANI (Palmer et al., 2020). Whole-genome ANI pairwise comparisons are most commonly performed through BLASTn or MUMmer alignment algorithms, named ANIb and ANIm, respectively (Richter and Rosselló-Móra, 2009). While ANIm is advantageous for preliminary analysis of extensive sequence data, ANIb shows more robust results (Richter and Rosselló-Móra, 2009). To follow the previous recommendation of the ICSP to evaluate taxonomic relatedness with two different metrics (Wayne et al., 1987), the use of both ANI and dDDH should be considered since they measure different genome properties especially depending on the chosen dDDH formula (Auch et al., 2010; Li et al., 2015; Volpiano et al., 2021). It is important to note that GGDC provides dDDH values calculated with different formulae (Meier-Kolthoff et al., 2013), dDDH formula 1 (GBDP formula d ₀ ), formula 2 (GBDP formula d4), and dDDH formula 3 (GBDP formula d ₆ ). Formula 2 is recommended for evaluating incompletely sequenced genomes, which comprise a large proportion of the current sequences available in genomic databases.

Our previous work agreed with Jin et al. (2020) findings, which indicated that dDDH was more discriminatory for Burkholderia species delineation. Both studies showed that some Burkholderia type species exhibit ANI values above the 95% threshold while sharing dDDH values below the 70% cutoff for species delineation. Diagnostic differential phenotypic traits corroborate the validity of these strains as distinct species. ANI values above the proposed threshold of 95% were also observed among pairwise comparisons of different type species when Mullins and Mahenthiralingam (2021) investigated 4,000 Burkholderia s.l. genome assemblies. These authors recommended the adjustment of the ANI threshold to 96% to discriminate Paraburkholderia fungorum from Paraburkholderia agricolais; Paraburkholderia caledonica from Paraburkholderia strydomiana; Paraburkholderia phytofirmans from Paraburkholderia dipogonis; Paraburkholderia hospita from Paraburkholderia steynii and Paraburkholderia terrae, while the last two could be differentiated using the cutoff value of 97%. Here we show that these species are effectively discriminated by evaluating both ANI and dDDH with the universal threshold values of 95% and 70%, respectively (Table 1). Moreover, phylogenomics also separated these species into distinct clades (Figure 1). Exceptions will be highlighted below.

Table 1 -
Pairwise whole genome comparisons performed in this work and taxonomic recommendations for Burkholderia sensu lato strains.

Putative type strains	Reference strains	JSpecies		FastANI	dDDH			Proposition
		ANIm	ANIb		formula 1	formula 2	formula 3
Paraburkholderia insulsa LMG 28183	Paraburkholderia fungorum LMG 16225	98.51	97.57	98.42	74.80	85.80	79.40	P. insulsa is a later heterotypic synonym of P. fungorum
Paraburkholderia agricolaris BaQS159	Paraburkholderia fungorum LMG 16225	95.83	94.06	95.55	56.70	64.60	59.10	Different species
Paraburkholderia dipogonis ICMP 19430	Paraburkholderia phytofirmans PsJN	95.95	94.17	95.80	55.20	66.10	57.90	Different species
Paraburkholderia terrae DSM 17804	Paraburkholderia hospita DSM 17164	95.42	93.83	95.17	61.70	62.00	63.20	Different species
Paraburkholderia steynii HC1.1ba	Paraburkholderia hospita DSM 17164	95.13	94.2	94.56	54.40	59.70	56.10	Different species
Paraburkholderia steynii HC1.1ba	Paraburkholderia terrae DSM 17804	96.78	96.0	96.52	60.60	71.2	63.90	Different species
Paraburkholderia strydomiana WK1.1f	Paraburkholderia caledonica LMG19076	95.86	94.42	95.63	65.90	65.50	67.70	Different species
Paraburkholderia aspalathi LMG 27731	Paraburkholderia nemoris LMG 31836	95.73	94.22	95.64	61.8	64.9	63.9	Different species
Paraburkholderia dioscoreae PDMSB31	Paraburkholderia xenovorans LB400	94.9	92.8	94.70	56.3	58.4	57.6	Different species
Paraburkholderia pallida 7MH5	Paraburkholderia oxyphila NBRC 105797	95.65	93.53	95.32	56.8	63.6	59	Different species
“Paraburkholderia atlantica” CCGE1002	Paraburkholderia youngii JPY169	94.92	93.14	94.57	57.7	59.6	59	“P. atlantica” could be validly accepted**
“Paraburkholderia caffeinitolerans” LMG 28688	“Paraburkholderia dokdonensis” DCR-13	99.94	97.82*	99.77	50.1	98.4	56.9	“P. caffeinitolerans” could be validly accepted**; “P. dokdonensis” belongs to the same species
Burkholderia oklahomensis C6786	“Burkholderia mayonis” BDU6	95	94.23	95.08	68.6	58.5	68.5	“B. mayonis” could be validly accepted**
Burkholderia orbicola TAtl-371	Burkholderia cenocepacia NCTC 13227	95.12	93.94	95.21	69.60	59.60	69.70	Different species
“Burkholderia semiarida” CCRMBC74	Burkholderia cenocepacia NCTC 13227	94.38	93.8	94.19	59.5	55.5	59.8	“B. semiarida” could be validly accepted**
“Burkholderia semiarida” CCRMBC74	Burkholderia orbicola TAtl-371	94.35	93.9	94.36	70.3	55.3	69.2
“Burkholderia semiarida” CCRMBC74	“Burkholderia sola” CCRMBC51	94.36	93.86	94.44	73.7	55.9	72.2
“Burkholderia sola” CCRMBC51	Burkholderia cenocepacia NCTC 13227	94.61	94.04	94.53	62.1	57.1	62.5	“B. sola” could be validly accepted**
“Burkholderia sola” CCRMBC51	Burkholderia orbicola TAtl-371	94.78	94.41	94.83	70.7	58.8	70.5
Burkholderia plantarii ATCC 43733	“Burkholderia perseverans” INN12	95.47	94.73	95.98	76.7	60.6	76.1	“B. perseverans” could be validly accepted**
Burkholderia mallei ATCC 23344	Burkholderia pseudomallei ATCC 23343	99.17	98.8*	99.29	83.1	92.5	87.6	Same species; B. mallei exhibits a genome reduction
“Burkholderia reimsis” BE51	Burkholderia cepacia ATCC 25416	97.65	96.84	97.48	78.10	77.90	80.90	“B. reimsis” was misidentified. Strain BE51 belongs to B. cepacia
Caballeronia terrestris LMG 22937	Caballeronia humi LMG 22934	94.99	93.65	94.60	69.60	59.30	69.60	Different species
	Caballeronia humi KEMC 7302-068	84.6*	75.63*	80.53	17	22.10	16.9	The genomic assembly GCF_007474635.1 of KEMC 7302-068 should be avoided
Caballeronia humi LMG 22934	Caballeronia humi KEMC 7302-068	84.6*	75.6*	80.56	17.30	22	17.20

• Bold values are above the cutoff for species delineation; * Alignment fractions below 65%; **Could be validly accepted as a new species once other ICSP requirements are met.

Figure 1 -
Phylogenetic tree of representative Paraburkholderia spp. based on the alignment of 273 orthologous protein sequences recognized by GET-HOMOLOGUES and reconstructed through the maximum likelihood approach of GET-PHYLOMARKERS. Borderline species investigated in this work were highlighted in bold. Burkholderia cepacia ATCC 25416^T was set as the outgroup. All bootstrap values are shown.

Burkholderia s.l. genomes exhibiting ANI borderline values were detected with FastANI followed by ProKlust analysis (^{Figure S1} and ^{Table S3}) and pangenome analyses (^{Figure S2}). By further using ANI and dDDH, our results confirmed i) the presence of a synonym within Burkholderia s.l.; ii) reinforced that some Paraburkholderia are distinct species that share ANIb values >95%; and iii) recommended the acceptance of new species. Our results and some other recommendations for the taxonomic study of this group are detailed below.

ANI and dDDH discriminate Paraburkholderia spp.

Some Paraburkholderia species share ANI values within the threshold for species circumscription (95-96%), forming identity clusters in the ProKlust analysis (^{Figure S1}, ^{Table S3}): Paraburkholderia insulsa, P. agricolaris, and P. fungorum; P. dipogonis and P. phytofirmans; P. hospita, P. steynii, and P. terrae; P. strydomiana and P. caledonica; Paraburkholderia aspalathi and Paraburkholderia nemoris; Paraburkholderia pallida and Paraburkholderia oxyphila (Table 1). The close relationship of these species could also be observed through phylogenomics (Figure 1). However, most of these genome sequences share dDDH values below the 70% threshold.

In accordance with our results, the authors who described the forest soil isolate P. nemoris as a new species (Vanwijnsberghe et al., 2021), observed orthoANI values above the species threshold and dDDH values below the species cutoff when compared to P. aspalathi (former Burkholderia aspalathi) (Mavengere et al., 2014). Similarly, the forest soil species P. pallida and P. oxyphila (former Burkholderia oxyphila) shared ANI values above 95% and dDDH values below 65%. Noteworthily, the use of ANI metrics alone would not be enough to separate these closely related species. Therefore, we highlight that the combined investigation of ANI and dDDH is useful for discriminating closely related Paraburkholderia species.

Paraburkholderia agricolaris is a soil-dwelling amoebae symbiont (Brock et al., 2020), while P. terrae and P. hospita type strains were isolated from soil and have similar genomic features and eco-phenotypes, interacting with soil fungi (Pratama et al., 2020). Paraburkholderia steynii, P. strydomiana, and P. dipogonis are among the plant symbionts included in the symbiovars sv. africana owing to their capacity to nodulate Papilionoideae legumes from South Africa and New Zealand (Paulitsch et al., 2020b), leading to their classification into the sv. Papilionideae (Bellés-Sancho et al., 2023). The studies that described these strains as new species also mentioned high 16S rRNA gene and ANI identity values (Sheu et al., 2015; Beukes et al., 2019). For instance, P. steynii and P. terrae type strains and P. strydomiana and P. caledonica share a similarity of 100% in the 16S rRNA gene sequence (Beukes et al., 2019). Therefore, classifying strains into these species might be problematic without genome metrics. These authors could only differentiate the new species using conventional and digital DDH. Similarly, we showed that P. strydomiana and P. caledonica shared ANIb and dDDH values below species boundaries (Table 1) and should be considered different species following Wayne et al. (1987) rationale. These authors showed that genomically similar isolates formed monophyletic clades in the phylogenetic reconstructions with the new species, reinforcing them as distinct species. However, dDDH formula 2 among P. steynii and P. terrae was above the threshold for species circumscription (71.2%; 68.4 to 74.3% of confidence interval).

An exception given to P. steynii and P. terrae

By evaluating the taxonomic status of P. terrae and P. steynii with care, we observed that they could not be differentiated by combining the evaluation of the universal threshold values for ANI and dDDH. The original work that describes P. steynii as a new species provides comparisons with P. terrae type strain showing some phenotypic differences, DDH values below 70%, and average dDDH values of 65.2% (Beukes et al., 2019). However, it is well known that phenotypic tests and the conventional DDH methodology are unreliable. Besides that, it is uncommon to consider the average of the three dDDH methodologies for species circumscription. These authors also highlight a remarkable difference among not only P. steyni and P. terrae, but also among P. strydomiana and P. caledonica: both P. strydomiana and P. steynii were able to nodulate the leguminous plant Hypocalyptus sophoroides, whereas the closely related type strain P. caledonica NBRC 102488^T and P. terrae NBRC 100964^T could not (Beukes et al., 2019). The inability of nodulation of these strains was claimed since the authors could not find the common nodulation loci, nodABCD, in their genomes.

To reinforce this finding, we annotated with Prokka all available genome sequences of P. steynii, P. terrae, P. hospita, Paraburkholderia caribensis, and other closely related species and performed pangenome analysis. Figure 2 shows that P. steynii HC1ba^T harbors a similar profile of nod and nif genes compared to two strains of P. caribensis, and the type strains of Paraburkholderia piptadeniae, Paraburkholderia franconis, Paraburkholderia diazotrophica, and Paraburkholderia sabiae, all South American mimosoid-nodulating species. Paraburkholderia hospita strains harbor a different nodD gene. A recent review describes differences in the origin of symbiotic genes of Paraburkholderia spp. isolated from nodules of South American and South African legumes (Bellés-Sancho et al., 2023). Regarding our taxonomic focus, we could show that P. terrae strains lack nodD and nif genes. Even though the type strain was characterized as a new nitrogen-fixing (diazotrophic) species by cultivating it in a nitrogen-free medium and amplifying the nifH gene through PCR (Yang et al., 2006), we could not confirm this data through genome analysis. The ability to nodulate legumes could be a significant difference among these closely related strains. Thus, these species are kept separated and constitute an exception for our dataset since they could not be differentiated using the universal cutoffs of ANI and dDDH. As mentioned above, an adjusted ANI threshold of 97% was proposed to delineate these species (Mullins and Mahenthiralingam, 2021).

Figure 2 -
Phylogenetic tree of Paraburkholderia strains and heatmap showing the presence (green) of the symbiotic genes nod, nif, and fix in each genome sequence. The phylogeny was based on the alignment of 88 orthologous protein sequences recognized by GET-HOMOLOGUES and reconstructed through the maximum likelihood approach of GET-PHYLOMARKERS. Type strains are shown in bold. Paraburkholderia phymatum STM815^T was set as the outgroup. All bootstrap values are shown.

A previous proposal reclassified these strains as P. terrae subspecies terrae and P. terrae subspecies steynii due to some differential phenotypic traits (Madhaiyan et al., 2022). However, we have some concerns about this proposition. The subspecies rules are less clear than the bacterial species circumscription. For instance, there is a proposal to delineate subspecies based on a dDDH threshold of 79% (Meier-Kolthoff et al., 2014), which is far from the value obtained for these strains. Moreover, the annotation file of P. steynii HC1ba^T has been recently removed from RefSeq, raising concerns regarding genome quality. Of note, the subspecies status is shown as not validated in the LPSN website, while the GTDB (Genome Taxonomy Database) considers them as synonyms. These issues could be clarified once more genomes of P. steynii strains are sequenced.

Considering that the genome of P. steynii should contain nod and nif genes, we suggest reclassifying strain YR281 to P. terrae due to the absence of nif and some nodD genes in its genome. Moreover, YR281 shares higher ANI values with other P. terrae strains (above 97%) than with P. steynii (96.3%). All suggested reclassifications of non-type strains of this work are shown in supplementary ^{table S2}. Two strains, P. terrae 19C8 and P. caribensis PCAR477, were grouped in clusters based on phylogenomic and ANI analyses (Figures 2 and ^S3). These strains showed ANIb values below 95% compared to other Paraburkholderia type species and shared ANI similarities of 98.7% among them (^{Figure S3}). Therefore, they belong to a new Paraburkholderia species.

Paraburkholderia insulsa as a later heterotypic synonym of Paraburkholderia fungorum

Our results indicated that these two species shared an ANIb value of 97.57%, ANIm of 98.51%, and dDDH values >74.8% (Table 1). The whole genome phylogenetic reconstruction also indicates that these species are highly similar, sharing the same most recent ancestor (Figure 1). Paraburkholderia fungorum was initially isolated from the white-rot fungus Phanerochaete chrysosporium, but was later found in various human and veterinary clinical samples (Coenye et al., 2001). It was described as a new species of Burkholderia in 2001 and then moved to the new genus Paraburkholderia according to phylogenetic clustering (Dobritsa and Samadpour, 2016). In 2015, Rusch et al. (2015) proposed the new species P. insulsa as a unique strain isolated at 30 m distance from an arsenic-rich hydrothermal vent in Papua Nova Guinea. This strain showed high 16S rRNA gene similarity with P. fungorum (99.8%), P. phytofirmans (98.8%), P. caledonica (98.4%), and Paraburkholderia sediminicola (98.4%), all previously belonging to Burkholderia. A few phenotypic differences were observed among them, including lipid composition, carbohydrate utilization, and enzyme profiles. However, DDH values indicated that P. insulsa PNG-April was a different species due to reassociation values below the 70% threshold (35-36.7% with P. fungorum DSM 17061^T, 10.3 and 20.5% with P. phytofirmans DSM 17436^T). Since DDH exhibits low reproducibility, this value may not be reliable. Besides that, minor phenotypic differences could be a result of intraspecies differences. Therefore, according to the genome similarities found here and in previous works (Mullins and Mahenthiralingam, 2021; Bach et al., 2022b), P. insulsa (Rusch et al., 2015) should be considered a later heterotypic synonym of P. fungorum (Coenye et al., 2001). Similarly, recent work has made the same proposition (Madhaiyan et al., 2022).

Phylogenomics and genome metrics indicate that “Paraburkholderia atlantica”, “Paraburkholderia caffeinitolerans”, “Paraburkholderia bonniea”, and “Paraburkholderia hayleyella” indeed represent new species

Our phylogenomic and ANI analyses (Figures 1 and ^S2) indicated that, at least from a genomic standpoint, the species “P. atlantica”, “P. bonniea”, “P. caffeinitolerans”, and “P. hayleyella” represent new species (Gao et al., 2016; Brock et al., 2020; Paulitsch et al., 2020a). Despite displaying near-cutoff ANI values, not only Paraburkholderia dioscoreae and Paraburkholderia xenovorans but also Paraburkholderia youngii and “P. atlantica” could be differentiated by dDDH. Hence, based on genome metrics, the Brazilian Atlantic Forest species “P. atlantica” (Paulitsch et al., 2020a) could be validly accepted as a new species once other ICSP requirements are met. Both “P. atlantica” and P. youngii were previously classified as Paraburkholderia tuberum sv. mimosae and formed clearly separated clusters in MLSA and ANI analyses. Interestingly, these species contain strains able to fix nitrogen and nodule mimosoid legumes of South and Central America, which also led to their classification into the symbiovar sv. atlantica (Mavima et al., 2022).

“Paraburkholderia caffeinitolerans” was isolated from a Chinese tea plantation soil and showed caffeine degrading abilities (Gao et al., 2016). Years later, a Korean strain isolated from the rhizosphere of Campanula takesimana was described as the new species “Paraburkholderia dokdonella” (Jung et al., 2019). However, genome metrics indicated that “P. dokdonella” is a later heterotypic synonym of “P. caffeinitolerans”, which could be validly accepted as a new species. This result corroborated previous findings (Mullins and Mahenthiralingam, 2021). Madhaiyan et al. (2022) have recently reinforced the proposal of “P. dokdonella” as a new species by correcting its name to “Paraburkholderia dokdonensis” and providing culture collection deposit certificates. However, we have some concerns regarding this genome sequence. “Paraburkholderia dokdonensis” shows an atypical genome size (4.4 Mbp) compared to other Paraburkholderia spp. (7-10 Mbp), which resulted in ANIb alignment fractions of 57%. It remains to be evaluated if this is due to a genome reduction or an anomalous genome assembly.

In general, bacterial endosymbionts, intracellular pathogens, or obligate pathogens harbor reduced genomes (González-Torres et al., 2019). For instance, Paraburkholderia agricolaris, “P. bonniea”, and “P. hayleyella” were isolated from the amoebae Dictyostelium discoideum and were found to remain in symbiosis during all host life stages. Both “P. bonniea” and “P. hayleyella” harbor reduced genome sizes probably related to gene losses commonly associated with an adaptation to the symbiotic lifestyle (Brock et al., 2020). Intriguingly, their G+C content (58.7 and 59.2%) was also lower than other Burkholderia s.l (^{Table S2}), except for Robbsia andropogonis, whose genome G+C content is 59.1%. ANI values between “P. bonniea” and “P. hayleyella” and R. andropogonis are 77 and 76.7%, respectively. “Paraburkholderia bonniea” and “P. hayleyella” lack culture collection deposit certificates, and thus are still not validly published.

Likewise, the fungal endosymbionts Mycetohabitans spp. exhibit genome sizes ranging from 3.2 to 3.8 Mbp. Another well-known example of genome reduction within Burkholderia s.l. is B. mallei, the obligate pathogen that causes glanders in horses, occasionally also infecting humans and other animals (Godoy et al., 2003). Since the proposal of Burkholderia as a new genus in 1992, B. mallei and B. pseudomallei are recognized as a single species (Yabuuchi et al., 1992). This could be observed using ProKlust FastANI clusters, phylogenomics, and additional genome metrics (Figure ^S4 and Tables 1 and ^S3). Ideally, these strains should be reclassified as B. mallei subspecies mallei and B. mallei subspecies pseudomallei. However, they are historically kept as different species due to the differences in the disease they cause. While B. mallei is an obligate pathogen mainly affecting horses, B. pseudomallei opportunistically causes melioidosis in humans and other animals (Godoy et al., 2003). Burkholderia mallei ATCC 23344^T harbors a genome of 5.8 Mbp, whereas the genome of B. pseudomallei ATCC 23343^T has a size of 7 Mbp.

Phylogenomics and genome metrics indicate that “Burkholderia mayonis”, “Burkholderia semiarida” and “Burkholderia sola” could be validated as new species

“Burkholderia mayonis” is a soil isolate from tropical northern Australia and was characterized as a new species of the B. pseudomallei complex through biochemical and genomic differences (Hall et al., 2022). Even though “B. mayonis” shares ANI values of 95% with Burkholderia oklahomensis, dDDH discriminated them as separate species (Table 1). “Burkholderia perseverans” belongs to the cluster of plant pathogens such as Bukholderia glumae, Bukholderia gladioli, and Bukholderia plantarii (^{Figure S4}). “Burkholderia perseverans” was isolated from leaf litter of Brazilian’s Restinga ecosystem and exhibits antifungal properties. It was differentiated from B. plantarii type strain through ANI, dDDH, and biochemical tests, including lack of growth in TSA medium containing 3% NaCl, citrate assimilation, and β-galactosidase activity (Andrade et al., 2021). These authors also characterized another two “B. perseverans” isolates that formed a separated cluster in the phylogenetic tree.

Here we show that differentiating “B. perseverans” from B. plantarii using genome metrics requires attention. Despite the widely used ANIb and dDDH formula 2 being below the species circumscription cutoffs, this is not the case for ANIm and the other two dDDH formulae (Table 1). If the criterium of using two metrics that measure different genomic properties is considered (Wayne et al., 1987), we should consider ANI and dDDH formulae 1 or 3. Formula 2 is similar to ANI and is especially recommended when comparing draft genomes, which is not true for these sequences (^{Table S2}). Therefore, using this criterium, we should consider B. plantarii and “B. perseverans” the same species. However, we agree with the proposal of the new species (Table 1) due to the following data: i) both genomes are complete and result in ANIb values below 95%. ANIb is more robust and the current preferential ANI methodology; ii) dDDH formula 2 is below 70% and is the developer’s recommended formula; iii) other “B. perseverans” isolates formed a separated clade in the phylogenetic tree; and iv) they exhibit diagnostic phenotypic traits (Andrade et al., 2021). This case highlights the eminent necessity for establishing more precise and detailed criteria for species delineation using genome metrics by the ICSP.

In 1997, multiple genomovars (I-V) were recognized within B. cepacia isolated from cystic fibrosis patients by whole-cell protein electrophoresis, DDH, and other phenotypic traits (Vandamme et al., 1997). Burkholderia cenocepacia was later proposed for “B. cepacia genomovar III”, encompassing the recA gene lineages IIIA, IIIB, IIIC, and IIID (Vandamme et al., 2003). These authors have already described DDH values of 58-83% among strains of lineages IIIA and IIIB. In our previous pangenome analysis (Bach et al., 2022b), these two clusters were also clearly separated by ANI, dDDH and phylogenomics. The cluster containing the type strain was called B. cenocepacia BCC08, while the other cluster was called Burkholderia sp. BCC05, corresponding to recA lineages IIIA and IIIB, respectively. ANI values between the strains from these groups were above 97.7%, while values among groups were in the twilight zone (95.4-95.7%). Likewise, dDDH values of type strain B. cenocepacia NCTC13227^T were above 89.9% within-cluster BCC08 (IIIA) and below 60.4% compared to strains belonging to cluster BCC05 (IIIB). Therefore, we suggested that BCC05 represents a new species.

In addition to recognizing these two clusters through ANIb and dDDH pairwise comparisons, Wallner et al. (2019) observed differences in the genome size, G+C content and protein coding sequence regions among them. More importantly, these authors evaluated gene content and observed that genomes from cluster BCC05 (IIIB) lack virulence traits present in BCC08 (IIIA) genomes, which was composed of clinical strains. They proposed the new species “Burkholderia servocepacia” to accommodate strains isolated from diverse sources (e.g., hospital, agricultural soil) that formed cluster IIIB. In 2022, Morales-Ruíz et al. (2022) performed genome metrics, phenotypic, and chemotaxonomic characterizations to accomplish the current mandatory rules for bacterial species descriptions and renamed “B. servocepacia” to Burkholderia orbicola, which is currently a validly accepted name. Here we extended our previous analyses to include the strain proposed as type, TAtl371. This strain clustered within BCC05 in ANI, dDDH, and phylogenomic analyses (Table 1, Figures 3 and ^S5).

Similarly, we have previously shown that the Bcc strains AZ4-2-10-S1D7 and XXVI belonged to new species, which we had called BCC03 (Bach et al., 2022b). More recently, two novel Bcc species were described isolated from the semi-arid north-east Brazilian region causing onion sour skin: “Burkholderia sola” and “Burkholderia semiarida” (Velez et al., 2023). The latter groups with the BCC03 strains in ANI and phylogenetic analyses (Figures 3 and ^S5). Here we show that, from the genomic standpoint, both species could be validly accepted (Table 1). Besides that, some other new species are still to be described in the vicinity of B. cenocepacia (^{Table S2}). A better definition and characterization of strains belonging to this group is critical since B. cenocepacia infections of cystic fibrosis patients exhibit poor outcomes (Pope et al., 2010).

Other recommendations

Genome metrics are revolutionizing bacterial taxonomy since the developed tools are easy to use, the analyses are portable, and the definition of clear thresholds makes the results more reliable. However, there are some concerns regarding the quality of genome sequences and the use of validly published type strains in the comparisons. To propose a new species, a representative strain is chosen as the type strain, which serves as a reference in subsequent taxonomical studies (Chun et al., 2018). Much relevant information about microbial species, such as the type strains, can be found on the LPSN website (https://www.bacterio.net/).

Noteworthily, one should always verify the quality of the compared genomes. Likewise, evaluating the alignment fraction or the percentage of aligned nucleotides is an essential quality check in genomic comparison analysis for taxonomic purposes (Li et al., 2015). In this study, all ANIb and ANIm results showed more than 65% of aligned nucleotides, except for “B. dokdonensis”, “B. mallei”, and the anomalous assembly of C. humi KEMC 7302-068^T (Table 1). Noteworthily, genomic databases are improving the curation process and removing anomalous assemblies, but there are still some errors that could mislead taxonomic analyses. The following are some examples within Burkholderia s.l.

“Burkholderia reimsis” BE51 belongs to B. cepacia

Strain BE51 was suggested as a new species in a genome announcement study without providing further evidence (Esmaeel et al., 2018). This genome is currently assigned as a “representative genome” in the RefSeq database. Our phylogenomic studies indicated that this strain was misidentified (Figure 3). Indeed, genome metrics ANI and dDDH showed that “Burkholderia reimsis” BE51 is a B. cepacia strain (Table 1 and ^S3). Therefore, the species “Burkholderia reimsis” should not be validly published.

Figure 3 -
Phylogenetic tree of Burkholderia strains based on the alignment of 695 orthologous marker sequences recognized by GET-HOMOLOGUES and reconstructed through the maximum likelihood approach of GET-PHYLOMARKERS. Type strains are shown in bold. Burkholderia territorii strains were set as the outgroup. All bootstrap values are shown.

Incongruencies between “Burkholderia paludis” MSh1 ^T genome sequences

Although our phylogenomic studies did not indicate problems with the genome sequence of “B. paludis” MSh1^T, Peeters et al. (2020) recently called attention to incongruencies with the sequence type gene markers obtained from the deposited genome of “B. paludis” MSh1^T, the resequenced genome of LMG 30113^T, and the information provided in the original description paper (Ong et al., 2016). Therefore, one should interpret genomic data from this species cautiously, and the name should only be validly published if these incongruencies are solved.

The genome assembly of Caballeronia humi KEMC 7302-068 should be avoided

Caballeronia terrestris LMG 22937^T and Caballeronia humi LMG 22934^T share high ANIb (93.65%) and dDDH values (69.60%), yet below species boundaries (Table 1), confirming that they are separate species. The NCBI RefSeq database correctly indicates the genome assembly GCF_001544475.1 of C. humi LMG 22934^T as the representative genome. However, another deposit of C. humi KEMC 7302-068^T, GCF_007474635.1, might be used preferentially by researchers seeking genomes with higher completeness, which is the case of the latter (50% instead of 25% of NCBI’s classification). Furthermore, here we showed that the assembly GCF_007474635.1 shows discrepant ANI and dDDH when compared to GCF_001544475.1 (Table 1), both putatively sequenced from the type strain. Therefore, the genome assembly GCF_007474635.1 should be avoided in taxonomic investigations. This assembly failed the NCBI database taxonomy check and has been recently removed from RefSeq.

Final remarks

Many studies have revisited the taxonomy of Burkholderiaceae using genome-based tools (Dobritsa and Samadpour, 2016; Estrada-de Los Santos et al., 2018; Wallner et al., 2019; Jin et al., 2020; Mullins and Mahenthiralingam, 2021; Bach et al., 2022b). These works have given valuable contributions to understanding the group since the reliable identification of a strain is an important step in exploring biotechnological potentials, being aware of biosafety risks, and choosing the most appropriate clinical protocols. However, we would like to highlight that other phenotypic investigations, beyond genomic analysis, should be performed to confirm the pathogenicity or host specificity of a strain. Furthermore, synonyms and putative new species within this group have already been recognized (Jin et al., 2020; Mullins and Mahenthiralingam, 2021; Bach et al., 2022b). Here we performed additional analyses to corroborate previous results, evaluated the presence of synonyms, and suggested some recommendations for the taxonomic study of this bacterial group.

ANI results varied among tools due to the implementation of different algorithms or the adoption of slight modifications in the formulae (Table 1). However, these differences should not be a problem since there is a recommendation to evaluate at least two different metrics to delineate species (Wayne et al., 1987). Here we highlighted the importance of using ANI and dDDH as a more discriminatory pipeline for Burkholderia s.l. strains that present borderline values in whole-genome comparisons. Thus, we recommend using a combination of the universal thresholds of 95 and 70% for ANI and dDDH calculations, respectively, to delineate species of this group reliably. In some cases, it was also necessary to evaluate phylogeny and the description of differential phenotypic traits that corroborate genomic differences. A previous comprehensive work has proposed using multiple ANI values to discriminate some Burkholderia s.l. species unequivocally (Mullins and Mahenthiralingam, 2021). This procedure enables the screening of large datasets once the dDDH tool is limited to a few comparisons per time, hindering its extensive adoption in genome-based taxonomic projects. However, following a universal threshold should be preferential to standardize communication among different areas. Considering these pragmatic criteria for the evaluation of genome metrics, here we revised the current Burkholderia s.l. taxonomy and reclassified P. insulsa as a later heterotypic synonym of P. fungorum, corroborated that closely related strains belong to different species, recommended the validation of species names, and showed incongruencies in names and genome assemblies.

Acknowledgments

EB received a scholarship from Conselho Nacional de Desenvolvimento Científico e Tecnológico, Brazil, Grant number 155771/2018-3.

References

• Andrade JP, de Souza HG, Ferreira LC, Cnockaert M, De Canck E, Wieme AD, Peeters C, Gross E, De Souza JT and Marbach PAS (2021) Burkholderia perseverans sp. nov., a bacterium isolated from the Restinga ecosystem, is a producer of volatile and diffusible compounds that inhibit plant pathogens. Braz J Microbiol 52:2145-2152.
• Auch AF, Klenk H-P and Göker M (2010) Standard operating procedure for calculating genome-to-genome distances based on high-scoring segment pairs. Stand Genomic Sci 2:142-142.
• Bach E, Passaglia LMP, Jiao J and Gross H (2022a) Burkholderia in the genomic era: From taxonomy to the discovery of new antimicrobial secondary metabolites. Crit Rev Microbiol 48:121-160.
• Bach E, Sant’Anna FH, dos Santos Seger GD and Passaglia LMP (2022b) Pangenome inventory of Burkholderia sensu lato, Burkholderia sensu stricto, and the Burkholderia cepacia complex reveals the uniqueness of Burkholderia catarinensis Genomics 114:398-408.
• Bellés-Sancho P, Beukes C, James EK and Pessi G (2023) Nitrogen-fixing symbiotic Paraburkholderia species: Current knowledge and future perspectives. Nitrogen 4:135-158.
• Beukes CW, Steenkamp ET, Van Zyl E, Avontuur J, Chan WY, Hassen AI, Palmer M, Mthombeni LS, Phalane FL and Sereme TK (2019) Paraburkholderia strydomiana sp. nov. and Paraburkholderia steynii sp. nov.: Rhizobial symbionts of the fynbos legume Hypocalyptus sophoroides Antonie Van Leeuwenhoek 112:1369-1385.
• Brock DA, Noh S, Hubert AN, Haselkorn TS, DiSalvo S, Suess MK, Bradley AS, Tavakoli-Nezhad M, Geist KS and Queller DC (2020) Endosymbiotic adaptations in three new bacterial species associated with Dictyostelium discoideum: Paraburkholderia agricolaris sp. nov., Paraburkholderia hayleyella sp. nov., and Paraburkholderia bonniea sp. nov. PeerJ 8:e9151.
• Chun J, Oren A, Ventosa A, Christensen H, Arahal DR, da Costa MS, Rooney AP, Yi H, Xu X-W and De Meyer S (2018) Proposed minimal standards for the use of genome data for the taxonomy of prokaryotes. Int J Syst Evol Microbiol 68:461-466.
• Coenye T, Laevens S, Willems A, Ohlén M, Hannant W, Govan JR, Gillis M, Falsen E and Vandamme P (2001) Burkholderia fungorum sp. nov. and Burkholderia caledonica sp. nov., two new species isolated from the environment, animals and human clinical samples. Int J Syst Evol Microbiol 51:1099-1107.
• Contreras-Moreira B and Vinuesa P (2013) GET_HOMOLOGUES, a versatile software package for scalable and robust microbial pangenome analysis. Appl Environ Microbiol 79:7696-7701.
• Dobritsa AP and Samadpour M (2016) Transfer of eleven species of the genus Burkholderia to the genus Paraburkholderia and proposal of Caballeronia gen. nov. to accommodate twelve species of the genera Burkholderia and Paraburkholderia Int J Syst Evol Microbiol 66:2836-2846.
• Esmaeel Q, Issa A, Sanchez L, Clément C, Jacquard C and Barka EA (2018) Draft genome sequence of Burkholderia reimsis BE51, a plant-associated bacterium isolated from agricultural rhizosphere. Microbiol Resour Announc 7:e00978-18.
• Estrada-de Los Santos P, Vinuesa P, Martínez-Aguilar L, Hirsch AM and Caballero-Mellado J (2013) Phylogenetic analysis of Burkholderia species by multilocus sequence analysis. Curr Microbiol 67:51-60.
• Estrada-de Los Santos P, Palmer M, Chávez-Ramírez B, Beukes C, Steenkamp ET, Briscoe L, Khan N, Maluk M, Lafos M and Humm E (2018) Whole genome analyses suggests that Burkholderia sensu lato contains two additional novel genera (Mycetohabitans gen. nov., and Trinickia gen. nov.): Implications for the evolution of diazotrophy and nodulation in the Burkholderiaceae. Genes 9:389.
• Gao Z-Q, Zhao D-Y, Xu L, Zhao R-T, Chen M and Zhang C-Z (2016) Paraburkholderia caffeinitolerans sp. nov., a caffeine degrading species isolated from a tea plantation soil sample. Antonie Van Leeuwenhoek 109:1475-1482.
• Garrity GM, Parker CT and Tindall BJ (2015) International code of nomenclature of prokaryotes. Int J Syst Evol Microbiol 90:1254416.
• Godoy D, Randle G, Simpson AJ, Aanensen DM, Pitt TL, Kinoshita R and Spratt BG (2003) Multilocus sequence typing and evolutionary relationships among the causative agents of melioidosis and glanders, Burkholderia pseudomallei and Burkholderia mallei J Clin Microbiol 41:2068-2079.
• González-Torres P, Rodríguez-Mateos F, Antón J and Gabaldón T (2019) Impact of homologous recombination on the evolution of prokaryotic core genomes. MBio 10:e02494-18.
• Goris J, Konstantinidis KT, Klappenbach JA, Coenye T, Vandamme P and Tiedje JM (2007) DNA-DNA hybridization values and their relationship to whole-genome sequence similarities. Int J Syst Evol Microbiol 57:81-91.
• Gyaneshwar P, Hirsch AM, Moulin L, Chen W-M, Elliott GN, Bontemps C, Estrada-de los Santos P, Gross E, dos Reis Jr FB and Sprent JI (2011) Legume-nodulating betaproteobacteria: Diversity, host range, and future prospects. Mol Plant Microbe Interact 24:1276-1288.
• Hall CM, Baker AL, Sahl JW, Mayo M, Scholz HC, Kaestli M, Schupp J, Martz M, Settles EW and Busch JD (2022) Expanding the Burkholderia pseudomallei Complex with the Addition of Two Novel Species: Burkholderia mayonis sp. nov. and Burkholderia savannae sp. nov. Appl Environ Microbiol 88:e01583-21.
• Jain C, Rodriguez-R LM, Phillippy AM, Konstantinidis KT and Aluru S (2018) High throughput ANI analysis of 90K prokaryotic genomes reveals clear species boundaries. Nat Commun 9:5114.
• Jin Y, Zhou J, Zhou J, Hu M, Zhang Q, Kong N, Ren H, Liang L and Yue J (2020) Genome-based classification of Burkholderia cepacia complex provides new insight into its taxonomic status. Biol Direct 15:6.
• Jung M-Y, Kang M-S, Lee K-E, Lee E-Y and Park S-J (2019) Paraburkholderia dokdonella sp. nov., isolated from a plant from the genus Campanula. J Microbiol 57:107-112.
• Li X, Huang Y and Whitman WB (2015) The relationship of the whole genome sequence identity to DNA hybridization varies between genera of prokaryotes. Antonie Van Leeuwenhoek 107:241-249.
• Lin Q, Lv Y, Gao Z and Qiu L (2020) Pararobbsia silviterrae gen. nov., sp. nov., isolated from forest soil and reclassification of Burkholderia alpina as Pararobbsia alpina comb. nov. Int J Syst Evol Microbiol 70:1412-1420.
• Lopes-Santos L, Castro DBA, Ferreira-Tonin M, Corrêa DBA, Weir BS, Park D, Ottoboni LMM, Rodrigues Neto J and Destéfano SAL (2017) Reassessment of the taxonomic position of Burkholderia andropogonis and description of Robbsia andropogonis gen. nov., comb. nov. Antonie Van Leeuwenhoek 110:727-736.
• Madhaiyan M, Sriram S, Kiruba N and Saravanan VS (2022) Genome-based Reclassification of Paraburkholderia insulsa as a Later Heterotypic Synonym of Paraburkholderia fungorum and Proposal of Paraburkholderia terrae subsp. terrae subsp. nov. and Paraburkholderia terrae subsp. steynii subsp. nov. Curr Microbiol 79:358.
• Mavengere NR, Ellis AG and Le Roux JJ (2014) Burkholderia aspalathi sp. nov., isolated from root nodules of the South African legume Aspalathus abietina Thunb. Int J Syst Evol Microbiol 64:1906-1912.
• Mavima L, Beukes CW, Palmer M, De Meyer SE, James EK, Maluk M, Muasya MA, Avontuur JR, Chan WY and Venter SN (2022) Delineation of Paraburkholderia tuberum sensu stricto and description of Paraburkholderia podalyriae sp. nov. nodulating the South African legume Podalyria calyptrata Syst Appl Microbiol 45:126316.
• Meier-Kolthoff JP, Auch AF, Klenk H-P and Göker M (2013) Genome sequence-based species delimitation with confidence intervals and improved distance functions. BMC Bioinformatics 14:60.
• Meier-Kolthoff JP, Hahnke RL, Petersen J, Scheuner C, Michael V, Fiebig A, Rohde C, Rohde M, Fartmann B and Goodwin LA (2014) Complete genome sequence of DSM 30083 T, the type strain (U5/41 T) of Escherichia coli, and a proposal for delineating subspecies in microbial taxonomy. Stand Genomic Sci 9:2.
• Morales-Ruíz L-M, Rodríguez-Cisneros M, Kerber-Díaz J-C, Rojas-Rojas F-U, Ibarra JA and Estrada-de Los Santos P (2022) Burkholderia orbicola sp. nov., a novel species within the Burkholderia cepacia complex. Arch Microbiol 204:178.
• Mullins AJ and Mahenthiralingam E (2021) The hidden genomic diversity, specialized metabolite capacity, and revised taxonomy of Burkholderia Sensu Lato. Front Microbiol 12:726847.
• Ong KS, Aw YK, Lee LH, Yule CM, Cheow YL and Lee SM (2016) Burkholderia paludis sp. nov., an Antibiotic-Siderophore Producing Novel Burkholderia cepacia Complex Species, Isolated from Malaysian Tropical Peat Swamp Soil. Front Microbiol 7:2046.
• Palleroni NJ, Kunisawa R, Contopoulou R and Doudoroff M (1973) Nucleic acid homologies in the genus Pseudomonas Int J Syst Evol Microbiol 23:333-339.
• Palmer M, Steenkamp ET, Blom J, Hedlund BP and Venter SN (2020) All ANIs are not created equal: Implications for prokaryotic species boundaries and integration of ANIs into polyphasic taxonomy. Int J Syst Evol Microbiol 70:2937-2948.
• Parks DH, Imelfort M, Skennerton CT, Hugenholtz P and Tyson GW (2015) CheckM: Assessing the quality of microbial genomes recovered from isolates, single cells, and metagenomes. Genome Res 25:1043-1055.
• Paulitsch F, Dall’Agnol RF, Delamuta JRM, Ribeiro RA, da Silva Batista JS and Hungria M (2020a) Paraburkholderia atlantica sp. nov. and Paraburkholderia franconis sp. nov., two new nitrogen-fixing nodulating species isolated from Atlantic forest soils in Brazil. Arch Microbiol 202:1369-1380.
• Paulitsch F, Delamuta JRM, Ribeiro RA, da Silva Batista JS and Hungria M (2020b) Phylogeny of symbiotic genes reveals symbiovars within legume-nodulating Paraburkholderia species. Syst Appl Microbiol 43:126151.
• Payne GW, Vandamme P, Morgan SH, LiPuma JJ, Coenye T, Weightman AJ, Jones TH and Mahenthiralingam E (2005) Development of a recA gene-based identification approach for the entire Burkholderia genus. Appl Environ Microbiol 71:3917-3927.
• Peeters C, Depoorter E, Canck ED and Vandamme P (2020) Genome sequence-based curation of PubMLST data challenges interspecies recombination in the Burkholderia cepacia complex. Future Microbiol 15:1091-1093.
• Pope CE, Short P and Carter PE (2010) Species distribution of Burkholderia cepacia complex isolates in cystic fibrosis and non-cystic fibrosis patients in New Zealand. J Cyst Fibros 9:442-446.
• Pratama AA, Jiménez DJ, Chen Q, Bunk B, Spröer C, Overmann J and van Elsas JD (2020) Delineation of a subgroup of the genus Paraburkholderia, including P. terrae DSM 17804T, P. hospita DSM 17164T, and four soil-isolated fungiphiles, reveals remarkable genomic and ecological features-proposal for the definition of a P. hospita species cluster. Genome Biol Evol 12:325-344.
• Richter M and Rosselló-Móra R (2009) Shifting the genomic gold standard for the prokaryotic species definition. Proc Natl Acad Sci 106:19126-19131.
• Rusch A, Islam S, Savalia P and Amend JP (2015) Burkholderia insulsa sp. nov., a facultatively chemolithotrophic bacterium isolated from an arsenic-rich shallow marine hydrothermal system. Int J Syst Evol Microbiol 65:189-194.
• Sant’Anna FH, Bach E, Porto RZ, Guella F, Sant’Anna EH and Passaglia LM (2019) Genomic metrics made easy: What to do and where to go in the new era of bacterial taxonomy. Crit Rev Microbiol 45:182-200.
• Sawana A, Adeolu M and Gupta RS (2014) Molecular signatures and phylogenomic analysis of the genus Burkholderia: Proposal for division of this genus into the emended genus Burkholderia containing pathogenic organisms and a new genus Paraburkholderia gen. nov. harboring environmental species. Front Genet 5:429.
• Seemann T (2014) Prokka: Rapid prokaryotic genome annotation. Bioinformatics 30:2068-2069.
• Sheu S-Y, Chen M-H, Liu WY, Andrews M, James EK, Ardley JK, De Meyer SE, James TK, Howieson JG and Coutinho BG (2015) Burkholderia dipogonis sp. nov., isolated from root nodules of Dipogon lignosus in New Zealand and Western Australia. Int J Syst Evol Microbiol 65:4716-4723.
• Vandamme P, Holmes B, Vancanneyt M, Coenye T, Hoste B, Coopman R, Revets H, Lauwers S, Gillis M and Kersters K (1997) Occurrence of multiple genomovars of Burkholderia cepacia in cystic fibrosis patients and proposal of Burkholderia multivorans sp. nov. Int J Syst Evol Microbiol 47:1188-1200.
• Vandamme P, Holmes B, Coenye T, Goris J, Mahenthiralingam E, LiPuma JJ and Govan JR (2003) Burkholderia cenocepacia sp. nov.- a new twist to an old story. Res Microbiol 154:91-96.
• Vanwijnsberghe S, Peeters C, De Ridder E, Dumolin C, Wieme AD, Boon N and Vandamme P (2021) Genomic aromatic compound degradation potential of novel Paraburkholderia Species: Paraburkholderia domus sp. nov., Paraburkholderia haematera sp. nov. and Paraburkholderia nemoris sp. nov. Int J Mol Sci 22:7003.
• Velez LS, Aburjaile FF, de Farias ARG, Baia ADB, de Oliveira WJ, Silva AMF, Benko-Iseppon AM, Azevedo V, Brenig B and Ham JH (2023) Burkholderia semiarida sp. nov. and Burkholderia sola sp. nov., two novel B. cepacia complex species causing onion sour skin. Syst Appl Microbiol 46:126415.
• Vinuesa P, Ochoa-Sánchez LE and Contreras-Moreira B (2018) GET_PHYLOMARKERS, a software package to select optimal orthologous clusters for phylogenomics and inferring pan-genome phylogenies, used for a critical geno-taxonomic revision of the genus Stenotrophomonas Front Microbiol 9:771.
• Volpiano CG, Sant’Anna FH, Ambrosini A, de São José JFB, Beneduzi A, Whitman WB, de Souza EM, Lisboa BB, Vargas LK and Passaglia LMP (2021) Genomic metrics applied to Rhizobiales (Hyphomicrobiales): Species reclassification, identification of unauthentic genomes and false type strains. Front Microbiol 12:614957.
• Wallner A, King E, Ngonkeu EL, Moulin L and Béna G (2019) Genomic analyses of Burkholderia cenocepacia reveal multiple species with differential host-adaptation to plants and humans. BMC Genomics 20:803.
• Wayne LG, Brenner DJ, Colwell RR, Grimont PAD, Kandler O, Krichevsky MI, Moore LH, Moore WEC, Murray Rge and Stackebrandt E (1987) Report of the ad hoc committee on reconciliation of approaches to bacterial systematics. Int J Syst Evol Microbiol 37:463-464.
• Yabuuchi E, Kosako Y, Oyaizu H, Yano I, Hotta H, Hashimoto Y, Ezaki T and Arakawa M (1992) Proposal of Burkholderia gen. nov. and transfer of seven species of the genus Pseudomonas homology group II to the new genus, with the type species Burkholderia cepacia (Palleroni and Holmes 1981) comb. nov. Microbiol Immunol 36:1251-1275.
• Yang H-C, Im W-T, Kim KK, An D-S and Lee S-T (2006) Burkholderia terrae sp. nov., isolated from a forest soil. Int J Syst Evol Microbiol 56:453-457.

• Author Contributions

  EB Conceptualization, Methodology, Investigation, Formal Analysis, Writing and Original Draft Preparation; CGV Conceptualization, Methodology, Investigation, Formal Analysis; FHS Writing- Review and Editing; LMPP Supervision, Writing - Review and Editing

Supplementary material

The following online material is available for this article:

Figure S1 - Genomic clusters detected using pairwise ANI values from 824 Burkholderiales genomes.

Figure S2 - Heatmap of ANI values obtained from pairwise comparisons of genomes of representative Paraburkholderia spp.

Figure S3 - Heatmap of ANIb values obtained by pyANI comparison of genomes of selected Paraburkholderia strains.

Figure S4 - Phylogenetic tree of Burkholderia strains based on the alignment of 467 orthologous protein sequences recognized by GET-HOMOLOGUES and reconstructed through the maximum likelihood approach of GET-PHYLOMARKERS.

Figure S5 - Heatmap of ANI values obtained from pairwise comparisons of genomes of Burkholderia strains.

Table S1 - Sequence data from the genomes of Burkholderiales type strains analysed in this work using ProKlust.

Table S2 - List of Burkholderia s.l. genomes used in this work together with quality features and reclassifications.

Table S3 - Clusters of identity/similarity formed by the Burkholderia sensu lato genomes analysed in this work using FastANI coupled with ProKlust.

Edited by

Associate Editor:

Loreta Brandão de Freitas

Publication Dates

• Publication in this collection
  03 Nov 2023
• Date of issue
  2023

History

• Received
  25 Apr 2023
• Accepted
  13 Sept 2023