A semi-selective differential medium for identifying strains of Pantoea from environmental samples

Abstract

The genus Pantoea of the Erwinaceae is a genetically and metabolically diverse group whose representatives span a broad range of clinical and nonclinical environments. We sought to develop a culturing medium to facilitate the identification of Pantoea from environmental samples, and to distinguish its members from other closely related Gram-negative genera. We developed a semiselective differential medium, Pantoea Differential Medium (PDM), which contains crystal violet, sodium citrate, tryptone, as well as protease peptone. The efficacy of the medium was evaluated by assessing the growth and phenotype of 42 bacterial strains, including 18 strains of Pantoea along with other representatives of the Enterobacterales. All Pantoea strains, as well as Kosakonia sp. and Duffyella gerundensis formed orange-yellow pigmented colonies on the medium, while all other representative members of the Enterobacterales evaluated formed purple to pink colonies. The medium was also used to distinguish Pantoea from other bacteria in mixed cultures from environmental samples, with 44 of the 48 orange-yellow colonies being identified as members of Pantoea. PDM provides a means for rapidly identifying members of Pantoea from environmental samples, and differentiating them from many of the closely related members of the Erwiniaceae.

In this study, a new semiselective differential medium was developed that will allow for more comprehensive studies aimed at exploring the environmental distribution and general biology of the metabolically versatile group, Pantoea.

Introduction

Members of the bacterial genus Pantoea have been isolated from both environmental and clinical settings, and have been harnessed for agricultural and commercial applications (Walterson and Stavrinides 2015, Williams and Stavrinides 2020, 2022, Williams et al. 2020). Environmental surveys to identify and isolate members of Pantoea have provided additional insight into the diversity of the species, and have resulted in the identification of new isolates with distinct metabolic capabilities.

The identification of Pantoea isolates from cultured environmental samples can be challenging, which has prompted the development of various selective and semiselective media to facilitate quick identification and isolation. Some Pantoea-specific culture media has been formulated based on the relative hallophilicity of Pantoea isolates, such as the semiselective medium, PA 20, which contains NaCl (20% w/v), thallium nitrate, andD-arabitol (Goszczynska et al. 2006, Brady et al. 2009, Silini-Chérif et al. 2012, Kini et al. 2019). PA 20 supports the growth of Pantoea ananatis and Pantoea stewartii strains, but only some Pantoea agglomerans strains (Norris et al. 1976, Goszczynska et al. 2006). Species such as Pantoea vagans and Pantoea eucalypti do not utilize D-arabitol, while other Pantoea species have been suggested to be inhibited by thallium nitrate (Norris et al. 1976, Goszczynska et al. 2006, Brady et al. 2009, Silini-Chérif et al. 2012, Kini et al. 2019).

The semiselective medium, PGSA, whose key constitutents are NaCl (65% w/v), crystal violet, and sodium thiosulfate was developed to identify and isolate P. ananatis, P. agglomerans, and P. stewartii from infected rice crops and seeds (Kini et al. 2019). The medium also appears to support the growth of Pantoea cedenensis, as well as a variety of unclassified Pantoea spp., while inhibiting the growth of species such as Sphingomonas spp. (Kini et al. 2019). Likewise, the semiselective medium, LOMAC, developed to selectively isolate a broader range of Pantoea spp., is primarily composed of lysine, ornithine, mannitol, arginine, as well as charcoal, which absorbs chemicals that are potentially toxic to Pantoea (Kuranishi et al. 2019). However, it also supports the growth of a variety of Gram-negative environmental isolates, including other members of the Erwinaceae that also exhibit the characteristic yellow pigmentation (Kuranishi et al. 2019).

In this study, we developed a new differential medium called “Pantoea Differential Medium (PDM)” that inhibits the growth of many Gram-positive and fungal species, but allows for a diversity of Pantoea species to be distinguished from many closely related species on an agar plate (Chen and Day 1974, Stewart et al. 1977, Hall and Hamilton 1982, Mani and Bharagava 2016, Kuranishi et al. 2019). Importantly, it allows for the differentiation of even nonpigmented Pantoea isolates, such as those of Pantoea septica, which retain a more yellow pigmentation on PDM relative to other members of the Erwiniaceae.

Materials and methods

Bacterial strains and culturing conditions

The bacterial strains used in the development of PDM are shown in Table 1. Cultures were streaked onto 1X Lysogeny Broth agar plates (LB; Miller, BD Biosciences Franklin Lakes, NJ) from 20% glycerol stocks maintained at −80°C (50% glycerol, 50% 1X liquid LB bacteria culture). Incubation conditions for the environmental and clinical strains were 30°C and 37°C, respectively.

Table 1.

Bacterial strains used in this study.

Strain		Isolation	Source
Acinetobacter baumannii	ATCC17A18
Bacillus sp.	E2012-02	Wild rose leaves
Bacillus subtilis	14VB542579B	Catheter urine
Citrobacter	12GC134883	Human, edema, CHF, cellulitis
Duffyella gerundensis	EM595	Peri-orbital swelling	St. Boniface General Hospital
Duffyella sp.	E2023-17	Wild carrot	This study
Enterobacter sp.	ATCC700323		Regina General Hospital
Enterococcus faecium	K0190882		Dr. Andrew Cameron, University of Regina
Erwinia amylovora	EA321	Hawthorn
Erwinia billingiae	EhWF18		Dr. Gwyn Beattie, Iowa State
Escherichia coli	ATCC35218		Regina General Hospital
Klebsiella pneumoniae	ATCCBAA1705		Regina General Hospital
Kosakonia sp.	12 202	Musk melon	ICMP
Lactococcus lactis	HD1		Heather Dietz, University of Regina
Micrococcus sp.	13DB759184A
Mixta calida	18DB967130	Urine, midstream	St. Boniface General Hospital
	19KB424268	Catheter urine	St. Boniface General Hospital
	18BF523624	Catheter urine	St. Boniface General Hospital
	18BF710974	Urostomy urine	St. Boniface General Hospital
	BB957621B1	Human, male, CAPD dialysate, peritonitis	St. Boniface General Hospital
	BB957621A2	Human, male, CAPD dialysate, peritonitis	St. Boniface General Hospital
Pantoea agglomerans	E2023-1	Medicago lupulina (black medic)	This study
	E2023-2	Common dandelion	This study
	E2023-3	Common dandelion	This study
	E2023-5	Stick	This study
	E2023-6	Flea beetle	This study
	E2023-7	Trembling aspen	This study
	E2023-8	Arugula	This study
	E2023-9	Arugula	This study
	E2023-10	Arugula	This study
	E2023-11	Arugula	This study
	E2023-12	Flea beetle	This study
	E2023-13	Flea beetle	This study
	E2023-14	Flea beetle	This study
	E2023-15	Wild carrot	This study
	E2023-16	Wild carrot	This study
	E2023-20	Broomweed	This study
	E2023-22	Soil	This study
	E2023-23	Black spruce pine cone	This study
	E2023-24	Leaf	This study
	E2023-25	Canada thistle	This study
	E2023-27	Lamb’s quarter	This study
	E2023-28	Elm tree sprout	This study
	E2023-29	Hawthorn	This study
	E2023-30	Canada thistle	This study
	E2023-32	Aloe vera	This study
	E2023-34	Sedum	This study
	E2023-35	Baby’s breath	This study
	E2023-36	Crab apple	This study
	E2023-37	Pilea peperomiodes	This study
	E2023-38	Canada thistle	This study
	E2023-41	Strawberry plant	This study
	E2023-42	Arugula	This study
	E2023-43	Arugula	This study
	E2023-44	Arugula	This study
	E2023-45	Arugula	This study
	E2023-46	Arugula	This study
	E2023-47	Arugula	This study
	E2023-48	Arugula	This study
	SN01080	Slug	Nadarasah and Stavrinides (2014)
	H42501	Human, male, blood	Sunnybrook Hospital
	SP00101	Raspberry	Nadarasah and Stavrinides (2014)
	12 531	Gypsophila	ICMP
	B015092	Urine midstream	Sask. Disease Control Lab
	DC434	Maize pathogen	Dr. David Coplin, Ohio State
	SP01202	Strawberry leaf and stem foxtail barley	Nadarasah and Stavrinides (2014)
	E2012-29
Pantoea allii	E2023-31	Dead nettle	This study
Pantoea ananatis	M232A	Maize	Dr. Steven Lindow, UC Berkley
	E2023-4	Cattail	This study
Pantoea dispersa	E2012-01	Wild rose leaves
Pantoea eucalypti	B011489	Superficial wound	Sask. Disease Control Lab
Pantoea eucrina	E2023-33	Crab apple blossom	This study
	E2023-39	Dead neetle	This study
	E2023-40	Garden poppy	This study
Pantoea septica	18GB96456	Catheter urine	St. Boniface General Hospital
	18GB350152B	Endotracheal	St. Boniface General Hospital
	18GB929244	Wound, foot	St. Boniface General Hospital
	13DB759184C	Nephrostomy urine	St. Boniface General Hospital
	17VB530109	Catheter urine	St. Boniface General Hospital
	13DB759184B	Nephrostomy urine	St. Boniface General Hospital
Pantoea spp.	E2023-21	Broomweed	This study
Pseudomonas aureofaciens	Pau1
Pseudomonas sp.	E2023-26	Canada thistle	This study
Staphylococcus aureus	K1-7	Clinical	Dr. Chris Yost, University of Regina
Salmonella enterica	ATCCH028
Salmonella typhimurium	TA98
Streptococcus mutans	UAIS9:wt	Clinical	Heather Dietz, University of Regina

Pantoea Differential Medium

One liter of PDM contains 16 g Bacto^TM Tryptone Pancreatic Digest of Casein (Life Technologies Corporation, Detroit, MI), 8 g Criterion Proteose Peptone #3 (Hardy Diagnostics, Santa Maria, CA), 10 g NaCl (Fischer Scientific, Waltham, MA), 32 g K₂HPO₄ (EMD Chemicals, Germany), 16 g KH₂PO₄ (VWR Life Science, Solon, OH), 4.04 g sodium citrate (J.T. Baker Chemical Co., Phillisburg, NJ), and 8 ml Crystal Violet (Thermo Electron Corporation, Waltham, MA). The ddH₂O was brought to a boil at 100°C, at which time the heat was reduced, a stir bar added, and each component added sequentially, after which 15 g of agar was added (BD Biosciences Franklin Lakes, NJ). The medium was autoclaved at 121°C for 30 min. After autoclaving, the medium had a pH of 6.6.

Molecular typing of environmental isolates

Environmental samples were obtained from a variety of sources including plants, water, soil, and insects in Regina, SK, Canada, and streaked directly onto PDM. The plates were incubated at 30°C for 24 h, and the development of yellow pigmentation was monitored over a subsequent 24 h period at room temperature. The growth and pigmentation of the reference strains were evaluated on the differential medium at least 3 times, with similar results across experiments.

Bacterial DNA was extracted from strains that had yellow pigmentation on PDM and LB, and the cpn60 gene, which allows for species-level identification, was amplified using colony PCR (Gob et al. 1996, Marston et al. 1999, Brousseau et al. 2001). PCRs were carried out in 25 µl reactions with Standard Taq DNA polymerase according to the manufacturer’s protocol (New England Biolabs Inc., Ipswich, MA) using cpn60_ent + 1 (ATGGCAGCWAAAGACGTAAAATTCGG) and cpn60_ent-1330 (CGCRACYTTRATACCSACGTTCTG) using an annealing temperature of 59°C. Gel electrophoresis using 1% agarose was used to visualize the amplicons. Samples were sequenced by Genome Quebec (Montreal, Quebec). For some strains, the 16S rRNA gene was amplified in a colony PCR reaction (as above) using primers 16S_+27F (AGAGTTTGATCMTGGCTCAG) (Lane et al. 1985, Frank et al. 2008) and 16S_-1400R (ACGGGCGGTGTGTACAA) (Soutar and Stavrinides 2019) at an annealing temperature of 57.8°C. Sequence data have been deposited in Genbank under accession numbers PV699648–PV699692, and PV688339. Sequence data are also available in the supplementary material in FASTA format.

Phylogenetic analysis

Consensus alignments for the forward and reverse sequencing reads of each environmental sample were generated using CLC Workbench (Version 22.0). A ClustalW alignment, as implemented by MEGAX (10.2.4) was then used to generate a Maximum Likelihood phylogeny using the General Time Reversible model with gamma distribution and invariable sites (8 Gamma categories). The initial ML tree was made using the maximum parsimony method with the Subtree-Pruning-Grafting (Extensive, SPR Level 5) heuristic search method, and no branch swap filter.

Results

Development of PDM

We developed a medium that differentiated a panel of Pantoea isolates from other members of the Erwinaceae, and allowed for the identification of some frequently nonpigmented species, such as strains of P. septica. The nutritional components for this differential medium, PDM, were based on Terrific Broth (Gibco^TM, Billings, MT), but sodium citrate was added as a carbon source (Brady et al. 2010). Glycerol and yeast extract were removed to limit the production of metabolic products (Costa et al. 2002). Proteose peptone was supplemented as a nitrogen source, and crystal violet was used to inhibit some Gram-positive bacteria and fungi, and to bind the salts in the medium (Chen and Day 1974, Stewart et al. 1977, Hall and Hamilton 1982, Goszczynska et al. 2006, Mani and Bharagava 2016, Kuranishi et al. 2019).

The efficacy of PDM was assessed using 42 bacterial strains, including representatives of P. agglomerans, Pantoea allii, P. eucalypti, Pantoea dispersa, P. ananatis, P. stewartii, and P. septica. Members of the Erwiniaceae used to evaluate the differentiating ability of the medium included Duffyella, Erwinia, and Mixta, as well as various Gram-positive strains (Table 1). All Pantoea strains from across the different species groups formed yellow or orange colonies on the differential medium after incubation for 48–72 h, including strains of P. septica (Fig. 1). Kosakonia sp. and Duffyella gerundensis also formed yellow or orange colonies, while Mixta calida, which is typically yellow on LB, forms purple colonies on the differential medium (Palmer et al. 2018).

Figure 1.

Pigmentation of reference strains on PDM (A) and Lysogeny Broth (B). From left to right: P. ananatis (M232A), P. agglomerans (E2012-29), P. dispersa (E2012-01), P. eucalypti (B011489), P. agglomerans (SN01080), P. agglomerans (H42501), P. agglomerans (SP00101), P. agglomerans (12 531), P. agglomerans (B015092), P. agglomerans (DC434), P. agglomerans (SP01202), Citrobacter (12GC134883), P. septica (18GB96456), P. septica (18GB350152B), P. septica (18GB929244), P. septica (13DB79184C), P. septica (17VB530109), P. septica (13DB79184B), P. septica (16G18504195), D. gerundensis (EM595), E. amylovora (Ea321), E. billingiae (EhWF18), M. calida (18MB36935), M. calida (18LA85142A), M. calida (18VB347619), M. calida (18VB347619), M. calida (BB957621B1), M. calida (BB957621A2), A. baumannii (ATCC17A18), E. coli (ATCC35218), Enterobacter sp. (ATCC700323), K. pneumoniae (ATCCBAA1705), Kosakonia sp. (12 202), Pseudomonas aureofaciens (Pau1), S. enterica (ATCCH028), S. typhimurium (TA98), Micrococcus sp. (K0190882), Bacillus sp. (E2012-02), E. faecium (K0190882), L. lactis (HD1), S. aureus (K 1–7), S. mutans (UAIS9:wt). Pigmentation of environmental isolates E2023-01 through E2023-48 from Regina, SK, Canada on PDM (C) and Lysogeny Broth (D). Contrast was adjusted evenly on all images to better highlight the differences in colours. The color spectrum (E) of colonies grown on PDM are based on observations from the members of the Enterobacteriaceae and Erwiniaceae used in this study.

Among the Enterobacteriaceae, Citrobacter sp., and Salmonella Typhimurium formed purple and gray colonies on PDM, respectively, while all other members of the Enterobacteriaceae developed purple, pink, or gray colonies (Figure 1). Pseudomonas aureofaciens, which produces an orange-yellow carotenoid on LB medium (Peix et al. 2007, Huang et al. 2011) formed gray-purple colonies on PDM. The Gram-positive strains that were evaluated exhibited no growth (Fig. 1).

Isolation and identification of candidate Pantoea strains from the environment

Given that PDM provided the ability to differentiate between Pantoea and its close relatives, we evaluated its efficacy using specimens collected from plants, trees, insect, and soil. Colonies that had yellow or orange pigmentation were typed using cpn60 and a maximum-likelihood phylogeny was generated (Fig. 2). Of the 48 orange to yellow colonies that formed on PDM, 44 were identified as strains of P. agglomerans, P. allii, P. ananatis, P. eucrina, as well as a representative of a candidate new species, Pantoea sp. E2023-21. The isolated P. agglomerans strains formed a single monophyletic group with the reference strain, while isolates of P. allii, P. ananatis, and P. eucrina clustered with their respective reference strains within distinct clades.

Figure 2.

A maximum likelihood phylogeny based on the cpn60 gene from isolated environmental strains generated using the General Time Reversible model with gamma distribution and invariable sites, with 500 bootstrap replicates. Only bootstrap values >70% are shown.

The P. allii was identified from a leaf sample of Canada Thistle (Cirsium arvense), P. ananatis from Bulrush (Typha latifolia), and P. dispersa isolates from a blossom tree, a poppy leaf, and a dead nettle. One strain we isolated from Broomweed (Gutierrezia sarothrae), Pantoea sp. E2023-21, may constitute a new species as it formed a more divergent sister lineage to the P. rodasi type strain (Fig. 2). One strain obtained from Queen Anne’s Lace (Daucua carota) that appeared orangy-yellow when plated on PDM was identified as a species of Duffyella (Soutar and Stavrinides 2022). The newly isolated Duffyella strain grouped with, but was not identical to, the D. gerundensis reference strain in the phylogeny. One strain that could not be identified using cpn60 was identified as Pseudomonas sp. through 16S rRNA typing.

Discussion

In this study, we developed a new differential medium, PDM, for rapidly distinguishing Pantoea isolates from some of their closest relatives. A panel of bacterial strains was assessed on the differential medium, and we found that only strains of Pantoea, Kosakonia, and Duffyella formed yellow to orange colonies. Notably, PDM allows for the differentiation of P. septica, a species primarily associated with the clinical environment, and whose members often produce white or pale yellow colonies on LB medium (Pizzolante et al. 2018). PDM was also effective in differentiating between species of Pantoea and Mixta, which share similar pigmentation on LB medium and were once considered to be part of the same genus (Palmer et al. 2018). PDM appeared to inhibit the growth of all Gram-positive species we evaluated, and may also inhibit fungal species given the antifungal properties of crystal violet (Chen and Day 1974, Stewart et al. 1977, Hall and Hamilton 1982, Mani and Bharagava 2016). Previous studies also used crystal violet to inhibit Gram-positive species, but it was speculated that other components of their medium, such as added thallium nitrate or perhaps the pH also contributed to the inhibition of Gram-positive species (Goszczynska et al. 2006, Kini et al. 2019).

We used our differential medium to isolate and identify strains of Pantoea from environmental samples. Molecular typing of 48 candidate Pantoea strains yielded 44 strains belonging to P. agglomerans, P. allii, P. ananatis, and P. eucrina, with strains of P. agglomerans representing the majority. Additionally, we isolated a strain of P. ananatis, a species that is a plant and opportunistic human pathogen (De Baere et al. 2004, Coutinho and Venter 2009). Notably, some strains of P. ananatis, such as BRT175, have been explored for their antimicrobial metabolites against various agricultural pathogens, such as Erwinia amylovora, and their novel biosurfactants (Walterson et al. 2014, Smith et al. 2016). Two species that are less frequently isolated from environmental surveys, P. eucrina, a species that includes both plant and human pathogens, and P. allii, a species associated with disease of onion, were also recovered (Brady et al. 2010, 2011, Vahling-Armstrong et al. 2016, Lotte et al. 2018, Kølle et al. 2025).

Members of the Enterobacteriaceae including Citrobacter, Enterobacter, Escherichia, and Klebsiella were easily distinguishable from Pantoea on PDM. The only non-Pantoea strain isolated from our environmental samples that formed orangy-yellow colonies on PDM belonged to Duffyella sp., the type strain of which was previously Erwinia gerundensis (Soutar and Stavrinides 2022). Although Pantoea and Duffyella both belong to the Erwinaceae, Pantoea was phenotypically distinct from the other strains of the Erwiniaceae that were assayed on PDM, including closely related Erwinia and Mixta strains. As there are relatively few Duffyella strains available in the public databases, our medium may allow for the isolation of more Duffyella strains from the environment. Additionally, P. aureofaciens Pau1, appeared purple when plated on PDM. However, our study identified an environmental Pseudomonas strain that appeared yellow on PDM, suggesting some Pseudomonads would be false positives. A strain of Kosakonia of the Enterobacteriaceae also produced a yellow pigment when assayed on PDM, and it is possible that members of Kosakonia may also be misidentified as Pantoea. It is therefore important that gene-based typing methods be used to validate any candidate Pantoea isolates cultured on PDM.

PDM is an efficacious medium for the isolation of Pantoea strains from environmental samples, although future refinement of its consitutents may provide a means to better distinguish between Pantoea, Duffyella and some Pseudomonas. The relatively high selectivity of this medium along with the distinctive phenotype of nonpigmented Pantoea strains on PDM will allow for more comprehensive studies aimed at exploring the diversity and biology of this ubiquitous group.

Author contributions

Abigail Apperley: Data curation, Conceptualization, Investigation, Validation. Adria Bateman: Conceptualization, Methodology, Formal analysis, Investigation, Writing—original draft, Writing—review & editing. John Stavrinides: Validation, Formal analysis, Writing—original draft, Writing—review & editing, Supervision, Project administration, Funding acquisition.

Conflict of interest

No conflict of interest declared.

Funding

The authors gratefully acknowledge the financial support from the Natural Sciences and Engineering Research Council of Canada (RGPIN-02863–2021). AA and AB were supported by the Natural Sciences and Engineering Research Council of Canada’s Undergraduate Student Research Award (USRA).

Data availability

Sequence data have been deposited in Genbank under accession numbers PV699648-PV699692, and PV688339. Sequence data are also available in the supplementary material in FASTA format.