DNA Markers for Detection and Genotyping of Xanthomonas euroxanthea

Abstract

Xanthomonas euroxanthea is a bacterial species encompassing both pathogenic and non-pathogenic strains and is frequently found colonizing the same host plants as X. arboricola. This presents the need to develop a detection and genotyping assay able to track these bacteria in microbial consortia with other xanthomonads. Eight X. euroxanthea-specific DNA markers (XEA1-XEA8) were selected by comparative genomics and validated in silico regarding their specificity and consistency using BLASTn, synteny analysis, CG content, codon usage (CAI/eCAI values) and genomic proximity to plasticity determinants. In silico, the selected eight DNA markers were found to be specific and conserved across the genomes of 11 X. euroxanthea strains, and in particular, five DNA markers (XEA4, XEA5, XEA6, XEA7 and XEA8) were unfailingly found in these genomes. A multiplex of PCR targeting markers XEA1 (819 bp), XEA8 (648 bp) and XEA5 (295 bp) was shown to successfully detect X. euroxanthea down to 1 ng of DNA (per PCR reaction). The topology of trees generated with the concatenated sequences of three markers (XEA5, XEA6 and XEA8) and four housekeeping genes (gyrB, rpoD, fyuA and acnB) underlined the equal discriminatory power of these features and thus the suitability of the DNA markers to discriminate X. euroxanthea lineages. Overall, this study displays a DNA-marker-based method for the detection and genotyping of X. euroxanthea strains, contributing to monitoring for its presence in X. arboricola-colonizing habitats. The present study proposes a workflow for the selection of species-specific detection markers. Prospectively, this assay could contribute to unveil alternative host species of Xanthomonas euroxanthea; and improve the control of phytopathogenic strains.

1. Introduction

Following extensive genotyping and comparative genomics studies performed on walnut-associated bacterial isolates, it has been shown that not all the isolates could be identified as Xanthomonas arboricola pv. juglandis, the phytopathogen commonly acknowledged as causing walnut bacterial blight (WBB) [1,2,3]. Further studies showed that some walnut-associated Xanthomonas isolates were taxonomically distinct from any of the other described Xanthomonas species.

These were proposed as members of the new species Xanthomonas euroxanthea [4]. Pathogenicity assays indicated that X. euroxanthea encompasses non-pathogenic and pathogenic strains that can cause WBB-like symptoms, being in a privileged position to investigate genetic determinants of pathogenesis and its evolution [4,5,6].

Interestingly, apart from its occurrence in walnuts (Juglans regia), recent evidence was gathered reporting the isolation of X. euroxanthea from distinct plant host species, such as Carya illinoensis (pecan; strains CPBF 761 and CPBF 766), that together with walnut (Juglans regia; strains CPBF 367, CPBF 424^T, CPBF 426 and CFBP 7653) belong to the Juglandaceae family; Solanum lycopersicum (tomato plants; strains BRIP 62409, BRIP 62411, BRIP 62415 and BRIP 62418) a member of the Solanaceae family [7,8,9]; and Phaseolus vulgaris (common bean; strain CFBP 7622, previously misclassified as X. arboricola [10,11]), a Fabaceae plant species.

More strikingly is that X. arboricola strains were also isolated from all the mentioned plant species, suggesting that both X. euroxanthea and X. arboricola share the same host plants, including the same plant specimen, which raises questions regarding co-colonization and niche-specific adaptations [7,8,12]. Still, taking into consideration the recent taxonomic refinement of X. arboricola species and the unearthing of six new X. euroxanthea strains (2949, 2955, 2957, 2974, 3640 and F2) [11], it is foreseeable the identification of additional X. euroxanthea isolates from plant species that have passed unnoticed so far.

Altogether, the common ecological niche of these two closely related species, plus their cosmopolitan distribution and co-occurrence in different host plant species resulted in misclassifying some X. euroxanthea isolates as X. arboricola [8,10,11], which calls for the need to develop methods for the detection and genotyping of this bacterial species. Over the years, different diagnostic and molecular typing tools have been developed for a variety of xanthomonads [13], namely for X. arboricola including their most studied pathovars juglandis [14,15] and pruni [16,17,18,19], aiming to address their diversity within a geographic or epidemiological context.

Generally, these approaches consist of culture-based detection of the phytopathogen by PCR, followed by multilocus sequence analysis (MLSA) of several housekeeping genes to define haplotypes. In fact, multiplex-PCR and dot-blot hybridization assays showed that three X. arboricola pv. juglandis specific DNA markers (XAJ1, XAJ6 and XAJ8) were absent from X. euroxanthea strains CPBF 367, CPBF 424^T and CPBF 426, which formed a distinct MLSA cluster [12,14].

A robust method to detect X. euroxanthea while undoubtedly distinguishing it from X. arboricola has not been described so far. Furthermore, while several genotyping techniques have been used to assess the diversity of phytosanitary regulated Xanthomonas, as recently reviewed [13], the data currently available regarding genotyping of X. euroxanthea is scarce and limited to MLSA studies performed on Xanthomonas isolated from walnut trees [12].

In this study, comparative genomics and in silico validation tools were combined as previously described [20] to select eight X. euroxanthea-specific DNA markers located in conserved genomic regions. While XEA1, XEA5 and XEA8 DNA markers were chosen to optimize a multiplex PCR detection method for the reliable detection of the target bacteria, altogether the number of SNPs recorded for DNA markers XEA5, XEA6 and XEA8 within the studied X. euroxanthea strains, revealed an allelic variation capable to discriminate X. euroxanthea strains as efficiently as the housekeeping genes used in MLSA. Ultimately, this work may unlock the possibility to conciliate bacterial detection and diversity assessment using the same DNA markers, which may be particularly useful to survey X. euroxanthea populations in environmental samples.

2. Materials and Methods

2.1. In Silico Selection and Validation of X. euroxanthea-Specific DNA Markers

A synteny analysis of 11 X. euroxanthea strains and other 24 representative xanthomonads, downloaded from the NCBI database (Table 1), was performed using MaGe v3.15.3 [21]. This led to the identification of X. euroxanthea-specific coding DNA sequences (CDSs) that are concomitantly present in X. euroxanthea genomes and absent from non-X. euroxanthea strains.

Table 1.

Bacterial strains used for MaGe synteny analysis to retrieve Xanthomonas euroxanthea specific coding sequences (CDSs).

Xanthomonas Species and Pathovars	Strains	GenBank, NCBI Accession/WGS Prefix
X. euroxanthea	CPBF 367	LR861803.1
X. euroxanthea	CPBF 424T	LR994544.1
X. euroxanthea	CPBF 426	LR861805.1
X. euroxanthea	CPBF 761	HG999363.1
X. euroxanthea	CPBF 766	HG999364.1
X. euroxanthea	CFBP 7622	MIGF01.1
X. euroxanthea	CFBP 7653	MIGK01.1
X. euroxanthea	BRIP 62409	QEZJ01.1
X. euroxanthea	BRIP 624011	QEZI01.1
X. euroxanthea	BRIP 62415	QEZH01.1
X. euroxanthea	BRIP 62418	QEZG01.1
X. arboricola	CPBF 1494	HG999362.1
X. arboricola	CPBF 765	HG999365.1
X. arboricola pv. juglandis	CPBF 427	LR861807.1
X. campestris pv. campestris	LMG 568PT	NC_003902
X. campestris pv. campestris	8004	NC_007086
X. campestris pv. campestris	B100	NC_010688
X. citri pv. citri	306	NC_003919
X. citri pv. bilvae	NCPPB 3213PT	CDHI01
X. phaseoli pv. phaseoli	CFBP 412	NZ_CP020964.2
X. citri subsp. aurantifolli	ICPB 10535	ACPY01
X. vasicola pv. musacearum	BCC282	RRCQ01
X. vasicola pv. musacearum	NCPPB 4381	ACHT01
X. oryzae pv. oryzae	PX099A	NC_010717.1
X. oryzae pv. oryzae	MAFF 311018	NC_007705.1
X. oryzae pv. oryzae	KACC 10331	NC_006834.1
X. oryzae pv. oryzicola	BLS256	NZ_AAQN
X. translucens pv. translucens	569	VIWM01.1
X. translucens pv. translucens	LMG 876PT	NZ_CAPJ.1
X. sacchari	NCPPB 4393	AGDB01.1
X. hortorum pv. gardneri	ICMP 7383	CP018731.1
X. hortorum pv. gardneri	LMG 962T	NZ_AEQX.1
X. vesicatoria	LMG 911T	NZ_AEQV
X. euvesicatoria pv. perforans	91-118	NZ_AEQW
X. albilineans	GPE PC73	NC_013722

Then, a BLAST was performed in NCBI (accessed in 22 November 2021 at https://www.ncbi.nlm.nih.gov/) using as query the putative X. euroxanthea-specific CDSs and the database nr/nt to further confirm their specificity. Sequences with hits pertaining solely to X. euroxanthea were considered putative X. euroxanthea-specific genomic regions (Table S1) and were used for primer and specific DNA marker design using Geneious^® 9.1.8 [22]. The affinity and complementarity of these primers to target X. euroxanthea CDSs was checked using NCBI’s Primer-BLAST tool [22,23]. The obtained eight DNA markers were then evaluated for specificity by a BLASTn analysis in Geneious^® 9.1.8 and NCBI (using the database nr/nt).

The genomic context of the markers was assessed to ensure its consistency across the diversity of X. euroxanthea strains. Particularly, a comparative genomics analysis was performed by local alignment of CDSs in MaGe (Figure S1a–d), to evaluate the syntenic context. Additionally, features, such as the CAI/eCAI values [24]; GC content (deducted from MaGe); chromosomal location; and proximity to determinants of genomic plasticity, namely transposons, integrases, recombinases and phage-related ORFs (Geneious^® 9.1.8), were annotated. The number of Single Nucleotide Polymorphisms (SNPs) included in the eight DNA markers and housekeeping genes of the 11 X. euroxanthea genomes considered in this study were summed up recurring to Geneious^® 9.1.8 and normalized using the formula $\left(\frac{\mathrm{SNPs} \mathrm{number}}{\mathrm{total} \mathrm{nucleotide} \mathrm{sequence} \mathrm{length}}\times 100\right)$. The described procedure followed for specific X. euroxanthea DNA markers design is systematized in Figure 1.

Figure 1.

Flowchart for the selection and validation of X. euroxanthea-specific DNA markers.

2.2. Bacterial Strains, Culture Conditions and DNA Extraction

The bacterial strains used for the validation of the eight X. euroxanthea-specific markers are listed in Table 2 and include seven X. euroxanthea strains; and other closely related and niche-sharing strains, namely 10 strains of X. arboricola, seven strains representing six pathovars of X. arboricola and 11 strains belonging to non-arboricola Xanthomonas species. Bacterial strains were cultured as previously described [14] or in peptone-sucrose-agar (PSA) medium (10 g peptone; 10 g sucrose; 1 g glutamic acid; 15 g agar and distilled water up to 1.0 L) at 28 °C. DNA was extracted from pure cultures using the EZNA Bacterial DNA Purification kit (Omega Bio-Tek, Norcross, GA, USA), according to the manufacturer’s instructions and quantified using a DS-11 microvolume spectrophotometer (DeNovix, Wilmington, DE, USA).

Table 2.

List of bacterial strains used for validation of the Xanthomonas euroxanthea-specific DNA markers.

Xanthomonas Species and Pathovars	Strains 1	Geographic Origin	Year of Isolation
X. euroxanthea	CPBF 367	Portugal (Loures)	2016
X. euroxanthea	CPBF 424T	Portugal (Loures)	2016
X. euroxanthea	CPBF 426	Portugal (Loures)	2016
X. euroxanthea	CPBF 761	Portugal (Alcobaça)	2016
X. euroxanthea	CPBF 766	Portugal (Alcobaça)	2016
X. euroxanthea	CFBP 7622	USA	1985
X. euroxanthea	CFBP 7653	France	2008
X. arboricola	CPBF 122	Portugal (Ponte da Barca)	2015
X. arboricola	CPBF 237	Portugal (Ponte de Lima)	2015
X. arboricola	CPBF 554	Portugal (Carrazeda de Ansiães)	2016
X. arboricola	CPBF 765	Portugal (Alcobaça)	2016
X. arboricola	CPBF 796	Portugal (Alcobaça)	2016
X. arboricola	CPBF 1494	Portugal (Alcobaça)	2014
X. arboricola	CPBF 1483	Portugal (Alcobaça)	2014
X. arboricola	CPBF 1514	Portugal (Estremoz)	2014
X. arboricola	CPBF 1567	Portugal (Bombarral)	2015
X. arboricola	CPBF 1586	Portugal (Loures)	2015
X. arboricola pv. juglandis	CPBF 427	Portugal (Loures)	2016
X. arboricola pv. juglandis	CPBF 1521	Portugal (Loures)	2014
X. arboricola pv. celebensis	LMG 677PT	New Zealand	1960
X. arboricola pv. corylina	LMG 689PT	USA	1939
X. arboricola pv. fragariae	LMG 19145PT	Italy	1993
X. arboricola pv. populi	CFBP 3123PT	Netherlands	1979
X. arboricola pv. pruni	LMG 852PT	New Zealand	1953
X. citri pv. citri	LMG 9322T	USA	1989
X. campestris pv. campestris	LMG 568PT	United Kingdom	1957
X. axonopodis pv. dieffenbachiae	LMG 695PT	Brazil	1965
X. fragariae	LMG 708T	USA	1960
X. oryzae pv. oryzicola	LMG 797PT	Malaysia	1964
X. translucens pv. translucens	LMG 876PT	USA	1933
X. vesicatoria	LMG 911T	New Zealand	1955
X. euvesicatoria pv. euvesicatoria	LMG 922	USA	1939
X. hortorum pv. gardneri	LMG 962T	Yugoslavia	1953
X.euvesicatoria pv. perforans	NCPPB 4321T	USA	1933
X.oryzae pv. oryzae	LMG 5047PT	India	1965

¹ CPBF: Portuguese Collection of Phytopathogenic Bacteria, Instituto Nacional de Investigação Agrária e Veterinária, I.P. Oeiras, Portugal. CFBP: French Collection for Plant-associated Bacteria, Institut National de la Recherche Agronomique, Angers, France. LMG: Belgian Coordinated Collections of Microorganisms/LMG Bacteria Collection, Universiteit Gent—Laboratorium voor Microbiologie, Gent, Belgium. NCPPB: National Collection of Plant Pathogenic Bacteria, Fera Science Ltd., York, UK. Superscript following strain names indicate ^T the type strain of a species and ^PT the pathotype strain for a pathovar.

2.3. Experimental Validation of Putative X. euroxanthea-Specific DNA Markers by Multiplex PCR

A multiplex PCR targeting the most promising DNA markers was optimized to validate a method to rapidly identify X. euroxanthea isolates. XEA1, XEA5 and XEA8 were the chosen markers with distinct amplicon lengths of 819, 295 and 648 bp, respectively; and their broad occurrence in the tested X. euroxanthea strains, apart from XEA1 (absent in CFBP 7622). A 20 µL PCR reaction mix consisted of 1 × DreamTaq Buffer (ThermoFisher Scientific, Waltham, MA, USA), 0.2 mM of each deoxynucleotide triphosphate (dNTP) (Grisp, Porto, Portugal), 0.2 mM of each forward and reverse primers (Table 3), 1.5 U of DreamTaq DNA Polymerase (ThermoFisher Scientific, Waltham, MA, USA) and 25 ng of DNA template.

Table 3.

Selected Xanthomonas euroxanthea-specific DNA markers (XEA1-XEA8), corresponding primer pair sequences, expected amplicon sizes and best BLASTn hits of amplicons with non-X. euroxanthea genomes.

DNA Markers	CDS (MaGe) 1	Locus Tag (NCBI)	Gene Annotation (MaGe)	Primers	Sequences (5′→3′)	Length (bp)	Best BLASTn Hit with Non-X. euroxanthea (E Value/Query Coverage)
XEA1	XE424_v1_a0582	XTG_000508	Conserved protein of unknown function	XEA1F	CTGCCGAGCGTGAAATCCAG	819	-
				XEA1R	CCTTCAGTTGCACCGAACGC
XEA2	XE424_v1_a2605	XTG_002379	Conserved protein of unknown function	XEA2F	AGTCCACCAATGCCATCGCC	425	-
				XEA2R	AGTCCACCAATGCCATCGCC
XEA3	XE424_v1_a2606	XTG_002380	Conserved protein of unknown function	XEA3F	CGGATCGGACAATGACGCTG	612	-
				XEA3R	GCTCTACATCGCCGCTGGAG
XEA4	XE424_v1_a1415	XTG_001287	TetR/AcrR family transcriptional regulator	XEA4F	GACGCATCCGCCCACGACC	341	Dickeya zeae A586-S18-A17 (2 × 10−50/95%)
				XEA4R	TAGGCGGCAGACCCCTTCC
XEA5	XE424_v1_a0462	XTG_000401	MarR family transcriptional regulator	XEA5F	AACGACGCTGACCTGGACC	295	Sphingomonas sp. AP4-R1 (5 × 10−13/73%)
				XEA5R	CGACACCGCACGACCCCG
XEA6	XE424_v1_a0617	XTG_000542	Conserved protein of unknown function	XEA6F	GCGGCTGCAGCGTCGTTG	237	Xanthomonas sp. GW (2 × 10−4/19%)
				XEA6R	TCACCTGATGATCGAAGCCTGG
XEA7	XE424_v1_a1414	n/a	Protein of unknown function	XEA7F	GGACGCGCCATGATCTGCC	212	-
				XEA7R	GGTGTCCGAGGMTCAGGTGC
XEA8	2	2	2	XEA8F	ATCGCCTCTGGATGACGGC	648	Dickeya zeae A586-S18-A17 (1 × 10−95/73%)
				XEA8R	GGTGATGTCGGCAAGCTCG

n/a: not available; ¹: no significant hit has been found. ²: XEA8 DNA marker was designed within the genomic subsequent and partially overlapping CDSs for markers XEA7 and XEA4.

Sterile distilled water was used as the negative control. PCR cycling parameters consisted of a first amplification cycle of 5 min at 95 °C, followed by 35 cycles of 95 °C for 30 s, 61 °C for 15 s and 72 °C for 30 s as well as a final DNA extension at 72 °C for 10 min. The same DNA samples were used as template in PCR reactions using each of the markers individually (XEA1, XEA5 and XEA8) and 1.0 U of DreamTaq DNA polymerase per PCR reaction.

PCR products were separated by electrophoresis on a 0.8% agarose gel (1 × TAE buffer) and visualized using Xpert Green DNA stain (Grisp, Porto, Portugal) with a Molecular Imager Gel Doc XR+ System (Bio-Rad, Hercules, CA, USA). The obtained PCR products for each marker and each strain were purified using the Illustra GFX GEL Band Purification kit (GE Healthcare, Buckinghamshire, UK), following the reference protocol available and sequenced on both strands (STAB Vida, Caparica, Portugal) to confirm their identity and determine the number of SNPs.

2.4. PCR Detection Limit

The detection limit of the multiplex PCR was determined using 10 μL from each of the 10-fold dilutions of X. euroxanthea CPBF 424^T chromosomal DNA prepared in distilled sterile water, ranging from 100 ng to 1 pg per PCR reaction. Multiplex PCR conditions were kept as described above.

2.5. Typing Potential of X. euroxanthea-Specific DNA Markers

Unrooted trees using XEA5, XEA6 and XEA8 markers (transversal to all strains of X. euroxanthea) (Table S2) and partial sequence analysis of the housekeeping genes acnB, fyuA, gyrB and rpoD (Table S3) of 11 X. euroxanthea strains were built to infer the typing potential of these DNA markers. Markers XEA5 (295 bp), XEA6 (237 bp) and XEA8 (648 bp) sequences were concatenated using the Geneious^® v. 9.1.7 and used to build a maximum-likelihood tree based on Tamura-Nei model on MEGA X [25], as previously described [14].

The nucleotide sequences of housekeeping genes acnB, fyuA, gyrB and rpoD were retrieved from the 11 X. euroxanthea genomes, aligned and trimmed to 513, 640, 828 and 793 bp, respectively, and subsequently concatenated using the Geneious^® v. 9.1.7 to build a maximum-likelihood tree as described for the markers. Since these three markers are X. euroxanthea specific, and no homologous marker could be found to define a coherent outgroup, the relatedness of the X. euroxanthea strains was inferred by an unrooted tree.

3. Results

3.1. In Silico Selection of DNA Markers for X. euroxanthea

The selection of X. euroxanthea-specific DNA markers was conducted according to the workflow detailed in Figure 1, which consisted on the use of a platform of comparative genomics (Mage) to screen for unique X. euroxanthea CDSs, further validated by BLASTn.

A list of CDSs exclusively present in the genomes of 11 X. euroxanthea strains and absent from other 12 Xanthomonas species, including the closely related X. arboricola, was deducted from synteny analysis between the 24 genomes (Table 1) along the full length of X. euroxanthea CPBF 424^T genome used as reference (Figure S1). The candidate CDSs (from the reference genome of X. euroxanthea CPBF 424^T) were submitted to a BLASTn to confirm their exclusivity in X. euroxanthea considering high stringency values of E-value, percentage of identity and query coverage and their suitability as X. euroxanthea-specific DNA markers appraised by the design of robust primers.

The data showed that, four CDSs had no significant BLAST hits; three CDSs showed no hits with other Xanthomonas sp. and solely one CDS coding for a putative conserved protein of unknown function matched with Xanthomonas sp. GW (E-value of 2 × 10⁻⁴ and query coverage of 19%) (Table 3). These seven CDSs were selected for the design of eight DNA markers, designated XEA1-XEA8; being that two consecutive and partly overlapping CDSs (XE424_v1_a1415 and XE424_v1_a1414) were used to design marker XEA8 (Table 3). Five of these CDSs are predicted to be proteins of unknown function and two to be transcriptional regulators (Table 3). Ultimately, five DNA markers, XEA4-XEA8, are unfailingly present in the genome of the 11 X. euroxanthea strains analyzed; XEA1 is present in eight genomes; and XEA2 and XEA3 are present in five strains (Figure 2).

Figure 2.

Distribution of six Xanthomonas euroxanthea (Xea)-specific DNA markers (XEA1, XEA2, XEA3, XEA5, XEA6 and XEA8) in 11 X. euroxanthea genomes. The presence/absence of six XEA DNA markers was assessed by BLASTn analysis in Geneious, allowing to disclose three patterns, A to C, that do not translate strain-host affinities.

3.2. Genomic Analysis Unearths the Stability of XEA DNA Markers

To further assess the uniqueness of these X. euroxanthea-specific CDSs used for DNA marker design, several features, including SNPs number, CAI/eCAI values, GC content, chromosomal location and chromosomal proximity to genomic plasticity-determinants were surveyed (Figure 3).

Figure 3.

Circular map of Xanthomonas euroxanthea strain CPBF 424^T chromosome. Outside to inner circles are showing genome coordinates (bp); X. euroxanthea-specific DNA markers XEA1-XEA8 (red); housekeeping genes gyrB, rpoD, fyuA and acnB (yellow); transposases (green); recombinases (light blue); integrases (dark blue) and phage-related ORFs (purple). For each XEA DNA marker and housekeeping gene the number of SNPs (calculated based on 11 genomes of X. euroxanthea), GC content and CAI/eCAI values are shown.

Within the 11 X. euroxanthea genomes considered in this study, the number of SNPs for each marker and housekeeping gene was determined and normalized to the full length of the sequence (i.e., DNA marker or housekeeping gene) as a percentual value indicative of the allelic diversity of the sequences (Figure 3).

CAI/eCAI values of CDSs range between 0.874 (for XE424_v1_a2606 corresponding to DNA marker XEA3) to 1.009 (for XE424_v1_a0617 corresponding to DNA marker XEA6); which are values similar to the range attained with housekeeping genes, namely, 0.918 (acnB) to 1.135 (gyrB) (Figure 3). GC content values for X. euroxanthea-specific CDSs varied between 55.5% (XE424_v1_a2606, XEA3) to 72.5% (XE424_v1_a0617, XEA6), which parallels with X. euroxanthea CPBF 424^T genome GC content value of 65.9% (Figure 3).

Moreover, chromosomal location studies revealed that the XEA markers are scattered throughout the first half of the chromosome and not in the vicinity of genomic plasticity-determinants, including transposases, recombinases, integrases and phage-related ORFs (Figure 3).

The synteny analysis performed with MaGe allowed to investigate the genomic context of the most representative DNA markers, i.e., XEA1, XEA5, XEA6 and XEA8, across 11 X. euroxanthea genomes and comparatively to Xanthomonas arboricola strains CPBF 427 (X. arboricola pv. juglandis) and CPBF 765 (Figure 4) and other Xanthomonas spp. representative strains (Figure S1).

Figure 4.

Comparative syntenic maps of four Xanthomonas euroxanthea-specific DNA marker-harboring-regions (a) XEA1 (designed from a conserved protein of unknown function sequence), (b) XEA5 (design from a MarR family transcriptional regulator), (c) XEA6 (designed from a conserved protein of unknown function sequence) and (d) XEA8 (designed from a protein of unknown function and a TetR/AcrR family transcriptional regulator sequences) DNA markers across 11 X. euroxanthea (Xea) and two Xanthomonas arboricola (Xa) strains.

The data showed that XEA1, XEA5, XEA6 and XEA8 are located in highly syntenic regions across all the X. euroxanthea genomes studied, underlining the absence of genomic rearrangements (Figure 4). Furthermore, CDSs used for the design of markers XEA5, XEA6 and XEA8 are exclusive of X. euroxanthea, being either absent (CDSs encompassing XEA5 and XEA8) or truncated (CDS encompassing XEA6) in X. arboricola strains (Figure 4).

Particularly, the region upstream and including marker XEA1 appears to have suffered erosion in the X. arboricola strains analyzed (CPBF 427 and CPBF 765) and in three of the X. euroxanthea strains studied (BRIP 62411, BRIP 62418 and CFBP 7622). In parallel, this scenario is observed for the region upstream of XEA5 for all strains, except for CPBF 424^T; and downstream of XEA8 for the non-X. euroxanthea strains.

Within the flanking regions of the XEA markers, the CDSs annotated as unknown proteins are the ones where the presence/absence across strains is inconsistent, suggesting that they have been decaying. While, for XEA6 and XEA8 markers, these events are limited to a single CDS immediately flanking the markers, for XEA5 this paradigm is particularly clear as several CDSs annotated as unknown proteins are present in CPBF 424^T, have been lost by the other strains.

3.3. Multiplex PCR Allows for the Confident Identification of X. euroxanthea Strains

Reliable identification of X. euroxanthea isolates by multiplex PCR was optimized for DNA markers XEA1, XEA5 and XEA8, as these originate amplicons of distinguishable size (819, 295 and 648 bp, respectively) and are located in three independent genomic regions of X. euroxanthea. The three markers were successfully amplified in all X. euroxanthea strains analyzed, with exception for XEA1 in X. euroxanthea CFBP 7622 strain (Figure 5) as expected by in silico studies. In addition, no amplification was observed for any of the other 28 xanthomonads, namely 17 strains of X. arboricola, including different pathovars and 9 non-arboricola Xanthomonas species.

Figure 5.

Multiplex PCR using Xanthomonas euroxanthea-specific DNA markers XEA1 (819 bp), XEA8 (648 bp) and XEA5 (295 bp) on 7 X. euroxanthea strains, 10 Xanthomonas arboricola strains, 6 pathovars of Xanthomonas arboricola and 9 non-arboricola Xanthomonas species. Markers XEA5 and XEA8 were successful in detecting X. euroxanthea strains, while XEA1 identified all X. euroxanthea strains, except for CFBP 7622. No amplification was observed for any of the other xanthomonads tested, namely X. arboricola and other Xanthomonas species. C-: negative control.

3.4. Detection Limit of Multiplex PCR with XEA DNA Markers

The detection limit of the multiplex PCR targeting the X. euroxanthea-specific markers XEA1, XEA5 and XEA8, determined through serial dilution of chromosomal DNA, was 1 ng for PCR reaction (Figure S2). When assessing the PCR detection limit of each marker individually, while for XEA1 and XEA8 the detection limit was identical to the multiplex PCR, i.e., 1 ng/PCR reaction, it was ascertained that the limit of detection lowered to 100 pg/PCR reaction for XEA5.

3.5. Typing Potential of Informative XEA DNA Markers

The concatenated sequences of XEA5, XEA6 and XEA8 (1180 bp) for each X. euroxanthea strain studied were aligned and used to generate a maximum-likelihood tree to investigate the discriminatory potential of these X. euroxanthea-specific markers comparatively to four housekeeping genes, corresponding to a concatenated sequence length of 2774 bp, (acnB, fyuA, gyrB and rpoD genes) commonly used for MLSA (Figure 6).

Figure 6.

Maximum-likelihood phylogenetic tree based on concatenated sequences of (a) DNA markers XEA5, XEA6 and XEA8 (1180 bp); and (b) partial housekeeping gene sequences for acnB, fyuA, gyrB and rpoD (2774 bp) extracted from 11 X. euroxanthea genomes. The tree was constructed using the Tamura-Nei model using MEGA X (Kumar et al., 2018 [25]). Supporting values from 1000 bootstrap replicates are indicated near nodes.

Attending that these XEA detection markers are highly specific to X. euroxanthea, no homologous sequences have been found in other bacterial taxa, and therefore the allelic diversity determined by the number of SNP (6 SNP/295nt for XEA5, 10 SNP/237nt for XEA6, and 61 SNP/648nt for XEA8; Figure 3) is represented by an unrooted Maximum-likelihood tree. For both trees, each X. euroxanthea strain is represented by an independent tree branch, with exceptions for strains CPBF 426 and CPBF 761, which clustered together in a single branch. Furthermore, the topology of both trees does not reflect any clustering according to plant–host of isolation nor by pathogenicity or non-pathogenicity phenotypes (Figure 6).

4. Discussion

It is important to investigate the distribution and role played by X. euroxanthea and the closely related X. arboricola within the plant hosts that they frequently co-colonize [12]. Therefore, it is essential to develop a reliable and efficient method for the accurate detection and identification of X. euroxanthea strains and its differentiation from X. arboricola [12].

Additionally, early detection is a critical first step towards timely sanitary intervention aimed at eradicating the pathogen and at reducing the inoculum spread to other plants; thus lessening the disease-induced damage in crops from growth to postharvest processing of products and ensuring agricultural sustainability [26]. Likewise, discriminating one bacterial species from the other would be useful in the application of suitable management procedures.

Recommended standard diagnostic protocols from OEPP/EPPO for other plant-diseases caused by Xanthomonas spp. (namely, X. arboricola pv. corylina, X. axonopodis pv. dieffenbachiae, X. arboricola pv. pruni, X. axonopodis pv. citri, X. fragariae, X. oryzae, X. axonopodis pv. allii, X. euvesicatoria, X. hortorum pv. gardneri, X. perforans and X. vesicatoria) still require several steps-observation of disease symptoms, microscopic examination, pathogen isolation, pathogenicity tests and molecular tests [27,28,29,30,31,32,33,34]. Having in mind that only recently X. euroxanthea was proposed as a new species [4], isolated from different plant hosts [7,8,9,10], for which the nature of the bacteria–plant interaction is still unknown, and no distinct symptoms have been described [4], the herein proposed multiplex PCR is the tool available for accurate detection and identification of X. euroxanthea.

While multiplex PCRs have been proposed to detect X. arboricola pv. juglandis [14] or X. arboricola pv. pruni [35], the present work describes eight X. euroxanthea-specific DNA markers (XEA1-XEA8) and proposes a methodology to detect and genotype X. euroxanthea strains. In view of that, using a comparative genomics strategy and by assessing a pool of numerous xanthomonads genomes, including several X. euroxanthea and X. arboricola, it was possible to narrow down X. euroxanthea-specific genomic regions to only seven CDSs. Genomes alignment is emerging as a fast and convenient solution to rapidly identify species-specific DNA markers to implement PCR-based detection methods, as recently evidenced for Xanthomonas campestris pv. raphani [36].

The specificity of these putative X. euroxathea-specific DNA markers were further validated by BLASTn analysis using X. euroxanthea type strain CPBF 424^T as query sequence, followed by an in silico workflow essentially as described in previous studies [20,37], to determine the genomic context of each X. euroxanthea-specific loci and particularly to ensure that these putative DNA markers are within conserved and stable genomic regions.

Specifically, four CDSs showed no significant BLASTn hits outside of X. euroxanthea, three CDSs showed some similarity with non Xanthomonas species, and only one CDS matched with a Xanthomonas spp. but with a poor similarity as shown by an E-value of 2 × 10⁻⁴ and query coverage of 19%. Based on these data, eight primer pairs corresponding to eight markers (XEA1-XEA8), were designed to nest in these seven CDSs. Marker XEA8 was designed from two consecutive and partly overlapping CDSs.

Furthermore, the chromosomal distance between each of the X. euroxanthea DNA markers (XEA1-XEA8) and genomic plasticity-determinants, such as transposases, recombinases, integrases and phage-related ORFs attests to the low genomic plasticity and high stability of the DNA marker-harboring-regions, as previously suggested [20]. All seven specific CDSs on which XEA markers are founded appear to be well adapted to the codon usage and CG content, thereby, suggesting that these CDSs were not recently acquired by Horizontal Gene Transfer—HGT [20,38].

Such findings are consistent with the hypothesis that these CDSs are conserved in the X. euroxanthea genomes and likely present across all the X. euroxanthea strains regardless of infrasubspecific variability. A BLAST analysis of each DNA-marker (XEA1-XEA8) against the genomes of 11 X. euroxanthea showed that five DNA markers (XEA4-XEA8) are unfailingly present in the genome of the 11 X. euroxanthea strains analyzed.

One marker (XEA1) is absent in three out of the 11 X. euroxanthea genomes analyzed, and markers XEA2 and XEA3 are present in the four X. euroxanthea strains isolated from walnut (CPBF 367, CPBF 424^T, CPBF 426 and CFBP 7653) and in one strain isolated from pecan (CPBF 766). These results suggest that the CDSs used for the design of markers XEA4-XEA8 are within the core genome of the 11 X. euroxanthea studied, whereas the CDSs corresponding to XEA1-XEA3 markers, although X. euroxanthea-specific, are part of the accessory genome [39].

An interesting attribute of X. euroxanthea is the encompassment of pathogenic (CPBF 424^T) and non-pathogenic strains (CPBF 367) by the same host [4]. Considering the pathogenicity and non-pathogenicity phenotypes; different plant hosts of isolation; and the occurrence of markers XEA1-XEA8 across the 11 studied X. euroxanthea, one may conclude that these X. euroxanthea-specific markers are not biased by pathogenicity phenotype or plant host species.

To determine the allelic variation of the markers and infer their potential to discriminate X. euroxanthea strains, the number of SNPs of markers shown to be present in the 11 X. euroxanthea strains (i.e., XEA5, XEA6 and XEA8) was determined. The data indicate that XEA8 with a 9.4% of SNPs 61 SNP/648 bp) stands as the most informative X. euroxanthea-specific DNA marker for typing purposes, as it surpasses the 2% and 4.2% of SNPs recorded for XEA5 and XEA6 markers, respectively (6 SNP/295 bp for XEA5; and 10 SNP/237 bp for XEA6).

Although the allelic variation of XEA5 and XEA6 may be inferior to the allelic variation observed for the housekeeping genes commonly used for MLSA (acnB (6.2%), fyuA (8.3%), gyrB (7%) and rpoD (5.3%)), when combined with XEA8, it is nevertheless sufficient to separate the 11 X. euroxanthea strains as efficiently as the housekeeping genes. These results suggest that these three X. euroxanthea-specific DNA markers (XEA5, XEA6 and XEA8) may be used both for the detection and genotyping of X. euroxanthea. Since the allelic variation was conducted considering 11 genomes of X. euroxanthea, the number of SNPs may be underestimated, i.e., novel X. euroxanthea strains may reveal novel single nucleotide substitutions in the sequence of the mentioned markers.

The synteny analyses conducted to investigate the genomic context of the XEA markers across 11 X. euroxanthea and two strains of the closely related species X. arboricola, revealed that XEA markers and their flanking CDSs are syntenic for X. euroxanthea. It is worth mentioning that, within the flanking regions of XEA markers, CDSs annotated as unknown proteins are intermittently absent when compared to CDSs of proteins of known function, suggesting genomic erosion.

This paradigm is further supported by the genomic context of the XEA5 marker where two CDSs, namely, the toxin RTX-I translocation ATP-binding protein and the membrane fusion protein (MFP) family protein remain present in CPBF 424^T, CPBF 766, BRIP 62418 and CPBF 427 despite the overall loss of the flanking unknown proteins. The fact that the CDS of the XEA6 marker is present in X. euroxanthea strains and truncated in X. arboricola strains suggests a common and recent ancestrality of these two species.

Distinctively, markers designed from CDSs annotated as a family transcriptional regulator, namely XEA5 and XEA8, are likely to portray an essential role in cell life and consequently remain conserved in X. euroxanthea, which is further supported by their inclusion as part of the core genome.

Overall, the data gathered regarding synteny, CAI/eCAI values and the GC content of X. euroxanthea-specific CDSs concerning markers XEA1, XEA5, XEA6 and XEA8, suggest that these markers and their flanking regions are not the result of recent HGT acquisitions events by X. euroxanthea, which increases its reliability for X. euroxanthea detection and identification, as has been hypothesized for other bacteria [20,37].

A multiplex PCR detection assay targeting XEA1, XEA5 and XEA8 markers was successfully developed for the identification of X. euroxanthea. The annealing temperature and extension time were optimized to steer clear of non-specific amplifications of DNA, which was also reportedly been done in previous studies [40].

This multiplex PCR was shown to be both specific and efficient given that neither non-specific amplifications (bands of unexpected sizes) in X. euroxanthea nor amplification in non-target Xanthomonas spp. tested were observed. As predicted by the in silico assessments, the amplicons corresponding to XEA1, XEA5 and XEA8 markers were obtained for all X. euroxanthea strains, with the exception of XEA1 for X. euroxanthea CFBP 7622. These results emphasize the robustness of the multiplex PCR for the detection and identification of X. euroxanthea.

When assessing the PCR detection limit of each marker individually, it was observed that XEA5 lowered the limit of detection value to 10 pg/µL, indicating that the sensitivity of the simplex PCR targeting XEA5 marker is ten-fold higher than the multiplex PCR. The detection limit of 0.1 ng/µL attained for the multiplex PCR is similar to what has been reported in other studies (0.02 and 0.5 ng/µL) [40,41].

To investigate the genotyping potential of the chosen markers, the allelic variation of the three markers present in all studied X. euroxanthea strains (XEA5, XEA6 and XEA8) was compared with four housekeeping genes commonly used for MLSA (acnB, fyuA, gyrB and rpoD) and represented as unrooted trees. The trees obtained had a similar topology, as each strain was allocated to an independent tree branch, except for strains CPBF 426 and CPBF 761, which clustered together in a single branch, and no SNPs were observed within the concatenated sequences of the XEA markers (1180 bp) nor within the partial housekeeping genes (2774 bp).

Such data reveals that markers and housekeeping genes are similarly informative in discriminating X. euroxanthea strains; and that the studied X. euroxanthea strains are not clustered according to plant–host isolation, pathogenicity or non-pathogenicity phenotypes.

By comparing marker-presence profiles to plant host species from which the X. euroxanthea strains were isolated, three patterns identified as A, B and C (Figure 2) are observed. Attending that all XEA markers are in syntenic genomic regions that overlap with homologs regions in the closely related X. arboricola strains, we may hypothesize that the X. euroxanthea strains isolated from two Juglandaceae species (Juglans regia and Carya illinoinensis) with pattern A, i.e., possessing all XEA markers (XEA1-XEA8), are ancestors of strains included in pattern B (i.e., which lost XEA2 and XEA3) and isolated from Carya illinoinensis and Solanaceaceae (Solanum lycopersicum) and pattern C (i.e., which lost XEA1, XEA2 and XEA3), isolated from Solanum lycopersicum and a Fabaceae (Phaseolus vulgaris).

Thus, rather than an acquisition of XEA2 and XEA3 markers in five of the eleven studied X. euroxanthea strains (pattern A), these markers were likely lost in the other X. euroxanthea strains, followed by XEA1 marker shown to be absent in 3 out the 11 X. euroxanthea strains, leading to the emergence of recent X. euroxanthea lineage characterized by marker-profile C (BRIP 62411, BRIP 62418) and Phaseolus vulgaris (CFBP 7622). These data suggest that the loss of markers XEA2, XEA3 and XEA1 i.e., patterns A-C, occurred progressively as X. euroxanthea lineages extended to a new host, specifically, from walnut, to pecan, to tomato, to the common bean. These results are aligned with studies describing events of genome erosion as a consequence of bacterial adaptation to a new plant–host [42,43].

To summarize, we may infer that the evolutionary trend of XEA marker loss follows progressive host-jump events from walnut, to pecan, to tomato and finally to common bean.

5. Conclusions

The present study proposes eight specific, efficient and reliable DNA markers for the detection and identification of X. euroxanthea isolates. The allelic variation of some of these markers allows to conciliate the detection and genotyping of X. euroxanthea strains, contributing to survey these bacteria in ecological niches colonized by the closely related X. arboricola. The multiplex PCR was shown to be highly specific, as solely the target DNA (i.e., X. euroxanthea) was amplified; and efficient, as an amplicon was observed with all tested X. euroxanthea strains.

The present study also provided a successful workflow for the selection of molecular markers, which is able to be implemented in the selection of species-specific genomic regions for any other taxa.

By analyzing a marker’s presence across strains of different colonizing plant host species, we may infer that X. euroxanthea colonization of different plant host species occurred at different points in time.

Supplementary Materials

The following supporting information can be downloaded at: https://www.mdpi.com/article/10.3390/microorganisms10061078/s1, Table S1. MaGe labels of the seven selected X. euroxanthea-specific CDSs of 11 X. euroxanthea genomes used to design XEA1 to XEA8 markers. Table S2. Genomic coordinates of the eight X. euroxanthea-specific DNA markers of 11 X. euroxanthea genomes. Table S3. MaGe labels of four housekeeping genes from 11 X. euroxanthea genomes used in the construction of an unrooted tree. Figure S1. Synteny map of DNA markers (a) XEA1, (b) XEA5, (c) XEA6 and (d) XEA8 across the genomes of 11 X. euroxanthea and 24 other Xanthomonas spp. strains. Figure S2. The PCR detection limits assessed using purified DNA from CPBF 424^T. C-: negative control (sterile distilled water).

Author Contributions

Conceptualization: F.T., J.F.P. and L.M.; validation, writing—original draft preparation and formal analysis: K.G.S., F.T. and L.M.; investigation: K.G.S., L.M. and M.T.; writing—review and editing: K.G.S., F.T., J.F.P., L.M. and M.T.; supervision: F.T. and L.M.; project administration, funding acquisition and resources: F.T. All authors have read and agreed to the published version of the manuscript.

Data Availability Statement

Not applicable.

Conflicts of Interest

The authors declare no conflict of interest. The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript, or in the decision to publish the results.

Funding Statement

Funding for this work was provided by the EU Horizon 2020 Research and Innovation program (grant agreement 857251), European Structural & Investment Funds (ESIFs) through COMPETE 2020, National Funds through FCT (in the scope of the projects EVOXANT-PTDC/BIA-EVF/ 3635/2014-POCI-01-0145-FEDER-016600; and UIDB/50027/2020) and the Operational Thematic Program for Competitiveness and Internationalization (POCI) under the PORTUGAL 2020 Partnership Agreement through the European Regional Development Fund (FEDER). L.M. was supported by fellowship from FCT (SFRH/BD/137079/2018). J.F.P. acknowledges support from the Department of Life Sciences and Facility Management of the Zurich University of Applied Sciences (ZHAW) in Wädenswil. This article is based upon work from COST Action CA16107 EuroXanth, supported by COST (European Cooperation in Science and Technology). The APC was funded by EU’s Horizon 2020 Research and Innovation program through BIOPOLIS (grant agreement 857251).