Incongruence between multi-locus sequence analysis (MLSA) and whole-genome-based phylogenies: Pseudomonas syringae pathovar pisi as a cautionary tale

Average Nucleotide Identity and Digital DNA-DNA Hybridization Analysis Following PromethION Nanopore-Based Whole Genome Sequencing Allows for Accurate Prokaryotic Typing

Whole-genome sequencing (WGS) is revolutionizing clinical bacteriology. However, bacterial typing remains investigated by reference techniques with inherent limitations. This stresses the need for alternative methods providing robust and accurate sequence type (ST) classification. This study optimized and evaluated a GridION nanopore sequencing protocol, adapted for the PromethION platform. Forty-eight Escherichia coli clinical isolates with diverse STs were sequenced to assess two alternative typing methods and resistance profiling applications. Multi-locus sequence typing (MLST) was used as the reference typing method. Genomic relatedness was assessed using Average Nucleotide Identity (ANI) and digital DNA-DNA Hybridization (DDH), and cut-offs for discriminative strain resolution were evaluated. WGS-based antibiotic resistance prediction was compared to reference Minimum Inhibitory Concentration (MIC) assays. We found ANI and DDH cut-offs of 99.3% and 94.1%, respectively, which correlated well with MLST classifications and demonstrated potentially higher discriminative resolution than MLST. WGS-based antibiotic resistance prediction showed categorical agreements of ≥ 93% with MIC assays for amoxicillin, ceftazidime, amikacin, tobramycin, and trimethoprim-sulfamethoxazole. Performance was suboptimal (68.8-81.3%) for amoxicillin-clavulanic acid, cefepime, aztreonam, and ciprofloxacin. A minimal sequencing coverage of 12× was required to maintain essential genomic features and typing accuracy. Our protocol allows the integration of PromethION technology in clinical laboratories, with ANI and DDH proving to be accurate and robust alternative typing methods, potentially offering superior resolution. WGS-based antibiotic resistance prediction holds promise for specific antibiotic classes.

OrthoPhyl-streamlining large-scale, orthology-based phylogenomic studies of bacteria at broad evolutionary scales

There are a staggering number of publicly available bacterial genome sequences (at writing, 2.0 million assemblies in NCBI's GenBank alone), and the deposition rate continues to increase. This wealth of data begs for phylogenetic analyses to place these sequences within an evolutionary context. A phylogenetic placement not only aids in taxonomic classification but informs the evolution of novel phenotypes, targets of selection, and horizontal gene transfer. Building trees from multi-gene codon alignments is a laborious task that requires bioinformatic expertise, rigorous curation of orthologs, and heavy computation. Compounding the problem is the lack of tools that can streamline these processes for building trees from large-scale genomic data. Here we present OrthoPhyl, which takes bacterial genome assemblies and reconstructs trees from whole genome codon alignments. The analysis pipeline can analyze an arbitrarily large number of input genomes (>1200 tested here) by identifying a diversity-spanning subset of assemblies and using these genomes to build gene models to infer orthologs in the full dataset. To illustrate the versatility of OrthoPhyl, we show three use cases: E. coli/Shigella, Brucella/Ochrobactrum and the order Rickettsiales. We compare trees generated with OrthoPhyl to trees generated with kSNP3 and GToTree along with published trees using alternative methods. We show that OrthoPhyl trees are consistent with other methods while incorporating more data, allowing for greater numbers of input genomes, and more flexibility of analysis.

An ANI gap within bacterial species that advances the definitions of intra-species units

IMPORTANCE

Bacterial strains and clonal complexes are two cornerstone concepts for microbiology that remain loosely defined, which confuses communication and research. Here we identify a natural gap in genome sequence comparisons among isolate genomes of all well-sequenced species that has gone unnoticed so far and could be used to more accurately and precisely define these and related concepts compared to current methods. These findings advance the molecular toolbox for accurately delineating and following the important units of diversity within prokaryotic species and thus should greatly facilitate future epidemiological and micro-diversity studies across clinical and environmental settings.

Bacterial dispersal and biogeography as underappreciated influences on phytobiomes

Bacterial strains are not distributed evenly throughout the environment. Here I explore how differential distribution and dispersal patterns of bacteria could affect interactions and coevolutionary dynamics with plants, and highlight ways that variation could be taken advantage of to develop robust and effective microbial consortia to inoculate crops. Questions about biogeographical patterns in viruses, fungi, and other eukaryotes are equally as prevalent and important for agriculture, and are in some cases more thoroughly explored. For simplicity as well as to bring attention to bacterial biogeography and dispersal in the context of plant interactions, I focus solely on bacterial patterns and questions for this article. The next few years will no doubt bring great advances in our understanding of dispersal capabilities and population dynamics for many plant-associated bacteria, and one of the next looming challenges will be learning to harvest this diversity in ways that can benefit agriculture.

A complete genome sequence for Pseudomonas syringae pv. pisi PP1 highlights the importance of multiple modes of horizontal gene transfer during phytopathogen evolution

Hybrid assembly strategies that combine long-read sequencing reads from Oxford Nanopore's MinION device combined with high-depth Illumina paired-end reads have enabled completion and circularization of both plasmids and chromosomes from multiple bacterial strains. Here we demonstrate the utility of supplementing Illumina paired-end reads from a previously published draft genome of P. syringae pv. pisi PP1 with long reads to generate a complete genome sequence for this strain. The phylogenetic placement and genomic repertoire of virulence factors within this strain provides a unique perspective on virulence evolution within P. syringae phylogroup 2, and highlights that strains can rapidly acquire virulence factors through horizontal gene transfer by acquisition of plasmids as well as through chromosomal recombination.

Inference of Convergent Gene Acquisition Among Pseudomonas syringae Strains Isolated From Watermelon, Cantaloupe, and Squash

Pseudomonas syringae sensu stricto (phylogroup 2; referred to as P. syringae) consists of an environmentally ubiquitous bacterial population associated with diseases of numerous plant species. Recent studies using multilocus sequence analysis have indicated the clonal expansion of several P. syringae lineages, located in phylogroups 2a and 2b, in association with outbreaks of bacterial spot disease of watermelon, cantaloupe, and squash in the United States. To investigate the evolutionary processes that led to the emergence of these epidemic lineages, we sequenced the genomes of six P. syringae strains that were isolated from cucurbits grown in the United States, Europe, and China over a period of more than a decade, as well as eight strains that were isolated from watermelon and squash grown in six different Florida counties during the 2013 and 2014 seasons. These data were subjected to comparative analyses along with 42 previously sequenced genomes of P. syringae stains collected from diverse plant species and environments available from GenBank. Maximum likelihood reconstruction of the P. syringae core genome revealed the presence of a hybrid phylogenetic group, comprised of cucurbit strains collected in Florida, Italy, Serbia, and France, which emerged through genome-wide homologous recombination between phylogroups 2a and 2b. Functional analysis of the recombinant core genome showed that pathways involved in the ATP-dependent transport and metabolism of amino acids, bacterial motility, and secretion systems were enriched for recombination. A survey of described virulence factors indicated the convergent acquisition of several accessory type 3 secreted effectors (T3SEs) among phylogenetically distinct lineages through integrative and conjugative element and plasmid loci. Finally, pathogenicity assays on watermelon and squash showed qualitative differences in virulence between strains of the same clonal lineage, which correlated with T3SEs acquired through various mechanisms of horizontal gene transfer (HGT). This study provides novel insights into the interplay of homologous recombination and HGT toward pathogen emergence and highlights the dynamic nature of P. syringae sensu lato genomes.

Recombination of ecologically and evolutionarily significant loci maintains genetic cohesion in the Pseudomonas syringae species complex

Background

Pseudomonas syringae is a highly diverse bacterial species complex capable of causing a wide range of serious diseases on numerous agronomically important crops. We examine the evolutionary relationships of 391 agricultural and environmental strains using whole-genome sequencing and evolutionary genomic analyses.

Results

We describe the phylogenetic distribution of all 77,728 orthologous gene families in the pan-genome, reconstruct the core genome phylogeny using the 2410 core genes, hierarchically cluster the accessory genome, identify the diversity and distribution of type III secretion systems and their effectors, predict ecologically and evolutionary relevant loci, and establish the molecular evolutionary processes operating on gene families. Phylogenetic and recombination analyses reveals that the species complex is subdivided into primary and secondary phylogroups, with the former primarily comprised of agricultural isolates, including all of the well-studied P. syringae strains. In contrast, the secondary phylogroups include numerous environmental isolates. These phylogroups also have levels of genetic diversity typically found among distinct species. An analysis of rates of recombination within and between phylogroups revealed a higher rate of recombination within primary phylogroups than between primary and secondary phylogroups. We also find that “ecologically significant” virulence-associated loci and “evolutionarily significant” loci under positive selection are over-represented among loci that undergo inter-phylogroup genetic exchange.

Conclusions

While inter-phylogroup recombination occurs relatively rarely, it is an important force maintaining the genetic cohesion of the species complex, particularly among primary phylogroup strains. This level of genetic cohesion, and the shared plant-associated niche, argues for considering the primary phylogroups as a single biological species.

Electronic supplementary material

The online version of this article (10.1186/s13059-018-1606-y) contains supplementary material, which is available to authorized users.

Comparative genomics of Pseudomonas syringae reveals convergent gene gain and loss associated with specialization onto cherry (Prunus avium)

Genome-wide analyses of the effector- and toxin-encoding genes were used to examine the phylogenetics and evolution of pathogenicity amongst diverse strains of Pseudomonas syringae causing bacterial canker of cherry (Prunus avium), including pathovars P. syringae pv morsprunorum (Psm) races 1 and 2, P. syringae pv syringae (Pss) and P. syringae pv avii. Phylogenetic analyses revealed Psm races and P. syringae pv avii clades were distinct and were each monophyletic, whereas cherry-pathogenic strains of Pss were interspersed amongst strains from other host species. A maximum likelihood approach was used to predict effectors associated with pathogenicity on cherry. Pss possesses a smaller repertoire of type III effectors but has more toxin biosynthesis clusters than Psm and P. syringae pv avii. Evolution of cherry pathogenicity was correlated with gain of genes such as hopAR1 and hopBB1 through putative phage transfer and horizontal transfer respectively. By contrast, loss of the avrPto/hopAB redundant effector group was observed in cherry-pathogenic clades. Ectopic expression of hopAB and hopC1 triggered the hypersensitive reaction in cherry leaves, confirming computational predictions. Cherry canker provides a fascinating example of convergent evolution of pathogenicity that is explained by the mix of effector and toxin repertoires acting on a common host.

Genetic Diversity and Virulence of Wheat and Barley Strains of Xanthomonas translucens from the Upper Midwestern United States

Bacterial leaf streak (BLS) of wheat and barley, caused by Xanthomonas translucens pv. undulosa and X. translucens pv. translucens, has been of growing concern in small grains production in the Upper Midwestern United States. To optimize disease resistance breeding, a greater awareness is needed of the pathovars and genetic diversity within the pathogens causing BLS in the region. Multilocus sequencing typing (MLST) and analysis (MLSA) of four common housekeeping genes (rpoD, dnaK, fyuA, and gyrB) was used to evaluate the genetic diversity of 82 strains of X. translucens isolated between 2006 and 2013 from wheat, barley, rye, and intermediate wheatgrass. In addition, in planta disease assays were conducted on 75 strains to measure relative virulence in wheat and barley. All strains were determined by MLSA to be related to X. translucens pv. undulosa and X. translucens pv. translucens. Clustering of strains based on Bayesian, network, and minimum spanning trees correlated with relative virulence levels in inoculated wheat and barley. Thus, phylogeny based on rpoD, dnaK, fyuA, and gyrB correlated with host of isolation and was an effective means for predicting virulence of strains belonging to X. translucens pv. translucens and X. translucens pv. undulosa.

Absence of genome reduction in diverse, facultative endohyphal bacteria

Fungi interact closely with bacteria, both on the surfaces of the hyphae and within their living tissues (i.e. endohyphal bacteria, EHB). These EHB can be obligate or facultative symbionts and can mediate diverse phenotypic traits in their hosts. Although EHB have been observed in many lineages of fungi, it remains unclear how widespread and general these associations are, and whether there are unifying ecological and genomic features can be found across EHB strains as a whole. We cultured 11 bacterial strains after they emerged from the hyphae of diverse Ascomycota that were isolated as foliar endophytes of cupressaceous trees, and generated nearly complete genome sequences for all. Unlike the genomes of largely obligate EHB, the genomes of these facultative EHB resembled those of closely related strains isolated from environmental sources. Although all analysed genomes encoded structures that could be used to interact with eukaryotic hosts, pathways previously implicated in maintenance and establishment of EHB symbiosis were not universally present across all strains. Independent isolation of two nearly identical pairs of strains from different classes of fungi, coupled with recent experimental evidence, suggests horizontal transfer of EHB across endophytic hosts. Given the potential for EHB to influence fungal phenotypes, these genomes could shed light on the mechanisms of plant growth promotion or stress mitigation by fungal endophytes during the symbiotic phase, as well as degradation of plant material during the saprotrophic phase. As such, these findings contribute to the illumination of a new dimension of functional biodiversity in fungi.

Comparative genomics reveals genes significantly associated with woody hosts in the plant pathogen Pseudomonas syringae

The diversification of lineages within Pseudomonas syringae has involved a number of adaptive shifts from herbaceous hosts onto various species of tree, resulting in the emergence of highly destructive diseases such as bacterial canker of kiwi and bleeding canker of horse chestnut. This diversification has involved a high level of gene gain and loss, and these processes are likely to play major roles in the adaptation of individual lineages onto their host plants. In order to better understand the evolution of P. syringae onto woody plants, we have generated de novo genome sequences for 26 strains from the P. syringae species complex that are pathogenic on a range of woody species, and have looked for statistically significant associations between gene presence and host type (i.e. woody or herbaceous) across a phylogeny of 64 strains. We have found evidence for a common set of genes associated with strains that are able to colonize woody plants, suggesting that divergent lineages have acquired similarities in genome composition that may form the genetic basis of their adaptation to woody hosts. We also describe in detail the gain, loss and rearrangement of specific loci that may be functionally important in facilitating this adaptive shift. Overall, our analyses allow for a greater understanding of how gene gain and loss may contribute to adaptation in P. syringae.

Divorcing Strain Classification from Species Names

Confusion about strain classification and nomenclature permeates modern microbiology. Although taxonomists have traditionally acted as gatekeepers of order, the numbers of, and speed at which, new strains are identified has outpaced the opportunity for professional classification for many lineages. Furthermore, the growth of bioinformatics and database-fueled investigations have placed metadata curation in the hands of researchers with little taxonomic experience. Here I describe practical challenges facing modern microbial taxonomy, provide an overview of complexities of classification for environmentally ubiquitous taxa like Pseudomonas syringae, and emphasize that classification can be independent of nomenclature. A move toward implementation of relational classification schemes based on inherent properties of whole genomes could provide sorely needed continuity in how strains are referenced across manuscripts and data sets.

Diversity of Xanthomonas campestris Isolates from Symptomatic Crucifers in New York State

To assess the diversity of Xanthomonas campestris spp. infecting crucifers in New York, 154 isolates were collected over 10 years across the state. The goal was to determine if isolates of the pathogen were overwintering in New York and serving as primary inoculum in subsequent years, or if novel isolates were entering the state each year. Pure cultures of isolates were characterized using multilocus sequence analysis (MLSA), a greenhouse pathogenicity assay, repetitive element-polymerase chain reaction (Rep-PCR) using the BOX-A1R primer, and enzyme-linked immunosorbent assay. The MLSA scheme proved to be more efficient than Rep-PCR for a large sample population and for comparison with global isolates. X. campestris isolated from crucifers in New York comprised of X. campestris pv. campestris and X. campestris pv. raphani, with X. campestris pv. raphani being predominately isolated from transplants. Evidence for unique haplotypes persisting on the same farm for several years due to improper seedbed rotations was documented in addition to novel haplotypes being spread throughout states through infected transplants and seed. Rep-PCR confirmed the high diversity of X. campestris and was used to generate 15 unique fingerprint patterns from isolates collected in the first 5 years. A worldwide comparison of isolates suggests that the X. campestris pv. campestris population appears to be very homogenous with dominant haplotypes persisting for extended periods and being globally disseminated.

Phylogenomic analysis of the genus Ralstonia based on 686 single-copy genes

The genus Ralstonia contains species that are devastating plant pathogens, opportunistic human pathogens, and/or important degraders of xenobiotic and recalcitrant compounds. However, significant nomenclature problems exist, especially for the Ralstonia solanacearum species complex which consists of four phylotypes. Phylogenomics of the Ralstonia genus was investigated via a comprehensive analysis of 39 Ralstonia genomes as well as four genomes of Cupriavidus necator (more commonly known by its previous name Ralstonia eutropha). These data revealed 686 single-copy orthologs that could be extracted from the Ralstonia core-genome and used to reconstruct the phylogeny of the genus Ralstonia. The generated tree has strong bootstrap support for almost all branches. We also estimated the in silico DNA-DNA hybridization (isDDH) and the average nucleotide identity (ANI) values between each genome. Our data confirmed that whole genome sequence data provides a powerful tool to resolve the complex taxonomic questions of the genus Ralstonia, e.g. strains of Ralstonia solanacearum phylotype IIA and IIB may represent two subspecies of R. solanacearum, and strains of R. solanacearum phylotype I and III may be classified into two subspecies of Ralstonia pseudosolanacearum. Recently, strains of R. solanacearum phylotype IV were proposed to be reclassified into different subspecies of Ralstonia syzygii; our study, however, showed that phylotype IV strains had high isDDH values (83.8-96.1 %), indicating it may be not appropriate to classify these closely related strains into different subspecies. We also evaluated the performance of six chromosomal housekeeping genes (gdhA, mutS, adk, leuS, rplB and gyrB) used in Ralstonia phylogenetic inference. The multilocus sequence analysis of these six marker genes was able to reliably infer the phylogenetic relationships of the genus Ralstonia.

A user's guide to a data base of the diversity of Pseudomonas syringae and its application to classifying strains in this phylogenetic complex

The Pseudomonas syringae complex is composed of numerous genetic lineages of strains from both agricultural and environmental habitats including habitats closely linked to the water cycle. The new insights from the discovery of this bacterial species in habitats outside of agricultural contexts per se have led to the revelation of a wide diversity of strains in this complex beyond what was known from agricultural contexts. Here, through Multi Locus Sequence Typing (MLST) of 216 strains, we identified 23 clades within 13 phylogroups among which the seven previously described P. syringae phylogroups were included. The phylogeny of the core genome of 29 strains representing nine phylogroups was similar to the phylogeny obtained with MLST thereby confirming the robustness of MLST-phylogroups. We show that phenotypic traits rarely provide a satisfactory means for classification of strains even if some combinations are highly probable in some phylogroups. We demonstrate that the citrate synthase (cts) housekeeping gene can accurately predict the phylogenetic affiliation for more than 97% of strains tested. We propose a list of cts sequences to be used as a simple tool for quickly and precisely classifying new strains. Finally, our analysis leads to predictions about the diversity of P. syringae that is yet to be discovered. We present here an expandable framework mainly based on cts genetic analysis into which more diversity can be integrated.

Pseudomonas syringae CC1557: a highly virulent strain with an unusually small type III effector repertoire that includes a novel effector

Both type III effector proteins and nonribosomal peptide toxins play important roles for Pseudomonas syringae pathogenicity in host plants, but whether and how these pathways interact to promote infection remains unclear. Genomic evidence from one clade of P. syringae suggests a tradeoff between the total number of type III effector proteins and presence of syringomycin, syringopeptin, and syringolin A toxins. Here, we report the complete genome sequence from P. syringae CC1557, which contains the lowest number of known type III effectors to date and has also acquired genes similar to sequences encoding syringomycin pathways from other strains. We demonstrate that this strain is pathogenic on Nicotiana benthamiana and that both the type III secretion system and a new type III effector, hopBJ1, contribute to pathogenicity. We further demonstrate that activity of HopBJ1 is dependent on residues structurally similar to the catalytic site of Escherichia coli CNF1 toxin. Taken together, our results provide additional support for a negative correlation between type III effector repertoires and the potential to produce syringomycin-like toxins while also highlighting how genomic synteny and bioinformatics can be used to identify and characterize novel virulence proteins.

Genomics-Based Exploration of Virulence Determinants and Host-Specific Adaptations of Pseudomonas syringae Strains Isolated from Grasses

The Pseudomonas syringae species complex has recently been named the number one plant pathogen, due to its economic and environmental impacts, as well as for its role in scientific research. The bacterium has been repeatedly reported to cause outbreaks on bean, cucumber, stone fruit, kiwi and olive tree, as well as on other crop and non-crop plants. It also serves as a model organism for research on the Type III secretion system (T3SS) and plant-pathogen interactions. While most of the current work on this pathogen is either carried out on one of three model strains found on dicot plants with completely sequenced genomes or on isolates obtained from recent outbreaks, not much is known about strains isolated from grasses (Poaceae). Here, we use comparative genomics in order to identify putative virulence-associated genes and other Poaceae-specific adaptations in several newly available genome sequences of strains isolated from grass species. All strains possess only a small number of known Type III effectors, therefore pointing to the importance of non-Type III secreted virulence factors. The implications of this finding are discussed.