Population subdivision and the detection of recombination in non-typable Haemophilus influenzae

Identifying Bacterial Airways Infection in Stable Severe Asthma Using Oxford Nanopore Sequencing Technologies

Previous metagenomic studies in asthma have been limited by inadequate sequencing depth for species-level bacterial identification and by heterogeneity in clinical phenotyping. We hypothesize that chronic bacterial airways infection is a key "treatable trait" whose prevalence, clinical phenotype and reliable biomarkers need definition. In this study, we have applied a method for Oxford Nanopore sequencing for the unbiased metagenomic characterization of severe asthma. We optimized methods to compare performance of Illumina MiSeq, Nanopore sequencing, and RT-qPCR on total sputum DNA extracts against culture/MALDI-TOF for analysis of induced sputum samples from highly phenotyped severe asthma during clinical stability. In participants with severe asthma (n = 23) H. influenzae was commonly cultured (n = 8) and identified as the dominant bacterial species by metagenomic sequencing using an optimized method for Illumina MiSeq and Oxford Nanopore. Alongside superior operational characteristics, Oxford Nanopore achieved near complete genome coverage of H. influenzae and demonstrated a high level of agreement with Illumina MiSeq data. Clinically significant infection was confirmed with validated H. influenzae plasmid-based quantitative PCR assay. H. influenzae positive patients were found to have sputum neutrophilia and lower FeNO. In conclusion, using an optimized method of direct sequencing of induced sputum samples, H. influenzae was identified as a clinically relevant pathogen in severe asthma and was identified reliably using metagenomic sequencing. Application of these protocols in ongoing analysis of large patient cohorts will allow full characterization of this clinical phenotype. IMPORTANCE The human airways were once thought sterile in health. Now metagenomic techniques suggest bacteria may be present, but their role in asthma is not understood. Traditional culture lacks sensitivity and current sequencing techniques are limited by operational problems and limited ability to identify pathogens at species level. We optimized a new sequencing technique-Oxford Nanopore technologies (ONT)-for use on human sputum samples and compared it with existing methods. We found ONT was effective for rapidly analyzing samples and could identify bacteria at the species level. We used this to show Haemophilus influenzae was a dominant bacterium in the airways in people with severe asthma. The presence of Haemophilus was associated with a "neutrophilic" form of asthma - a subgroup for which we currently lack specific treatments. Therefore, this technique could be used to target chronic antibiotic therapy and in research to characterize the full breadth of bacteria in the airways.

Comparative pangenome analysis of capsulated Haemophilus influenzae serotype f highlights their high genomic stability

Haemophilus influenzae is an opportunistic pathogen adapted to the human respiratory tract. Non-typeable H. influenzae are highly heterogeneous, but few studies have analysed the genomic variability of capsulated strains. This study aims to examine the genetic diversity of 37 serotype f isolates from the Netherlands, Portugal, and Spain, and to compare all capsulated genomes available on public databases. Serotype f isolates belonged to CC124 and shared few single nucleotide polymorphisms (SNPs) (n = 10,999), but a high core genome (> 80%). Three main clades were identified by the presence of 75, 60 and 41 exclusive genes for each clade, respectively. Multi-locus sequence type analysis of all capsulated genomes revealed a reduced number of clonal complexes associated with each serotype. Pangenome analysis showed a large pool of genes (n = 6360), many of which were accessory genome (n = 5323). Phylogenetic analysis revealed that serotypes a, b, and f had greater diversity. The total number of SNPs in serotype f was significantly lower than in serotypes a, b, and e (p < 0.0001), indicating low variability within the serotype f clonal complexes. Capsulated H. influenzae are genetically homogeneous, with few lineages in each serotype. Serotype f has high genetic stability regardless of time and country of isolation.

Whole genome sequencing-based classification of human-related Haemophilus species and detection of antimicrobial resistance genes

Background

Bacteria belonging to the genus Haemophilus cause a wide range of diseases in humans. Recently, H. influenzae was classified by the WHO as priority pathogen due to the wide spread of ampicillin resistant strains. However, other Haemophilus spp. are often misclassified as H. influenzae. Therefore, we established an accurate and rapid whole genome sequencing (WGS) based classification and serotyping algorithm and combined it with the detection of resistance genes.

Methods

A gene presence/absence-based classification algorithm was developed, which employs the open-source gene-detection tool SRST2 and a new classification database comprising 36 genes, including capsule loci for serotyping. These genes were identified using a comparative genome analysis of 215 strains belonging to ten human-related Haemophilus (sub)species (training dataset). The algorithm was evaluated on 1329 public short read datasets (evaluation dataset) and used to reclassify 262 clinical Haemophilus spp. isolates from 250 patients (German cohort). In addition, the presence of antibiotic resistance genes within the German dataset was evaluated with SRST2 and correlated with results of traditional phenotyping assays.

Results

The newly developed algorithm can differentiate between clinically relevant Haemophilus species including, but not limited to, H. influenzae, H. haemolyticus, and H. parainfluenzae. It can also identify putative haemin-independent H. haemolyticus strains and determine the serotype of typeable Haemophilus strains. The algorithm performed excellently in the evaluation dataset (99.6% concordance with reported species classification and 99.5% with reported serotype) and revealed several misclassifications. Additionally, 83 out of 262 (31.7%) suspected H. influenzae strains from the German cohort were in fact H. haemolyticus strains, some of which associated with mouth abscesses and lower respiratory tract infections.

Resistance genes were detected in 16 out of 262 datasets from the German cohort. Prediction of ampicillin resistance, associated with bla_TEM-1D, and tetracycline resistance, associated with tetB, correlated well with available phenotypic data.

Conclusions

Our new classification database and algorithm have the potential to improve diagnosis and surveillance of Haemophilus spp. and can easily be coupled with other public genotyping and antimicrobial resistance databases. Our data also point towards a possible pathogenic role of H. haemolyticus strains, which needs to be further investigated.

Supplementary Information

The online version contains supplementary material available at 10.1186/s13073-022-01017-x.

Learning from -omics strategies applied to uncover Haemophilus influenzae host-pathogen interactions: Current status and perspectives

Haemophilus influenzae has contributed to key bacterial genome sequencing hallmarks, as being not only the first bacterium to be genome-sequenced, but also starring the first genome-wide analysis of chromosomes directly transformed with DNA from a divergent genotype, and pioneering Tn-seq methodologies. Over the years, the phenomenal and constantly evolving development of -omic technologies applied to a whole range of biological questions of clinical relevance in the H. influenzae-host interplay, has greatly moved forward our understanding of this human-adapted pathogen, responsible for multiple acute and chronic infections of the respiratory tract. In this way, essential genes, virulence factors, pathoadaptive traits, and multi-layer gene expression regulatory networks with both genomic and epigenomic complexity levels are being elucidated. Likewise, the unstoppable increasing whole genome sequencing information underpinning H. influenzae great genomic plasticity, mainly when referring to non-capsulated strains, poses major challenges to understand the genomic basis of clinically relevant phenotypes and even more, to clearly highlight potential targets of clinical interest for diagnostic, therapeutic or vaccine development. We review here how genomic, transcriptomic, proteomic and metabolomic-based approaches are great contributors to our current understanding of the interactions between H. influenzae and the human airways, and point possible strategies to maximize their usefulness in the context of biomedical research and clinical needs on this human-adapted bacterial pathogen.

Comparative genomic analysis identifies X-factor (haemin)-independent Haemophilus haemolyticus: a formal re-classification of 'Haemophilus intermedius'

The heterogeneous and highly recombinogenic genus Haemophilus comprises several species, some of which are pathogenic to humans. All share an absolute requirement for blood-derived factors during growth. Certain species, such as the pathogen Haemophilus influenzae and the commensal Haemophilus haemolyticus, are thought to require both haemin (X-factor) and nicotinamide adenine dinucleotide (NAD, V-factor), whereas others, such as the informally classified 'Haemophilus intermedius subsp. intermedius', and Haemophilus parainfluenzae, only require V-factor. These differing growth requirements are commonly used for species differentiation, although a number of studies are now revealing issues with this approach. Here, we perform large-scale phylogenomics of 240 Haemophilus spp. genomes, including five 'H. intermedius' genomes generated in the current study, to reveal that strains of the 'H. intermedius' group are in fact haemin-independent H. haemolyticus (hiHh). Closer examination of these hiHh strains revealed that they encode an intact haemin biosynthesis pathway, unlike haemin-dependent H. haemolyticus and H. influenzae, which lack most haemin biosynthesis genes. Our results suggest that the common ancestor of modern-day H. haemolyticus and H. influenzae lost key haemin biosynthesis loci, likely as a consequence of specialized adaptation to otorhinolaryngeal and respiratory niches during their divergence from H. parainfluenzae. Genetic similarity analysis demonstrated that the haemin biosynthesis loci acquired in the hiHh lineage were likely laterally transferred from a H. parainfluenzae ancestor, and that this event probably occurred only once in hiHh. This study further challenges the validity of phenotypic methods for differentiating among Haemophilus species, and highlights the need for whole-genome sequencing for accurate characterization of species within this taxonomically challenging genus.

Genomic characterization of Haemophilus influenzae: a focus on the capsule locus

Background

Haemophilus influenzae (Hi) can cause invasive diseases such as meningitis, pneumonia, or sepsis. Typeable Hi includes six serotypes (a through f), each expressing a unique capsular polysaccharide. The capsule, encoded by the genes within the capsule locus, is a major virulence factor of typeable Hi. Non-typeable (NTHi) does not express capsule and is associated with invasive and non-invasive diseases.

Methods

A total of 395 typeable and 293 NTHi isolates were characterized by whole genome sequencing (WGS). Phylogenetic analysis and multilocus sequence typing were used to characterize the overall genetic diversity. Pair-wise comparisons were used to evaluate the capsule loci. A WGS serotyping method was developed to predict the Hi serotype. WGS serotyping results were compared to slide agglutination (SAST) or real-time PCR (rt-PCR) serotyping.

Results

Isolates of each Hi serotype clustered into one or two subclades, with each subclade being associated with a distinct sequence type (ST). NTHi isolates were genetically diverse, with seven subclades and 125 STs being detected. Regions I and III of the capsule locus were conserved among the six serotypes (≥82% nucleotide identity). In contrast, genes in Region II were less conserved, with only six gene pairs from all serotypes showing ≥56% nucleotide identity. The WGS serotyping method was 99.9% concordant with SAST and 100% concordant with rt-PCR in determining the Hi serotype.

Conclusions

Genomic analysis revealed a higher degree of genetic diversity among NTHi compared to typeable Hi. The WGS serotyping method accurately predicted the Hi capsule type and can serve as an alternative method for Hi serotyping.

Molecular Signatures of Non-typeable Haemophilus influenzae Lung Adaptation in Pediatric Chronic Lung Disease

Non-typeable Haemophilus influenzae (NTHi), an opportunistic pathogen of the upper airways of healthy children, can infect the lower airways, driving chronic lung disease. However, the molecular basis underpinning NTHi transition from a commensal to a pathogen is not clearly understood. Here, we performed comparative genomic and transcriptomic analyses of 12 paired, isogenic NTHi strains, isolated from the nasopharynx (NP) and bronchoalveolar lavage (BAL) of 11 children with chronic lung disease, to identify convergent molecular signatures associated with lung adaptation. Comparative genomic analyses of the 12 NP-BAL pairs demonstrated that five were genetically identical, with the remaining seven differing by only 1 to 3 mutations. Within-patient transcriptomic analyses identified between 2 and 58 differentially expressed genes in 8 of the 12 NP-BAL pairs, including pairs with no observable genomic changes. Whilst no convergence was observed at the gene level, functional enrichment analysis revealed significant under-representation of differentially expressed genes belonging to Coenzyme metabolism, Function unknown, Translation, ribosomal structure, and biogenesis Cluster of Orthologous Groups categories. In contrast, Carbohydrate transport and metabolism, Cell motility and secretion, Intracellular trafficking and secretion, and Energy production categories were over-represented. This observed trend amongst genetically unrelated NTHi strains provides evidence of convergent transcriptional adaptation of NTHi to pediatric airways that deserves further exploration. Understanding the pathoadaptative mechanisms that NTHi employs to infect and persist in the lower pediatric airways is essential for devising targeted diagnostics and treatments aimed at minimizing disease severity, and ultimately, preventing NTHi lung infections and subsequent chronic lung disease in children.

Insights into the population structure and pan-genome of Haemophilus influenzae

The human-restricted bacterium Haemophilus influenzae is responsible for respiratory infections in both children and adults. While colonization begins in the upper airways, it can spread throughout the respiratory tract potentially leading to invasive infections. Although the spread of H. influenzae serotype b (Hib) has been prevented by vaccination, the emergence of infections by other serotypes as well as by non-typeable isolates (NTHi) have been observed, prompting the need for novel prevention strategies. Here, we aimed to study the population structure of H. influenzae and to get some insights into its pan-genome. We studied 305H. influenzae strains, enrolling 217 publicly available genomes, as well as 88 newly sequenced H. influenzae invasive strains isolated in Portugal, spanning a 24-year period. NTHi isolates presented a core-SNP-based genetic diversity about 10-fold higher than the one observed for Hib. The analysis of key factors involved in pathogenesis, such as lipooligosaccharides, hemagglutinating pili and High Molecular Weight-adhesins, suggests that NTHi shape its virulence repertoire, either by acquisition and loss of genes or by SNP-based diversification, likely towards host immune evasion and persistence. Discreet NTHi subpopulations structures are proposed based on core-genome supported with 17 candidate genetic markers identified in the accessory genome. Additionally, this study provides two bioinformatics tools for in silico rapid identification of H. influenzae serotypes and NTHi clades previously proposed, obviating laboratory-based demanding procedures. The present study constitutes an important genomic framework that could lay way for future studies on the genetic determinants underlying invasiveness and disease and population structure of H. influenzae.

Patients with Chronic Obstructive Pulmonary Disease harbour a variation of Haemophilus species

H. haemolyticus is often misidentified as NTHi due to their close phylogenetic relationship. Differentiating between the two is important for correct identification and appropriate treatment of infective organism and to ensure any role of H. haemolyticus in disease is not being overlooked. Speciation however is not completely reliable by culture and PCR methods due to the loss of haemolysis by H. haemolyticus and the heterogeneity of NTHi. Haemophilus isolates from COPD as part of the AERIS study (ClinicalTrials - NCT01360398) were speciated by analysing sequence data for the presence of molecular markers. Further investigation into the genomic relationship was carried out using average nucleotide identity and phylogeny of allelic and genome alignments. Only 6.3% were identified as H. haemolyticus. Multiple in silico methods were able to distinguish H. haemolyticus from NTHi. However, no single gene target was found to be 100% accurate. A group of omp2 negative NTHi were observed to be phylogenetically divergent from H. haemolyticus and remaining NTHi. The presence of an atypical group from a geographically and disease limited set of isolates supports the theory that the heterogeneity of NTHi may provide a genetic continuum between NTHi and H. haemolyticus.

Pneumococcal vaccine impacts on the population genomics of non-typeable Haemophilus influenzae

The implementation of pneumococcal conjugate vaccines (PCVs) has led to a decline in vaccine-type disease. However, there is evidence that the epidemiology of non-typeable Haemophilus influenzae (NTHi) carriage and disease can be altered as a consequence of PCV introduction. We explored the epidemiological shifts in NTHi carriage using whole genome sequencing over a 5-year period that included PCV13 replacement of PCV7 in the UK’s National Immunization Programme in 2010. Between 2008/09 and 2012/13 (October to March), nasopharyngeal swabs were taken from children <5 years of age. Significantly increased carriage post-PCV13 was observed and lineage-specific associations with Streptococcus pneumoniae were seen before but not after PCV13 introduction. NTHi were characterized into 11 discrete, temporally stable lineages, congruent with current knowledge regarding the clonality of NTHi. The increased carriage could not be linked to the expansion of a particular clone and different co-carriage dynamics were seen before PCV13 implementation when NTHi co-carried with vaccine serotype pneumococci. In summary, PCV13 introduction has been shown to have an indirect effect on NTHi epidemiology and there exists both negative and positive, distinct associations between pneumococci and NTHi. This should be considered when evaluating the impacts of pneumococcal vaccine design and policy.

Clinical and Molecular Epidemiology of Childhood Invasive Nontypeable Haemophilus influenzae Disease in England and Wales

INTRODUCTION

In countries with established Haemophilus influenzae type b (Hib) immunization programs, nontypeable H. influenzae (NTHi) is now responsible for nearly all invasive H. influenzae cases across all age groups.

METHODS

Public Health England (PHE) conducts enhanced national surveillance of invasive H. influenzae disease in England and Wales. Invasive NTHi isolates submitted to Public Health England from children of ages 1 month to 10 years during 2003-2010 were characterized by multilocus sequence typing (MLST). Detailed clinical information was obtained for all laboratory-confirmed cases of invasive NTHi disease in children during 2009-2013.

RESULTS

In England and Wales, there were 7797 cases of invasive H. influenzae disease diagnosed during 2000-2013 and 1585 (20%) occurred in children aged 1 month to 10 years, where NTHi was responsible for 31-51 cases (incidence, 0.53-0.92/100,000) annually. Detailed clinical follow-up of 214 confirmed NTHi cases diagnosed in this age-group during 2009-2013 revealed that 52% (n = 111) occurred in <2-year-old and 52% (n=110) had comorbidity. Bacteremic pneumonia was the most common clinical presentation (n = 99, 46%), 16% (n = 34) required intensive care and 11% (n = 23) died. Characterization by biotyping and MLST of 316 NTHi strains from children with invasive disease during 2003-2010 revealed a genetically heterogeneous population (155 MLSTs) with diverse biotypes and no association with comorbidity status, clinical disease or outcome.

CONCLUSIONS

The high level of genetic diversity in invasive NTHi strains highlights the difficulties in developing an effective vaccine against this pathogen.

The capsule biosynthesis locus of Haemophilus influenzae shows conspicuous similarity to the corresponding locus in Haemophilus sputorum and may have been recruited from this species by horizontal gene transfer

The newly described species Haemophilus sputorum has been cultured from the upper respiratory tract of humans and appears to have little pathogenic potential. The species encodes a capsular biosynthesis locus of approximately 12  kb composed of three distinct regions. Region I and III genes, involved in export and processing of the capsular material, show high similarity to the corresponding genes in capsulate lineages of the pathogenic species Haemophilus influenzae; indeed, standard bexA and bexB PCRs for detection of capsulated strains of H. influenzae give positive results with strains of H. sputorum. Three ORFs are present in region II of the sequenced strain of H. sputorum, of which a putative phosphotransferase showed homology with corresponding genes from H. influenzae serotype c and f. Phylogenetic analysis of housekeeping genes from 24 Pasteurellaceae species showed that H. sputorum was only distantly related to H. influenzae. In contrast to H. influenzae, the capsule locus in H. sputorum is not associated with transposases or other transposable elements. Our data suggest that the capsule locus of capsulate lineages of H. influenzae may have been recruited relatively recently from the commensal species H. sputorum by horizontal gene transfer.

Classification, identification, and clinical significance of Haemophilus and Aggregatibacter species with host specificity for humans

The aim of this review is to provide a comprehensive update on the current classification and identification of Haemophilus and Aggregatibacter species with exclusive or predominant host specificity for humans. Haemophilus influenzae and some of the other Haemophilus species are commonly encountered in the clinical microbiology laboratory and demonstrate a wide range of pathogenicity, from life-threatening invasive disease to respiratory infections to a nonpathogenic, commensal lifestyle. New species of Haemophilus have been described (Haemophilus pittmaniae and Haemophilus sputorum), and the new genus Aggregatibacter was created to accommodate some former Haemophilus and Actinobacillus species (Aggregatibacter aphrophilus, Aggregatibacter segnis, and Aggregatibacter actinomycetemcomitans). Aggregatibacter species are now a dominant etiology of infective endocarditis caused by fastidious organisms (HACEK endocarditis), and A. aphrophilus has emerged as an important cause of brain abscesses. Correct identification of Haemophilus and Aggregatibacter species based on phenotypic characterization can be challenging. It has become clear that 15 to 20% of presumptive H. influenzae isolates from the respiratory tracts of healthy individuals do not belong to this species but represent nonhemolytic variants of Haemophilus haemolyticus. Due to the limited pathogenicity of H. haemolyticus, the proportion of misidentified strains may be lower in clinical samples, but even among invasive strains, a misidentification rate of 0.5 to 2% can be found. Several methods have been investigated for differentiation of H. influenzae from its less pathogenic relatives, but a simple method for reliable discrimination is not available. With the implementation of identification by matrix-assisted laser desorption ionization-time of flight mass spectrometry, the more rarely encountered species of Haemophilus and Aggregatibacter will increasingly be identified in clinical microbiology practice. However, identification of some strains will still be problematic, necessitating DNA sequencing of multiple housekeeping gene fragments or full-length 16S rRNA genes.

Characteristics of Haemophilus influenzae invasive isolates from Portugal following routine childhood vaccination against H. influenzae serotype b (2002-2010)

We aimed to characterize Haemophilus influenzae invasive isolates recovered in Portugal over a 9-year period (2002-2010) following the inclusion of H. influenzae serotype b (Hib) conjugate vaccination in the National Immunization Program (NIP) in the year 2000 and compare the results with those obtained in a similar study from the pre-vaccination era (1989-2001) previously described by us. As part of a laboratory-based passive surveillance system, 144 invasive isolates obtained in 28 Portuguese hospitals were received at the National Reference Laboratory for Bacterial Respiratory Infections and were characterized. Capsular types and antibiotic susceptibility patterns were determined. The ftsI gene encoding PBP3 was sequenced for β-lactamase-negative ampicillin-resistant (BLNAR) isolates. Genetic relatedness among isolates was examined by multilocus sequencing typing (MLST). Most isolates (77.1%) were non-capsulated, a significant increase compared to the pre-vaccination era (19.0%, p < 0.001). Serotype b strains decreased significantly (from 81.0 to 13.2%, p < 0.001) and serotype f increased significantly (from 0.8 to 6.9%, p = 0.03). Ten percent of the isolates were β-lactamase producers, a value lower than that previously observed (26.9%, p = 0.005). Eight percent of all isolates were BLNAR. A high genetic diversity among non-capsulated isolates was found. By contrast, capsulated isolates were clonal. The implementation of Hib vaccination has resulted in a significant decline in the proportion of serotype b H. influenzae invasive disease isolates. Most episodes of invasive disease occurring in Portugal are now due to fully susceptible, highly diverse, non-capsulated strains. Given the evolving dynamics of this pathogen and the increase in non-type b capsulated isolates, continuous surveillance is needed.

Hierarchical and spatially explicit clustering of DNA sequences with BAPS software

Phylogeographical analyses have become commonplace for a myriad of organisms with the advent of cheap DNA sequencing technologies. Bayesian model-based clustering is a powerful tool for detecting important patterns in such data and can be used to decipher even quite subtle signals of systematic differences in molecular variation. Here, we introduce two upgrades to the Bayesian Analysis of Population Structure (BAPS) software, which enable 1) spatially explicit modeling of variation in DNA sequences and 2) hierarchical clustering of DNA sequence data to reveal nested genetic population structures. We provide a direct interface to map the results from spatial clustering with Google Maps using the portal http://www.spatialepidemiology.net/ and illustrate this approach using sequence data from Borrelia burgdorferi. The usefulness of hierarchical clustering is demonstrated through an analysis of the metapopulation structure within a bacterial population experiencing a high level of local horizontal gene transfer. The tools that are introduced are freely available at http://www.helsinki.fi/bsg/software/BAPS/.