Morning session:
Using bacterial genomic data to inform empirical antimicrobial treatment of shigellosis in England
Lewis Mason, Department of Genetics
Shigellosis is a gastrointestinal disease caused by species of Shigella bacteria. Shigella spp. spread by consumption of food/water contaminated with faeces. In higher-income nations, shigellosis typically occurs in international travellers or as a sexually transmissible illness among men-who-have-sex-with-men (MSM). Shigellosis is treated with antibiotics: ciprofloxacin, ceftriaxone, azithromycin, or pivmecillinam. Genomic data from the two transmission pathways indicate that pathways have unique AMR profiles. We gathered responses from 23 hospitals on their shigellosis treatment guidelines. We used genomic data to investigate how AMR genes in Shigella correlate to its species; patient travel location, sex, and age. We found 14/23 hospital trusts (61%) had guidance treating shigellosis. Of these, seven use ciprofloxacin, two use ceftriaxone, and one uses azithromycin. From genomic data, we found that > 40% of Shigella bacteria from MSM were resistant to ciprofloxacin and azithromycin between 2018-2020. >50% were resistant to ciprofloxacin (travel to Asia). >30% were resistant to ceftriaxone (travel to Africa). Categorised by species: azithromycin resistance was in 66% of S. sonnei and 3% in S. flexneri. We show that knowing the species of Shigella and patient demographics is useful in predicting which antibiotics are likely to be effective prior to receiving antimicrobial susceptibility testing results.
Modelling the impact of Universal Varicella Vaccination in England
Lauren Adams, Disease Dynamics Unit - Department of Veterinary Medicine
Universal varicella vaccination has been introduced in many countries since the vaccine was developed. Exogenous boosting suggests that exposure to varicella boosts an individual’s immunity preventing reactivation of the virus. Consequently, removal of naturally occurring varicella through a universal programme may cause an increase in shingles cases. We aim to use CDC data on shingles incidence to better understand exogenous boosting, apply this to a UK transmission model and determine the impact of universal varicella vaccination on varicella transmission.
We fit an age structured transmission model to US incidence data for varicella and zoster then calibrate it to annual incidence data to understand how zoster incidence changed following vaccination and determine a duration of protection due to boosting. We then apply this to a UK model to assess the impact of vaccination on varicella and zoster. Finally, we use the model results to inform a cost-effective analysis.
We find that exogenous boosting provides just 3 years protection against zoster, which is significantly less than what previously models have used. We find that a small increase in zoster cases is expected following the introduction of universal varicella vaccination, but this does not offset benefits gained from the reduction in varicella cases.
Acute response to pathogens in the early human placenta at single cell resolution
Elias Rafael Ruiz-Morales, Vento Lab, Wellcome Sanger Institute
The placenta acts as a maternal-fetal barrier that provides nourishment and protection from infections. However, certain pathogens can attach to and even cross the placenta, causing pregnancy complications with potential lifelong impacts on the child’s health. Here, we profiled at the single-cell level the placental responses to three pathogens associated with intrauterine complications - Plasmodium falciparum, Listeria monocytogenes and Toxoplasma gondii. Upon encountering the pathogens, we observed that all placental lineages initiate inflammatory responses, that may compromise placental function. Additionally, we investigated the responses of fetal macrophages, known as Hofbauer cells to each pathogen, and propose that they are the probable niche for T. gondii. Lastly, we unveiled the adaptive strategies of P. falciparum in the placental microenvironment, characterized by the modulation of protein export into the host erythrocyte and nutrient uptake pathways. In summary, we have delineated the cellular networks and signaling pathways that mediate acute placental inflammatory responses, which may contribute to complications during pregnancy.
Understanding the influence of therapeutic interventions on the ecology of the microbiota in the CF airways
Pok-Man Ho, Welch Group, Department of Biochemistry
Although we have multiple methods to catalogue microorganisms in different ecosystems, little is known about how the individual species interact with one another. Previous efforts in this regard have focused on pairwise interactions. However, there is so far no consensus on what ecological role(s) any given microorganism plays; its “profession” (niche) in the ecosystem. Defining the niche is crucial, because the microbiota are a risk factor in many chronic infection scenarios. With the ever-expanding threat of antimicrobial resistance, this information on “what each organism does” in the ecosystem is more important than ever, especially given the realisation that not all members of the microbial community promote disease; some appear to prevent it. Here we introduce our computational model based on the generalised Lotka-Volterra equation and implementation of a Bayesian simulation framework. By using time-series data of cystic fibrosis airway infections, our machine learning approach identified distinct ecological signals associated with different therapeutic interventions. Our results are consistent with the notion of “no consensus” in pairwise microbial interactions, and provide a systematic and highly-reproducible method to quantify microbial interactions. Although our approach does not provide mechanistic explanations for the observed interactions between species, it does enable testable hypotheses to be investigated.
Unraveling incongruent dengue infection intensities inferred from serological and case surveillance: roles of antibody kinetics, assay noise, and heterogeneous infection risk
Angkana Huang, Pathogen Dynamics Group, Department of Genetics
Uncovering rates at which susceptible individuals became infected, i.e. the Force of Infection (FOI), is essential for assessing susceptibility of populations and the study of correlates to predict future risks. For dengue, measuring FOI is of particular significance, as prior exposure is a strong risk factor for severe disease. Longitudinal serology are considered gold standard measurements, as they directly track the transition of seronegative individuals to seropositive (seroincidence). Cross-sectional serology can provide estimates of FOI by contrasting seroprevalence across ages. Age of reported cases, linked to waiting time to infection, can also be used to infer FOI.
Using 26 years of data from cohort studies and hospital-attended cases from Kamphaeng Phet province, Thailand, we found FOI estimates from the three sources to be highly inconsistent. Annual FOI estimates from seroincidence was 1.96 to 3.78-times higher than case-derived FOI. Correlation between seroprevalence-derived and case-derived FOI was poor (correlation coefficient=0.19) despite no systematic bias. Through extensive simulations and theoretical analysis, we show that the incongruences were results of the interplay between dengue antibody kinetics, assay noise, and heterogeneity in FOI across ages. Extending standard inference models to include these processes reconciled the FOI and susceptibility estimates.
Afternoon session:
Can you catch cancer? Can cancer catch you?
Kevin Gori, Transmissible Cancer Group, Department of Veterinary Medicine
To date, numerous infectious cancers have been discovered in the wild. The oldest known, Canine Transmissible Venereal Tumour (CTVT), is a disease that infects primarily free roaming dogs.
Originating as a tumour in an arctic sled dog over six thousand years ago, for millennia CTVT has spread among dog populations as a contagious allograft. But the transmission has not only been in one direction. We have recently identified the genetic signature of an ancient dog embedded in a series of modern tumour samples.
Using genomic sequence analysis and population genetics, we trace the source of this signal to a highly rearranged fragment of DNA that was incorporated by CTVT from a host dog that lived in the Middle East over two thousand years ago.
Learning the fitness dynamics of pathogens from phylogenies
Noemie Lefrancq, Salje/Parkhill group, Department of Genetics/Veterinary Medicine
The dynamics of pathogen genetic diversity, including the emergence of lineages with increased fitness, is a foundational concept of disease ecology with key public health implications. However, the identification of distinct lineages and estimation of associated fitness remain challenging, and is rarely done outside densely sampled systems. Here, we present a novel scalable framework that summarises changes in population composition in phylogenetic trees, allowing for the automatic detection of lineages based on shared fitness and evolutionary relationships. We present our approach on a broad set of viruses and bacteria (SARS-CoV-2, H3N2 influenza, Bordetella pertussis and Mycobacterium tuberculosis), which includes both well-studied and understudied threats to human health. We show our framework recovers the main known circulating lineages for each pathogen, as well as specific amino acid changes linked to fitness changes. Additionally, our framework showcases its potential for uncovering previously undetected lineages with increased fitness, including three co-circulating Bordetella pertussis lineages. Our findings are robust to uneven and limited observation. This widely applicable framework provides an avenue to monitor evolution in real-time to support public health action and explore fundamental drivers of pathogen fitness.
Using evolutionary information in gene finding and annotation in microbes
Zachary Ardern, Pathogens and Microbes, Wellcome Sanger Institute
Evolutionary relationships between genetic sequences provide a lot of useful information for genome annotation, but much of this information is not currently used. When used, evolutionary information is often included in an ad hoc way to follow up hypotheses rather than systematically. This talk will review methods and some new results from finding new genes in bacteria and viruses using sequence constraint and assigning functions to them using evolutionary methods. In particular many genes have been missed or incorrectly called by standard methods; particularly genes which are short or overlapping known genes in alternative reading frames. Massive scale genome sequencing data and improved methods in understanding molecular evolution can together be leveraged for improved functional genome annotation, particularly when suitable genome sequence sampling strategies are used.
The biofilm matrix of Pseudomonas aeruginosa is a distinct extracellular compartment which selectively traps key secreted proteins
Rahan Nazeer, Welch Group, Department of Biochemistry
Pseudomonas aeruginosa (PA) is both a ‘professional secretor’ and a prolific biofilm former. Consequently, many secreted factors need to pass through the biofilm matrix. This raises the question of whether any of these secretions are selectively trapped in the matrix, forming a distinct biofilm “matrixome”.
To address this, we developed a method to separate biofilm cells from the matrix and from the culture supernatant, yielding clearly-differentiated “exo-compartments” for proteomic analysis. This revealed that the matrix does selectively trap secreted factors, while allowing others to pass through into the “secretome”. One of the most abundant matrix-enriched proteins was an uncharacterised protein, PA2668. Encoded adjacent to PA2668 is an uncharacterised Type II Secretion system (T2SS), designated as the hpl cluster. Given that the virulent substrates of many T2SS are also encoded adjacent to their cognate secretion machinery, it seems likely that PA2668 is exported through the Hpl system. To investigate this further, we have purified PA2668 to raise antibodies and begin detailed functional characterisation. The hpl genes, highly conserved across PA isolates, have been deleted for proteomic definition of the Hpl secretome. LacZ fusions and DNA pulldowns are being used to assess how expression of the Hpl machinery and PA2668 are regulated.
Convergent evolution of Enterobacteriaceae in epidemiological networks with high antimicrobial use
Yi Ling Tam, Baker group, Department of Genetics
One of the major research focuses concerning antimicrobial resistance is to understand the evolutionary prerequisites that precede the emergence of resistance. Previous studies investigating antimicrobial resistance in multiple Shigella epidemics, for example those circulating the “Men who have sex with men” (MSM) community in the UK, showed that repeated horizontal transfer of plasmids carrying resistance genes among Shigella facilitated and enhanced resistance. Here, we aim at identitying non-AMR changes in the Shigella sonnei’s (S.sonnei) genomes that act as preliminary steps, or “stepping stones”, preceding the development of antibiotic resistance in the MSM community. Through genomic and phylogenetic analyses of 3,745 S.sonnei isolates obtained through routine UK surveillance, we identified a gene candidate methionine-tRNA synthetase (metG) that is related to antibiotic tolerance, which has been suggested as a mechanism for the development of resistance. Moreover, we also observed duplication signal of metG in some samples By further investigation, we hope this could help with identifying early warning signs of resistance development and targeting public health interventions.