CID Annual Symposium 2022 Programme: Posters | Cambridge Infectious Diseases

Morning Poster Session (10.10-11.00h)

Detection of Respiratory Pathogens and Antimicrobial Resistance Genes by High Throughput Nanofluidic PCR in Bronchoalveolar Lavage (BAL) Samples of ICU Patients

Leanne Kermack, Department of Medicine

The burden of respiratory infections is a significant global concern, causing 4 million deaths annually. Diagnosing these infections is challenging as patient symptoms are not always pathogen-specific. Furthermore, culture as the ‘gold standard’ has limited sensitivity and alongside antimicrobial sensitivity testing is time-consuming. Molecular approaches such as the TaqMan Array Card (TAC) represent a faster and more sensitive alternative. However, these are higher cost, require larger sample volumes, and cover a limited breadth of organisms. To address these shortcomings, this study developed a HT-qPCR respiratory panel able to detect 93 respiratory organisms (viruses, bacteria and fungi) and antimicrobial resistance genes in 96 samples simultaneously.

BAL samples (previously analysed by TAC and culture) were examined using this panel to compare results across the three methods. There was a correlation between detection via TAC and HT-qPCR panel for the majority of targets. 76.5% of targets were detected by both culture and the HT-qPCR panel. Of these, 4.7% were not also detected by TAC. In terms of concordance with culture, the difference between TAC and HT-qPCR was not statistically significant (p=0.28).

This HT-qPCR panel could be customised to detect both existing and emerging respiratory pathogens, supporting clinicians to diagnose and improve patient outcomes.

Engineered MATE multidrug transporters reveal two functionally distinct ion-coupling pathways in NorM from Vibrio cholerae

Boyan Bai, Department of Pharmacology

Multidrug and toxic compound extrusion (MATE) transport proteins confer multidrug resistance on pathogenic microorganisms and affect pharmacokinetics in mammals. Our understanding of how MATE transporters work has mostly relied on protein structures and MD simulations. However, the energetics of drug transport has not been studied in detail. Many MATE transporters utilise the electrochemical H + or Na + gradient to drive substrate efﬂux, but NorM-VC from Vibrio cholerae can utilise both forms of metabolic energy. To dissect the localisation and organisation of H + and Na + translocation pathways in NorM-VC we engineered chimaeric proteins in which the N-lobe of H + -coupled NorM-PS from Pseudomonas stutzeri is fused to the C-lobe of NorM-VC, and vice versa. Our ﬁndings in drug binding and transport experiments with chimaeric, mutant and wildtype transporters highlight the versatile nature of energy coupling in NorM-VC, which enables adaptation to ﬂuctuating salinity levels in the natural habitat of V. cholerae. (Communications Biology 4(1): 558, doi:10.1038/s42003-021-02081-6: Sagar Raturi, Asha V. Nair, Keiko Shinoda, Himansha Singh, Boyan Bai, Satoshi Murakami, Hideaki Fujitani, and Hendrik W. van Veen (Corresponding author))

Energetics of lipid transport by the ABC transporter MsbA is lipid dependent

Yakun Tang, Department of Pharmacology

The ABC multidrug exporter MsbA mediates the translocation of lipopolysaccharides and phospholipids across the plasma membrane in Gram-negative bacteria. Although MsbA is structurally well characterised, the energetic requirements of lipid transport remain unknown. Here, we report that similar to the transport of small-molecule antibiotics and cytotoxic agents, the flopping of physiologically relevant long-acyl-chain 1,2-dioleoyl (C18)-phosphatidylethanolamine in proteoliposomes requires the simultaneous input of ATP binding and hydrolysis and the chemical proton gradient as sources of metabolic energy. In contrast, the flopping of the large hexa-acylated (C12-C14) Lipid-A anchor of lipopolysaccharides is only ATP dependent. This study demonstrates that the energetics of lipid transport by MsbA is lipid dependent. As our mutational analyses indicate lipid and drug transport via the central binding chamber in MsbA, the lipid availability in the membrane can affect the drug transport activity and vice versa.
(doi: 10.1038/s42003-021-02902-8. Authors: Dawei Guo, Himansha Singh, Atsushi Shimoyama, Charlotte Guffick, Yakun Tang, Sam M Rowe, Timothy Noel, David R Spring, Koichi Fukase, Hendrik W van Veen)"

How to reduce West Nile infection risk for humans? Mathematical modelling to predict the effectiveness of intervention strategies

Elisa Fesce, Department of Veterinary Medicine

West Nile disease (WND) is a vector-borne disease that involves mosquitoes as vectors, and several bird species as definitive hosts. Humans are susceptible to WND and develop symptoms that can progress even to severe neurological disease. This, and the widespread of the disease, make WND of public concern in several countries in Europe, America and Asia. Prevention from mosquito bites is currently the only strategy available to protect humans from infection, but only a few studies have been performed to compare the effectiveness in reducing the number of infectious mosquitoes of different intervention strategies.
We then propose a mathematical model to compare the efficacy of six possible intervention strategies. The model was calibrated with data coming from the WND surveillance plan provided by Regione Lombardia and Istituto Zooprofilattico Sperimentale della Lombardia e dell’Emilia Romagna (IZSLER) (northern Italy), and six different intervention strategies were included to assess and compare their efficacy in reducing the number of infectious mosquitoes.
Results obtained showed that the efficacy of the intervention strategy depends on both the chosen type of intervention and its intensity. The elimination of larvae and eggs or adult mosquitoes showed the best efficacy, while the other interventions tested showed lower efficacy.

Whole-genome sequence analysis of Salmonella typhi provides evidence of phylogenetic linkage between cases of typhoid fever in Santiago, Chile in the 1980s and 2010-2016

Mailis Maes, Department of Medicine

In Santiago, Chile, the Typhoid Fever Control Program used vaccination and wastewater management to reduce the S. typhi incidence from >200/100,000 in the 1980s to <1/100,000 in the 2010s. We wanted to understand how the population of S. typhi has changed over this time, and what proportion of recent isolates represent “historic” 1980s genotypes versus “new introduction” genotypes. Whole-genome sequencing was performed on two collections of isolates: 113 from the 1980’s, and 99 from the 2010’s. With the exception of three separate introductions of H58 into the country, the overall distribution of genotypes was similar between the 1980s isolates and the recent (2011-2016) isolates. Both modern and historical isolates include a higher proportion of uncommon ancient genotypes (1.1 and 1.2.1) than found globally. In three 1.2.1 isolates (1980s carrier, 1980s case, 2013 isolate), we detected the same novel pHCM2 plasmid. Thus, among the few recent Santiago typhoid isolates are indications of transmission of historic genotypes, perhaps from chronic carriers. Sequence data indicates that 30 years after control of typhoid in Santiago, among the rare sporadic cases that still occur, a notable proportion are 1980s genotypes.

Rapid Detection of Bacteria Directly from Clinical Samples

Daniel Buhl, Department of Veterinary Medicine

Antimicrobial-resistant bacteria pose a major threat to the healthcare and agricultural system. Problems with appropriate treatment in these settings often arise through the unavailability of fast identification of the disease-causing species. Current methods for bacterial identification require culturing and therefore take at least a day or longer. This not only leads to a delay in treatment and the potential for resistance to develop but also results in severe losses in sectors such as dairy farming.
We established three rapid methods for the detection of mastitis pathogens directly from bovine milk without the need to culture the pathogen beforehand. Using novel mass spectrometry methods such as laser-assisted Rapid Evaporative Ionisation Mass Spectrometry (LA-REIMS) we were able to detect specific biomarkers in as little as 30 minutes. Interestingly, these molecules were found to originate from a pathogen-specific host-immune response rather than the bacteria themselves. We also combined portable PCR techniques using highly specific primers with a printed array strip for detection of major mastitis pathogens within 120 minutes. Furthermore, we constructed a whole-cell bacterial biosensor allowing for visual detection of pathogens without advanced equipment in less than 300 minutes, which also has potential as a targeted drug delivery system.

Effects of the G-quadruplex-binding drugs Quarfloxin and CX-5461 on the malaria parasite Plasmodium falciparum

Dr Holly Craven, Department of Pathology

The apicomplexan parasite Plasmodium falciparum causes the most devastating form of the malarial disease in humans. P. falciparum has an A/T-biased genome with few G-quadruplex (G4) forming sequences; nevertheless, G4s exist, with biological roles at DNA/RNA levels. The existence of these G4s presents the potential for repurposing of G4-binding anti-cancer agents as novel antimalarials. New antimalarials are urgently required because drug resistance commonly develops, including the current frontline artemisinin-based regimens. We previously showed that the G4-binding agent quarfloxin (QF) is potent and fast-acting against cultured P. falciparum. However, the mechanistic pathway appears to differ from that observed in humans, implying a valuable parasite-specific mode of killing. Here, we set out to further characterise this process for QF and a related drug CX-5461. QF was found to localise throughout the parasite, not in a specific organelle. Efforts at developing QF resistant parasites showed only mild tolerance can be acquired and is not heritable. We found that both QF and CX-5461 can cause DNA damage signatures in P. falciparum. Interestingly, while DNA damage was more severe with CX-5461, QF was the more cytotoxic drug. Therefore, the two modes of action may not be identical, and not limited to the induction of DNA damage.

Dissonance of temporal changes in dengue infection intensity inferred between serological studies, and passive case surveillance

Dr Angkana T. Huang, Department of Genetics

Force of infection (FOI) can be inferred from multiple data sources. Data from longitudinal serological studies are considered gold standards as they directly measure rates at which naive individuals seroconvert. Through introducing assumptions on gain and sustainment of seropositivity, cross-sectional serological data including individuals of variable age can draw information from differences in risk experienced across birth cohorts to provide estimates of FOI. Adding models to link between infection and generation of cases, FOI can be inferred from age-stratified case counts. Agreement between these estimates, however, are seldom assessed due to rarity of longitudinal serological data. Leveraging data from dengue cohort studies in Kamphaeng Phet province, Thailand (1998-2016) and case data from the provincial hospital, we explored relationships between the estimates. We found considerable concordance between seroconversion rates in cohort subjects and FOI inferred from hospital case counts (correlation coefficient (r)=0.53) but poor agreement with FOI inferred from serological cross-sections of the cohort study through simple serocatalytic models (r=0.07). Accounting for antibody kinetics, specifically incorporating nondurable detectability of monotypic responses and including only blood drawn pre-dengue season improved the agreement (r=0.25). Further investigations are underway to understand the processes underlying these discordances and their implications.

The role of antigenic and genetic diversity in driving the infection and disease risk of dengue virus

Dr Lin Wang, Department of Genetics

The infection and disease risks from dengue virus (DENV) are not well understood. Multiple competing factors drive DENV’s infection and disease processes, including the viral antigenic and genetic evolution and effects of pre-existing immunity. Combining case-based surveillance with viral surveillance might be useful to assess DENV’s infection and disease risks in the community.

We analyzed 18,236 laboratory-confirmed secondary DENV cases collected from 1990 to 2014 by a Bangkok children hospital. Infecting year and age were known for each case, with infecting serotype available for 51.7% of cases. To capture antigenic diversity within the population, we build a three-dimensional antigenic map using data from a large neutralization testing program. We further increase the resolution of population antigenic profile in any year by placing additional sequenced viruses onto the map, based on genetic similarity.

We developed a series of Bayesian models, progressively incorporating year, age, serotype, and antigenic information. We estimated the yearly serotype-specific force of infection, cross-protection effect, and importantly, the infection and disease risks over antigenic distance. We reconstructed age- and serotype-specific distributions of cases per year. Our study provides a unified framework to integrate case-based and viral surveillance data for assessing DENV’s infection and disease risks in the community.

Elucidating the spatiotemporal dynamics of Streptococcus pneumoniae using human mobility and genetic data.

Sophie Belman, Department of Genetics

Background: Streptococcus pneumoniae is a globally endemic, human obligate bacteria, and a leading cause of pneumonia and meningitis worldwide. Despite the enormous public health burden, the extent and mechanisms of pneumococcal spread across spatial scales remain largely unknown.
Methods: We analyze 6910 geolocated pneumococcal genomes collected from 2000 to 2014 in South Africa. We develop mechanistic phylogeographic models to reconstruct pathogen spread, incorporating the generation time distribution, human mobility data provided by Facebook, population size, and underlying heterogeneities in sampling.
Results: We found sequence pairs from the same genotype and province had 4.7[3.4-7.4] times the probability of having an MRCA within 3 years compared to sequence pairs from >1000km apart. We were able to accurately recover this observed pattern of spatial spread using our mechanistic model estimating that most (77%) of transmission events were local, with occasional long-distance transmissions. Our framework demonstrated the key role of travel to large cities in disease spread. After one year of sequential person-to-person transmission, the pneumococcus was 20 times more likely to have travelled to a major population hub than elsewhere.
Conclusion: Pneumococcus spreads slowly across South Africa, driven by limited long-distance human mobility and the long generation time of pneumococcus. Our framework provides an opportunity to explore whether similar patterns of spread are observed elsewhere with differing mobility patterns and the impact of vaccine introduction.

A new parameter for modelling the evolutionary pressure of vaccines

Maria A. Gutierrez, Department of Applied Mathematics and Theoretical Physics

While vaccination campaigns have played a crucial role in reducing incidence and deaths in the COVID-19 pandemic, we have been substantially challenged by the appearance of new viral variants, especially those that evade existing immunity.

Using a simple mathematical model for the epidemic dynamics, the research I will present at this flash talk discusses ideas on how vaccination, together with natural immunity, could exert pressure on viral evolution towards antigenic escape.

I will introduce a parameter to describe the within-host effect of vaccines on the appearance of variants. Building on the “Phylodynamics” framework of Grenfell et al 2004, I will explain how different immunological aspects might contribute to this parameter, how it may be related to known vaccine efficacy (VE) parameters and its potential implications for the population-level dynamics and optimal vaccine allocation.

Although the ideas that I present here have been motivated by the challenges of COVID-19, they may be applicable to other infectious diseases.

The matrix reloaded: biophysical and biochemical analysis of extracellular matrix components in Pseudomonas aeruginosa biofilms

Rahan Nazeer, Department of Biochemistry

Pseudomonas aeruginosa (PA) is an opportunistic human pathogen responsible for high morbidity and mortality in individuals predisposed towards infection, such as the immunocompromised, or people with cystic fibrosis (CF). PA-caused chronic infections in people with CF is associated with biofilm formation. However, the sessile lifestyle remains poorly understood. We know that the biofilm matrix – the “glue” holding biofilm cells together – confers tolerance to numerous external stresses such as antibiotics and immune factors, and also provides physical resistance to shear stress.

We now know that proteins, polysaccharides, lipids and nucleic acids constitute the biofilm matrix, and that each can play important roles in holding the biofilm together. However, the precise contribution of each matrix component to the biofilm’s mechanical stability remains unquantified. To interrogate this, we have made defined deletions in multiple biofilm-associated genes. Using these, we have been investigating how each gene product contributes towards the “stickiness” of the matrix. This is being done by measuring the force required to detach cells from the mutant biofilms using a state-of-the-art optical tweezers setup. In parallel, we have been examining how these genes contribute towards the “matrixome” of the biofilm (i.e., that complement of secretions spatially localised in the matrix compartment).

Molecular characterisation and diagnostic development for Oropouche fever, a neglected tropical viral disease

Dr. Monique Merchant, Department of Pathology

Oropouche fever is a neglected tropical viral disease endemic to Latin America and the Caribbean with the potential to cause future pandemics. Oropouche fever is caused by Oropouche virus (OROV), an orthobunyavirus with a tri-segmented (-)ssRNA genome and a propensity for genetic reassortment. OROV commonly infects sloths, birds, monkeys and rodents but has been known to spill over into human populations via transmission from biting insects.

After Dengue fever, Oropouche fever is the second-most prevalent insect-borne disease in Brazil. Over 500,000 infections have been documented as the result of numerous regional epidemics in South America since its discovery in 1955. Changes in climate and land use are increasingly bringing humans into contact with OROV reservoir and vector species, but currently, there is no readily available point-of-care diagnostic test for OROV infection.

By producing high-quality OROV antigens and anti-OROV nanobodies we aim to (1) interrogate the immune response to OROV infection, (2) develop cost-effective, in-field diagnostics, and (3) probe the use of recombinant OROV surface antigens as vaccine candidates. Preliminary progress on OROV antigen generation and characterisation will be discussed.

Afternoon Poster Session (14.45 - 15.30h)

Novel nano-litre high-throughput qPCR (HT-qPCR) to measure the abundance and diversity of antimicrobial resistance genes in human, animal and environmental samples

Avijit Dutta, Department of Medicine

Antimicrobial resistance (AMR) is one of the leading public health challenges causing 1.27 million deaths annually. One Health perspective studies AMR at the human-animal-environment interface. The majority of One Health AMR studies focus on phenotypic characterization or molecular detection of a limited number of antimicrobial resistance genes (ARGs) using conventional PCR. Here, we employed a novel nano-litre HT-qPCR approach using the Fluidigm Biomark HD to measure the abundance and diversity of 239 ARGs in human stool, animal scat, soil and water samples from informal settlements in Fiji. The ARG richness was calculated in both absolute (gene copies) and normalized (relative to 16S rRNA) means and the diversity of ARGs was studied using discriminant analysis of principal component (DAPC). Human stool samples harboured a significantly higher average ARG number (53.7±12.05, p<0.05), and abundance (log-transformed gene copy number 5.9±1.15, p< 0.0001) than other sample types. DPCA analysis indicated that the resistome of surface water samples was more closely related to human samples than to potable water, soil or animal samples. Our findings indicate that to contain AMR in low-income settings, critical consideration should be given to the efficient management of human excreta and water supply.

Drug-dependent inhibition of nucleotide hydrolysis in the heterodimeric ABC multidrug transporter PatAB from Streptococcus pneumoniae

Charlotte Guffick and Pei-Yu (Kelly) Hsieh, Department of Pharmacology

The bacterial ATP-binding cassette (ABC) multidrug exporter PatAB has a critical role in conferring antibiotic resistance in multidrug-resistant infections by Streptococcus pneumoniae. As with other heterodimeric ABC exporters, PatAB contains two transmembrane domains, that form a drug translocation pathway, and two nucleotide-binding domains, that bind ATP with only one site at which hydrolysis occurs. Here, we used mass spectrometry techniques to determine the subunit stoichiometry and investigate locations of drug binding in PatAB using drug mimetic azido-ethidium. Surprisingly, our analyses of labelled PatAB point to ethidium binding at the PatA nucleotide-binding domain, with the azido moiety crosslinked to residue Q521 in the H-like loop of the degenerate nucleotide binding site. Investigation into this compound and residue’s role in nucleotide hydrolysis pointed to a reduction in the activity of Q521A mutant and ethidium-dependent inhibition of hydrolysis for active proteins. While most transported drugs did not alter nucleotide hydrolysis in our reconstitution systems, further examples of ethidium-like inhibition were found with propidium, novobiocin and coumermycin A1, all through a non-competitive mechanism. These data cast light on potential mechanisms by which drugs can regulate nucleotide hydrolysis by PatAB, which may involve a novel drug binding site near the nucleotide-binding domains.

(doi:10.1111/febs.16366: Charlotte Guffick , Pei-Yu Hsieh, Anam Ali , Wilma Shi , Julie Howard , Dinesh K. Chinthapalli , Alex C. Kong, Ihsene Salaa, Lucy I. Crouch, Megan R. Ansbro, Shoshanna C. Isaacson, Himansha Singh, Nelson P. Barrera, Asha V. Nair, Carol V. Robinson, Michael J. Deery, and Hendrik W. van Veen (Corresponding author))."

Molecular mechanisms underlying the substrate transport by Major facilitator superfamily multidrug transporter LmrP

Sana Khalid, Department of Pharmacology

Multidrug resistant pathogenic bacteria are a major health concern globally. One important mechanism of drug resistance is based on the extrusion of antibiotics out of the cell by membrane proteins called multidrug transporters. LmrP is a secondary-active Major Facilitator Superfamily (MFS) multidrug transporter from a gram-positive bacterium. It can mediate the efflux of at least 22 clinically relevant antibiotics in a protonmotive force-dependent fashion by catalysing electrogenic antibiotic / proton antiport. In this mechanism, the catalytic carboxylates D142 and D235/E327 on the surface of the substrate-binding chamber are crucial, but their role in alternating access of the chamber is unknown. We used wildtype LmrP and carboxylate mutants (D142N, D235N and E327Q) in cysteine accessibility studies and total protein fluorescence assays. These labelling and fluorescence experiments were done at pH 6.0 and 8.0 which stabilise protonated and deprotonated states of LmrP, respectively. Protonation of D142, mimicked by the D142N mutation, stabilises LmrP in an inward-facing state compared to wildtype LmrP for which a mixed population of inward- and outward-facing conformations is present. Our results suggest that protonation and deprotonation of the catalytic carboxylates facilitate the switch between conformational states of LmrP that underly alternating access of the antibiotic binding chamber in the transporter.

Using optical tweezers to explore the molecular basis of attachment between Plasmodium falciparum merozoites and human erythrocytes

Emma Jones, Cambridge Institute for Medical Research

For Plasmodium falciparum to multiply, the parasites must invade human erythrocytes. Invasion is a complex process involving multiple receptor-ligand interactions. Invasion begins with attachment of the parasite to the erythrocyte, followed by reorientation to position the parasite to invade, then strong membrane wrapping occurs, finally, a tight junction is formed which passes around the merozoite forming the parasitophorous vacuole. Optical tweezers can be used to directly manipulate recently egressed merozoites and erythrocytes to quantify the strength of attachment between the cells, and the frequency with which such attachments occur. Multiple P. falciparum proteins are hypothesised to be involved in this step, but their precise role in attachment strength has not been explored at an individual cellular level. By using a range of inhibitors, antibodies and genetically modified P. falciparum strains with deletions in individual invasion-associated genes, we have quantitated the contribution of individual proteins to the strength of attachment. Initial results indicate that whilst disruption of most stages of invasion reduces the likelihood of attachment, so far only inhibitors of, or deletions in, members of the Erythrocyte Binding Like protein (EBA) family and the reticulocyte binding-like (RH) family of proteins, significantly affects the strength of attachment between merozoites and erythrocytes.

A review of the gaps and biases in data that represent spillover pathways of bat-borne viruses

Imogen Lindsley, Department of Veterinary Medicine

In order to correctly assess the frequency and route of spillover, the biases in the data need to be accounted for. A variety of publicly available databases do not account for all of these biases. In order to draw precise conclusions on the frequency of cross-species transmission events in nature, the biases in the data need to be re-assessed. My project assesses the gaps and sampling biases in the data that is being used to demonstrate the routes of transmission involved in the spillover of bat-borne viruses.

Contribution of Transposable Elements to the Genome of Schistosoma Mansoni

Toby Brann, Department of Pathology

Schistosomiasis, or Bilharzia, is a parasitic disease caused by trematodes of the Schistosoma genus. It is estimated that 250 million individuals worldwide suffer from schistosomiasis, with one billion people living in at-risk areas with a total loss of 1.9 million disability-adjusted life years. Transposable elements comprise approximately 40% of the Schistosoma mansoni genome and are thought to have some degree of functional contribution. These elements are repetitive, genetic entities that are potentially capable of mobilisation across an organism’s genome. In doing so, these elements may interrupt genes, provide regulatory sequences for expression, perpetuate recombination and increase or decrease an organism’s fitness. For example, in Drosophila, transposable elements have been demonstrated as contributing to adaptability and increased ability to live at altitude. By exploring how transposable elements affect the coding component of the Schistosoma mansoni genome we may better understand the transposable element contribution to loci relevant to pathology and adaptation, in addition to the evolutionary history and selective pressures of these elements. This work seeks to explore the transposable element content of schistosomes, in particular, Schistosoma mansoni and, specifically, the contribution of these elements to protein-coding sequences, genes and Schistosoma biology.

Trivalent MVA-based vaccine candidate against Ebola, Lassa, and Marburg virus elicits a robust humoral immune response against all three viruses

Nina Krause, Department of Veterinary Medicine

Ebola, Lassa, and Marburg virus cause severe diseases often with fatal outcomes and, preventative measures are limited. Considering production costs and delivery to people with limited access to health care, a trivalent vaccine is preferable. However, multivalent designs are associated with the concern of reduced immunogenicity of the individual antigens.
We designed a trivalent MVA-based vaccine candidate to elicit protection against all three viruses and evaluated its immunogenicity in Balb/c mice. We compared the effect of administrating each antigen alone, the three antigens as a mix of monovalent constructs, and the trivalent constructs.
There were no significant differences in the antigen-specific binding antibody titres of animals that received the trivalent construct versus animals receiving the respective antigens alone. These results were reflected in the pseudovirus neutralisation assays, where overall neutralising responses were modest. Differences between the magnitude of the T cell response towards the Lassa antigen versus the Marburg and Ebola antigen were observed.
We showed that a multivalent vaccine candidate induces a robust antibody response against several antigens. Cellular immunity appears to be more affected by a multivalent approach. Future studies aim to investigate whether this single vaccine yet protects from different Filoviruses, as well as Lassa virus challenge.

Disentangling the dynamics of cross-reacting pathogens in serological studies: a study of arbovirus transmission in Bangladesh

Megan O'Driscoll, Department of Genetics

The increasing emergence and spread of different arboviruses has led to frequent local co-circulation of multiple related pathogens. This poses a number of challenges for the interpretation of serological studies where cross-reactive antibodies can complicate serostatus classification. Recent advances in multiplex immunoassay technologies have provided a significant opportunity for the improved characterization of population immune profiles across multiple pathogens. We developed an analytical approach, fit in a Bayesian framework, for the analysis of multiplex serological data. We utilize the information provided by the population-level distribution of antibody titres across pathogens, age and space to estimate pathogen- and location-specific transmission intensity, seroprevalence, as well as the level of cross-reactivity between every pathogen pair. We apply our model to results of a cross-sectional seroprevalence study in Bangladesh, where sera from 8,112 individuals were collected and tested for IgG antibodies to dengue 1, 2, 3 and 4, Zika, Japanese encephalitis, yellow fever, West Nile, tick-borne encephalitis, and chikungunya. We estimate high levels of cross-reactivity between Japanese encephalitis and West Nile, and between dengue 1 and 3, consistent with close phylogenetic distances. We find transmission of all dengue viruses, Japanese encephalitis, and chikungunya, with substantial variability in transmission intensity across communities.

Individual, household and community drivers of dengue virus infection risk in Kamphaeng Phet province, Thailand

Gabriel Ribeiro dos Santos, Department of Genetics

Dengue virus (DENV) often circulates endemically. In such settings with high levels of transmission, it remains unclear whether there are risk factors that alter individual infection risk. We tested blood taken from individuals living in multigenerational households in Kamphaeng Phet province, Thailand for DENV antibodies (N=2376, mean age 31y). Seropositivity ranged from 45.4% among those 1-5y to 99.5% for those >30y. Using spatially explicit catalytic models, we estimated 11.8% of the susceptible population gets infected annually. We found 37.5% of the variance in seropositivity was explained by household membership with only 4.2% explained by spatial differences between households. The serostatus of individuals from the same household remained significantly correlated even when separated by up to 15 years in age. These findings show that despite highly endemic transmission, persistent differences in infection risk exist across households, the reasons for which remain unclear.

The Cystic Fibrosis airway microbiome: quantifying Pseudomonas aeruginosa interactions with other microbes.

Pok-Man Ho, Department of Biochemistry

Cystic Fibrosis (CF) is a genetic disease associated with exuberant airway mucus production. This mucus hosts a variety of microbes, mostly infamously, Pseudomonas aeruginosa (PA). PA infections often persist for decades and are associated with a progressive decline in microbiome diversity and patient prognosis. However, we still do not understand why PA dominates the microbiome. The aim of my project is to understand how PA interacts with the other microbiota; are the interactions competitive, synergistic or neutral? To address this, I have developed a Bayesian model based around the Lotka-Volterra Competition relationship. The model allows us to extract pairwise “interaction coefficients” in multi-species ecosystems. We are applying this model to understand better inter-species interactions in an in vitro co-cultivation model containing three CF-associated pathogens: PA, Staphylococcus aureus (SA) and Candida albicans (CA). Our data indicate that (surprisingly) SA prey upon PA, CA is mutualism with PA and CA competes with SA in the triple-species co-culture. Currently, we are also applying the model to interpret inter-species interactions in longitudinal data from the CF registry. A better understanding of how species interact in the CF airways may eventually lead to improved therapeutic interventions which maintain microbial diversity in the CF airways.

Nipah Virus Diversity Across Different Spatial Scales in South and Southeast Asia

Oscar Cortés Azuero, Department of Genetics

Nipah virus (NiV) is a bat-borne zoonotic virus with high fatality when it infects humans. It has been identified throughout South and Southeast Asia. However, its genetic diversity remains poorly understood. Here, we compiled a comprehensive dataset of all known genomes for NiV and associated location and host species information (N=208 from six countries). We reconstructed a time-resolved phylogeny and organised the sequences into distinct clusters. We then developed statistical approaches that tracked the spatial spread of lineages. Finally, we used the relationship between the number of lineages observed within a division as a function of the number of isolates collected to predict the underlying number of clusters in each location. We found existing NiV sequences could be divided into 4 major clades and 12 clusters. The probability that a pair of sequences came from the same cluster ranged from 0.35 (95%CI 0.3-0.44) when the sequences were sampled within 10km of each other to 0.12 (95%CI 0.08-0.19) when they were 500-1000km apart. Overall, we found the geographic spread has been slow. Our findings suggest widespread viral diversity in NiV and that different areas across the region have different circulating viruses, which can only be observed with extensive, uniform sampling.

Structure-based discovery of lipoteichoic acid synthase (LtaS) inhibitors

Dr Taufiq Rahman, Department of Pharmacology

LtaS is a key enzyme for the cell wall biosynthesis of Gram-positive bacteria. The latter if lacking lipoteichoic acid (LTA) exhibit impaired cell division and growth defects. Thus, LtaS appears to be an attractive anti-microbial target. The pharmacology around LtaS remains largely unexplored with only two small-molecule LtaS inhibitors reported, namely 'compound 1771' and the Congo Red dye but how these molecules interact with LtaS remains unclear. We used molecular docking with molecular dynamics (MD) simulation to model a plausible binding mode of compound 1771 to the extracellular catalytic domain of LtaS (eLtaS). The model was validated using Ala mutagenesis studies combined with ITC. Lead optimization driven by our computational model resulted in an improved version of compound 1771, namely EN-19 which showed greater affinity for binding to eLtaS than compound 1771 in biophysical assays. EN-19 reduced LTA production in S. aureus dose-dependently, induced aberrant morphology as seen for LTA-deficient bacteria and significantly reduced bacteria titres in the lung of mice infected with S. aureus. Analysis of our MD simulation trajectories revealed the possible formation of a transient cryptic pocket in eLtaS. Virtual screening against the cryptic pocket led to the identification of a new class of inhibitors that could potentiate β-lactams against MRSA. Our overall workflow and data should encourage further drug design campaigns against LtaS.

Diversity and distribution of the tryptophanase gene among gut bacteria

Ellis Kelly, Department of Genetics

Indole is the central metabolite produced from tryptophan by gut bacteria, through the action of the enzyme tryptophanase. Tryptophanase catalyzes the conversion of L-tryptophan to indole and ammonium pyruvate and is encoded by the tnaA gene. Numerous studies have shown that the microbiota-derived indoles play a significant role in bacterial intercellular signalling and impact upon cardiovascular, metabolic, and psychiatric diseases.
However, which gut microbes in the microbiome have tnaA, and therefore potentially produce indole, remains largely unknown. Here we explore the genetic diversity of tnaA in its most well-characterised bacterium, E. coli, and within the gut microbiota itself (Unified Human Gastrointestinal Genome (UHGG) collection). We found that three of the phyla accounted for >90% of the 470 TnaA identified. Bacteriodiota and Firmicutes, two of the dominant phyla in the gut microbiome, could be significant contributors of indole production in the gut. Further, the maintenance of tnaA in clusters in the gut environment suggests the importance of tryptophanase to individual species, and to the microbiota. The insights obtained from this study inform on the relationship between bacteria in the gut microbiota, and leave the potential for therapeutic treatments through tryptophanase regulation.