Pump-Priming Project Awards - Round 3 Awardees

Dr Valeria Brizzio

Research, Development and Innovation Manager
Sinergium Biotech S.A. (Argentina)

Collaborators:
Professor Simon Cutting, CSO of SporeGen Ltd (UK)

Dr Pablo Pérez, Department of Biological Sciences, School of Exact Sciences,
UNLP–CIDCA (CONICET, La Plata) (Argentina)

Dr Pablo Baldi, IDEHU, Immunology Department, School of Pharmacy and Biochemistry, University of Buenos Aires (Argentina)

Summary

Clostridium difficile infection (CDI) is one of the leading causes of nosocomial antibiotic-associated diarrhoea. Although CDI is usually considered a disease restricted to HIC, recent data suggests that rates in LMIC are similar to that of Europe or USA. Importantly, the primary method of control is by the use of antibiotics yet this pathogen acquires AMR as well as multidrug resistance and so developing a prophylactic is a priority. Parenteral vaccines have failed. Overarching evidence suggests that to prevent CDI an approach that prevents spore germination of the pathogen and colonization of the host GI-tract is required. Thus, mucosal vaccination that induces secretory IgA responses is a rational approach. Recently, a recombinant spore vaccine expressing the C-terminus of toxin A (TcdA26-39) has been shown to protect against CDI, induce mucosal responses and has been taken to Phase 1 studies. Nevertheless, this vaccine should be improved to ensure higher levels of protection and to overcome downstream regulatory hurdles arising from the use of GMOs. In the current project, we will develop a vaccine candidate using the oral spore (Bacillus subtilis) vaccine approach with a revised vaccine formulation that delivers additional C.difficile antigens constructed utilising a novel cloning system (THY-X-CISE™) that allows the creation of spores that are unable to proliferate in the environment. This latter point is essential for addressing biological containment of genetically modified organisms (GMOs) and ultimately facilitating regulatory approval of the vaccine candidate. 

Project outcome

Bacterial spores from Bacillus subtilis were engineered to produce a protein derived from the spore of Clostridium difficile known as CotE. This protein is involved in host colonization, has enzymatic properties, and is a putative protective antigen. Expression of the protein was achieved on the surface of the Bacillus spores enabling these spores to be used in an oral vaccination approach.

These recombinant spores were tested for oral immunization in mice and hamsters either alone or in combination with recombinant spores (known as PP108) previously obtained by Dr. Cutting’s group. The later spores express another antigen from C. difficile, namely the C terminus of toxin A (TcdA_26-39). The elicitation of systemic and mucosal humoral immune response (specific antibodies in serum and fecal extracts) was tested in both animal species. At the end of the immunization plan, hamsters were challenged with viable C. difficile by the oral route. Specific antibodies could not be detected in either mice or hamsters from any immunization group. However, a trend to increased survival and decreased disease after C. difficile challenge was observed in hamsters immunized with spores expressing CotE (but not in the group receiving PP108 or the combination of both spores).

Professor Adam Cunningham

Professor of Functional Immunity
Department of Immunology and Immunotherapy
University of Birmingham (UK)

Collaborators:
Dr Stephen Reece, Kymab Ltd (UK)

Prof Paul Kellam, Kymab Ltd (UK)

Prof Ian Henderson, Institute for Molecular Bioscience (Australia)

Summary

Most vaccines that work in humans work through antibodies. Identifying antibodies present in human sera after previous exposure to a pathogen or a toxin has been a valuable way to identify new vaccine candidates (antigens) recognised by humans. This has been invaluable for guiding vaccine design as part of a process called “reverse vaccinology”. Nevertheless, it has significant limitations, not least because vaccination artificially “skews” the antibody response, something not possible to detect in normal human sera after natural infection. The Kymab mouse, overcomes this limitation by replacing mouse antibody genes with human antibody genes. This means we can immunise mice to identify what parts of an antigen are recognised by humanised antibodies. This will lead us to examine how human antibodies “see” antigen to provide protection from a vaccine. This can accelerate vaccine design and development and reduce costs in vaccine development. This is a major reason Kymab has attracted significant investment from organisations like Wellcome and BMGF. We will test the potential of outer membrane vesicles from Acinetobacter baumannii to induce protective antibodies against infections with this organism. This pathogen is a significant cause of hospital-acquired infections in high and LMICs and shows significant resistance to multiple antibiotics. We will examine the strength of binding of the antibodies induced to distinct antigens within the OMV and the comparative capacity of the antibodies induced to kill Acinetobacter baumannii bacteria. This is the first step to developing a new vaccine targeted to populations most at risk of Acinetobacter baumannii infection.

Project outcomes

Use of human sera to identify protective antigens recognised after natural infection has been pioneered under the term “reverse vaccinology”. This has proven an effective approach as demonstrated by the development of vaccines against MenB, such as Bexsero. Nevertheless, after any natural infection there are a myriad of antigens recognised, most of which are not protective and so this can naturally constrain this approach. Moreover, since vaccination is an artificial skewing of the antibody response towards protective antigens, the reverse vaccinology approach does not inform on how the human antibody repertoire develops to vaccines, at least not until phase I, which is significantly downstream in the developmental process.

It would be helpful to examine the human antibody repertoire in an animal model, where we can assess this in much greater mechanistic detail. Kymab have developed a mouse where this is achieved through transferring the human B cell repertoire across into the murine host. Immunising these mice has provided a fantastic opportunity to examine the targeting of antigens by “human” antibodies at the supraphysiological levels induced after vaccination with purified subunit vaccines. Her, we have immunised the Kymab mouse alongside wild-type (WT) control mice to compare how antibody responses develop after immunisation.

We performed extensive analyses of the microarchitecture of the spleen in humanised mice to examine whether this was affected by the chimerisation and its relationship to WT mice. Overall, standard architecture was maintained. We observed normal B cell follicle and T zone formation and distribution across the tissue. Moreover, we observed that immunisation with a prototypical vaccine induced significant numbers of structures called germinal centres – the powerhouse of antibody selection and essential for the generation of optimal responses to vaccines. Through modifying specific technologies we could localise IgM and IgG plasma cells in humanised mice.

These studies provide a route through which to study the development and mechanisms of selection of humanised antibody responses to vaccines that are used in humans. It provides enhanced validation of the Kymab mouse as a tool to study the development of vaccines of relevance to humans.

Dr Nicholas Evans 

Senior Lecturer
Dept of Infection Biology
University of Liverpool (UK)

Collaborators:
Dr Caryn Fenner, Afrigen Biologics (South Africa)

Summary

Leptospirosis is a worldwide, severe infectious disease affecting several different species including man and ruminants, particularly in LMIC countries. Globally, cattle are the most severely afflicted in terms of numbers, resulting in severe economic losses, food security impact and substantial antimicrobial use, as well as increased zoonotic spread. Bovine leptospirosis (BL) vaccines are available, although there are several constraints inhibiting use in LMIC countries where most of the disease burden is present. Current BL vaccines have a limited range of specificity, only last for a short duration and require cold chain transport and storage which is problematic in many tropical, frequently LMIC, regions with substantial disease burden.

Bacterial outer membrane proteins (OMPs) are considered an important target to provide cross-protective and long lasting immunity against a range of Leptospira species and serovars. Immune evasion by spirochetes (such as leptospires) is considered to utilise adhesion of host molecules to the bacterial cell surface. Recent research, mutating OMPs to prevent binding to host molecules has increased immune efficacy of OMP vaccines for bacteria. We have identified several OMP amino acids key for adhesion to the host for a thermostable leptospire OMP which already has demonstrated some protective efficacy. Here, through mutating OMP proteins we aim to develop a novel thermostable vaccine with broad Leptospira specificity and enhanced efficacy and safety. Vaccines are considered key mechanisms to reduce AMR. Making vaccines more broadly protective, easily accessible and financially affordable can only increase uptake globally, especially in LMIC countries, therefore decreasing antibiotic use and AMR.

Project outcomes

We were able to generate de novo specific vaccine candidate genes within expression plasmids using cutting edge gene synthesis technology at the University of Liverpool (UoL). These genes are novel as we designed them so that the proteins they encode no longer attach to host skin or host wound healing molecules. This design was informed using previous collected data, which demonstrates a natural diversity of binding to different host molecules by different bacterial strains. It is anticipated that the production of these mutated vaccine components should enable better access of the proteins to the host immune system on vaccination.

These vaccine candidate genes were used to produce the putative vaccine components through expression in Escherichia coli as insoluble aggregates. These aggregates were refolded and purified so that pure protein was produced. Initial stability assays included investigations into maximum protein concentration as well as repeating previous work showing that when using electrophoretic mobility-shift assays that the thermostable protein only unfold after boiling in chaotropic 8M urea for 1 hour.

We have developed structural models for the novel leptospiral proteins. We have identified that the amino acids identified as key for binding to host molecules, were for the most part determined as surface exposed.

We also investigated the diversity of the thermostable vaccine candidates using phylogenetic trees to better understand how many variants might be needed to make a cross-protective vaccine. When comparing sequence diversity rom various cattle relevant serovars and species from across the world there were six distinct clusters suggesting that, dependent upon future immunological studies into cross-reactivity, that a hexavalent vaccine might be needed to ensure efficacy in LMIC cattle.

A global panel of cattle serum has been collected. The protein OMPs have been screened for antibody (IgG1 and IgG2) seroreactivity, using ELISA verifying interaction with the host immune system. Purified OMPs were subjected to ELISA-based functional assays to investigate loss of host binding (ongoing).

Vaccine candidates were produced at UoL and transported to Afrigen in South Africa to undertake stability testing in adjuvant formulations. Electrophoretic mobility-shift assay were used to determine the stability as a function of temperature and time (ongoing).

Further protein was produced at UoL and detergent exchange completed into different high purity detergents and transferred to the University of Manchester for crystal trials. Three different detergents were used for two different vaccine candidate preparations which were trialled under a wider variety of crystallisation conditions. No crystals were identified after initial incubation at room temperature although after subsequent refrigeration some conditions are now more suggestive of enabling crystals containing birefringent flecks.

We have carried out knowledge exchange and skills transfer to the LMIC vaccine developer, Afrigen. Using a number of meetings we described the protein production and lab techniques used involving both University of Liverpool and University of Kelantan (Malaysia). Further meetings have occurred between all groups to determine what funding might be applied for subsequently and we have plans for applications to both Innovate (UK) and the BBSRC for funds to enable future collaborative work across all groups.

Dr William Horsnell

Associate Professor
Division of Immunology & Institute of Infectious Disease and Molecular Medicine, University of Cape Town (South Africa)

Collaborators:
Prof Tim Mitchell, Institute of Microbiology and Infection, University of Birmingham (UK)

Dr Anna-Karin Maltais, Eurocine Vaccines AB (Sweden)

Summary

Streptococcus pneumoniae (pneumococcus) is a leading cause of bacterial pneumonia and sepsis in children, particularly in LMIC countries. The most effective public intervention to reduce pneumococcal diseases is vaccination. Despite the use of effective polysaccharide-based vaccines, pneumococcal disease still affects significant numbers of children worldwide. Underlying this continuing prevalence are increasing cases of disease caused by serotypes not protected by existing vaccines.

As pneumococcus initially colonises the nasopharynx, local immunisation here via nasal spray is an appropriate and feasible vaccination strategy to control the nasal carriage which gives rise to subsequent invasive disease. New protein antigen based mucosal vaccines delivered to the nasal mucosa can provide strong protection and can protect against serotypes not currently protected by existing vaccines.

Effectively vaccinating mothers can provide protection from birth and has been demonstrated to be effective against other streptococcal diseases (e.g. Group B Streptococcus). Moreover, mucosal vaccine delivery in mothers can also provide effective protection to offspring from bacterial infection.

In this project, we will use preclinical models to demonstrate that mucosal maternal vaccination with novel pneumococcal vaccines can protect offspring from the establishment of pathogenic pneumococcal infections.

Project outcomes

Protecting offspring from susceptibility to infection by vaccinating their mothers can be an effective way to provide protection from infection from very early in life. This is due to the fact that mothers can transfer immune molecules to offspring while pregnant and while breastfeeding. If maternal vaccination can protect against Streptococcus pneumonia infection is unknown. Here we test how maternal vaccination against Streptococcus pneumonia can influence offspring immunity.

In this project we found that mothers vaccinated with two very different vaccines against pneumococcal bacterial infections transferred a striking immune effect to their offspring. Offspring acquired maternal antibody when very young, as expected. However, we found that offspring still had raised antibody levels against Streptococcus pneumonia when they had grown up. This suggests that offspring are acquiring a complex immune education from their mothers. This was supported by our detection CD4 T cells having a raised ability to launch anti-pneumococcal associated responses in offspring born to vaccinated mothers. Moreover, we also detected raised B cell populations in offspring born to vaccinated mothers. These cells maybe the source of long term anti- Streptococcus pneumonia antibody responses.

These findings therefore identify maternal vaccination as potentially a very effective way to induce long lasting immunity in offspring against Streptococcus pneumonia infections.

Dr Ellen Jessouroun

Bacterial Development Program Manager
Technological Development Department
Bio-Manguinhos, Oswaldo Cruz Foundation-Fiocruz (Brazil)

Collaborators
Dr José Procópio Moreno Senna, Technological Development Department Bio-Manguinhos, Oswaldo Cruz Foundation-Fiocruz (Brazil)

Dr Barbara Bolgiano, Division of Bacteriology, National Institute for Biological Standards and Control (UK)

Dr Nicola Beresford, Division of Bacteriology, National Institute for Biological Standards and Control (UK)

Summary

Antimicrobial costs have been increasing steadily every year and special attention has been given by Brazilian public agencies, such as the National Agency of Sanitary Surveillance (ANVISA), to resistant bacterial infections. The study proposal is the development of a bivalent conjugate vaccine against A. baumannii and S. agalactiae which cause a high incidence of infections in hospitals in Brazil. The Streptococcus species also causes meningitis in newborn babies. This vaccine project targets two groups: adults susceptible to hospital-acquired infections, and pregnant women, who are at risk of S. agalactiae infection.  Taking advantage of the skills acquired by Bio-Manguinhos, from Oswaldo Cruz Foundation, in the development of conjugate vaccines for meningococcus, the study intends to develop a vaccine obtained by the chemical conjugation between the capsular polysaccharide of S. agalactiae and the major outer membrane (OM) protein, OmpA, from outer membrane vesicles (OMVs) from A. baumannii. The effectiveness of the vaccine, in inducing protection against the two target bacteria, will be assessed in mice. The proposal goal will be the search of correlates of immunity, through the evaluation of the induced antibody functionality and the changes to adhesion pattern of the two microorganisms to epithelial cell monolayers. Bio-Manguinhos will produce the experimental vaccines, immunize the mice and evaluate the functionality of the induced antibodies to protect against the two vaccine target pathogens. The adhesion test will be also performed by Bio-Manguinhos. NIBSC, the National Control Laboratory of the U.K. will evaluate the quality of glycoconjugate and OMV vaccines and components.

Project outcomes

The main findings of this project to develop a bivalent vaccine to protect against two organisms of significance in Brazil and globally are:

• The Streptococcus agalactiae polysaccharide production process was optimised and polysaccharide capsule has been obtained of the expected quality described in the literature for Group B Streptococcal serotype Ia. This component can potentially protect pregnant women and their unborn babies against neonatal meningitis.

• Outer membrane vesicles (OMVs) were obtained from Acinetobacter baumannii, which can potentially protect critically ill hospital patients from acquiring infections in intensive care settings.

• Characterisation and proteomic study of A. baumannii's OMV was a great advance, especially in evaluating the effect of detergent-treatment on vesicle size, lipid and protein composition and endotoxin content. This aided the production of safe conjugate vaccine material for the immunisation of mice, and also confirmed the presence of Outer Membrane Protein A (OmpA), the candidate carrier protein, in OMVs made with and without detergent.

• An adhesion inhibition test to evaluate if vaccine sera can prevent S. agalactiae and A. baumanni from binding to epithelial cells was established as an additional test to evaluate the immune response induced by experimental vaccines.

The project demonstrated that an experimental conjugate vaccine against two organisms (for which vaccines are not currently available) has the potential to protect against infection. Its further success will depend on improving conjugation reactions and characterisation of the obtained molecules, so a choice can be made to select the best carrier protein to be used. The work developed in a period of so many restrictions also showed the group's ability to seek work solutions that were not previously foreseen and reinforced the value of partnerships with groups with diverse experience and expertise inside and outside Fiocruz.

Professor Beate Kampmann

Theme Leader Vaccines & Immunity Theme and Professor of Paediatric Infectious diseases and immunology
Vaccine & Immunity Theme, MRC The Gambia at the London School of Hygiene & Tropical Medicine (The Gambia)

Collaborators:
Dr. Anja Saso, Vaccine and Immunity Theme, MRC Unit The Gambia at the London School of Hygiene & Tropical Medicine (The Gambia)

Summary

Whooping cough, known as pertussis, is a serious infection, especially in babies. Two types of pertussis vaccine exist: whole-cell (wP) and acellular (aP). aP vaccines may not be as effective in preventing whooping cough as wP vaccines, but the cause remains unknown. It may be due to differences in how these vaccines act at the surface (‘mucosa’) lining the inside of the nose and upper airway, where pertussis bacteria initially infect. Data from animal models show that immunisation with wP vaccines (but not aP) prevents pertussis bacteria infecting the nose, clearing them before any further symptoms develop and transmission to other individuals can occur. However, we do not know enough about the type of immune defence induced at the mucosa by the two different vaccines, and there is no data in children. Our project is set within a large study in The Gambia that is vaccinating babies with aP or wP vaccines. We will investigate components of the infant’s immune system activated by these vaccines at the mucosa, including cytokines and specific antibodies to pertussis bacteria. To achieve this, we will collect and analyse fluid from each infant’s nose. Blood samples are already being collected and we will add these valuable mucosal results to improve our insight into how these vaccines differ in their action. Our findings will help to design better pertussis vaccines in the future. This project links our opportunities and expertise in The Gambia to the UK and a wider European Consortium of academic and industry partners.

Project outcomes

Although vaccines to prevent pertussis (whooping cough) have been given to infants for decades, whooping cough remains a serious respiratory illness, estimated to contribute to 160,700 deaths in children younger than five years worldwide. Two types of safe and effective vaccines are currently in use, but how long they might be protective for appears to vary between them: compared to an acellular type of vaccine, it looks as if the pertussis whole cell vaccine provides longer lasting protection and appears to also protect against colonisation with Bordetella pertussis in the nose, and not just disease in the lungs. However, it is known to have more side effects, and in many countries was substituted with the acellular type of pertussis vaccine several decades ago. Rises in cases of countries that had undertaken this switch have been observed, especially in young babies, and consequently, pregnant women are now also vaccinated to protect newborn babies from this potentially serious infection through passive transfer of antibody during later stages of pregnancy. This has been shown to be very successful in reducing cases in newborn babies. Nevertheless, to date, it is not established what an optimal pertussis vaccine should actually look like, as there is no validated so-called correlate of protection, and immunity appears to rely on both antibody and cellular immune responses.

We set out to understand the differences in immunity induced by either type of pertussis vaccine in African children, who were born to women who had either received pertussis vaccine in pregnancy or not. During this controlled clinical trial in The Gambia, we also wanted to understand the role of antibody and cytokines at the mucosal surface - in this case in the nose - that is induced by either type of vaccine and which might play an important role in preventing the organism causing whooping cough (Bordetella pertussis) to take hold. We therefore collected samples directly from the nose of the infants using a special filter paper to absorb secretions, which could be used to measure a variety of immune parameters.

This method had never been employed in African children in this context. We now have data that show the production of antibody and cytokines (immune mediators) in nasal secretions, and our ongoing work will relate these findings to the antibody and immune mediators induced in blood of the same infants. This will allow us to gain a more comprehensive understanding of what happens at the local and systemic level in response to the vaccination and what the impact might be on other bacteria that also live in the nose.

The BactiVac funding has facilitated to extend our research from looking at immunity in the blood stream to measuring immunity in the nose. Whilst our work is still ongoing, we expect that these findings will help us to further optimise pertussis vaccines going forward and hopefully understand in more detail what type of vaccine induces the strongest and most durable protection.

Dr Kirsty Le Doare

Clinical Senior Lecturer
St George’s, University of London, Institute of Infection and Immunity (UK)

Collaborators:
Dr Per Fischer, MinervaX ApS (Denmark)

Dr Bengt Johansson Lindblom, Dept. Experimental Medical Science, Lund University (Sweden)

Dr Stephen Cose, MRC (Uganda)

Mr Thomas Hall, St George’s, University of London, Institute of Infection and Immunity (UK)

Summary

Group B Streptococcus (GBS) is a leading cause of neonatal infections and causes stillbirths and preterm births.  GBS can also cause infections in the elderly and immunocompromised host. Vaccine development is currently undergoing clinical trials and there are two vaccine candidates, one developed to make antibodies against the GBS coat (the capsule) and one against key proteins on the GBS surface. The capsule changes with each of the 10 types of GBS, but key proteins are the same for all types of GBS. Although GBS is a leading cause of disease, it is still relatively rare and testing a vaccine in such a rare disease would be too complex and too costly. It is now widely agreed that giving a vaccine against to pregnant women is a priority and that the vaccine could be licenced based on measuring the amount of antibodies needed to protect against infection using standardised laboratory tests followed by a large clinical trial after the vaccine is licenced to look at vaccine effectiveness. Such an approach has been used for vaccines like the meningitis B vaccine given to children under the age of 1 year. However, to develop this laboratory test we need to develop standard reagents that every laboratory wishing to test antibodies can use. We are already developing reagents to study antibodies against the capsule and this proposal aims to now develop reagents against key proteins so that we can finally get GBS vaccines to those who need them most.

Project outcomes

We have successfully developed an assay to simultaneously measure levels of antibodies against the N-terminal domains of three key proteins on the surface of GBS; AlphaC, Alp2 and Alp3 proteins (Alp2 and Alp3 have the same N-terminal domain (Alp2/3-N) meaning that the current version of the assay measures two specificities covering three proteins). These proteins are members of the Alpha-like protein (Alp) family, also containing Rib, Alp1 and Alp4. The Alps are expressed by GBS variant strains as allelic variants and essentially all invasive strains encode at least one family member. Alp4 is however extremely rare. The Alps are thought to be involved in helping GBS to cross into the bloodstream and cause disease in babies and mothers.

The multiplex assay developed is adapted to easily incorporate measurement of antibodies also against Rib-N and Alp-1, yielding a multiplex assay that can measure antibodies against all clinically relevant Alps/GBS strains. When we developed the assay, we were however aware of that some cross-reactivity between the different Alps exists and while we now have ruled out that this cross-reactivity interferes with the assay’s ability to determine the concentration of AlphaC-N and Alp2/3-N binding antibodies, it remains to be demonstrated that cross-reactivity does not impact on accuracy when measuring concentration of antibodies against the remaining Alp-Ns. However, one of the reagents we used (avidin beads) did give us some incorrect antibody readings. So, we developed an improved method of measuring the antibodies, to make sure that the results of our assay show only antibodies against the proteins of interest. In the future, the reagents we have developed will be available via NIBSC together with a standard protocol for the assessment of antibodies against key proteins on GBS. Combining all proteins into an antibody detection test may be achievable and could be used to measure the amount of antibodies that a pregnant woman needs to pass to her baby to protect against GBS disease (serocorrelates of protection) in the first months of life.

Professor Constantino III Roberto López-Macías

Principal Investigator
Medical Research Unit of Immunochemistry (Mexico)

Collaborators:
Professor Yvonne Perrie, Strathclyde Institute of Pharmacy and Biomedical Sciences, University of Strathclyde (UK)

Summary

Salmonella infections in humans are the cause of enteric fevers (Typhoid (TF) and Paratyphoid (PT)) and Non-Typhoidal Salmonellosis (NTS), which are major causes of morbidity and mortality in developing countries. There are no commercially available vaccines for PT and NTS and current vaccines for TF do not offer effective control of the disease. Thus a multi-valent vaccine that protects against all the different clinical manifestations remains an unmet public health need. Porins are outer membrane proteins from Salmonella enterica that generate immune responses in acute and convalescent TF and NTS patients. Purified porins have shown promise as a multivalent vaccine in preclinical testing and have also been shown to be safe and well tolerated in human volunteers. In addition, volunteers who received the porin-based vaccine carry circulating bactericidal antibodies up to 10 years after receiving one single immunisation. So far the porin-based vaccine formulation has been tested using conventional systemic delivery. In addition to being a non-invasive method for vaccine administration, mucosal delivery has the potential of eliciting protective immune responses at the pathogen site of entry. However, mucosal vaccination requires potent adjuvants and delivery systems to enhance immunogenicity, and to decrease the degradation rate. Particulate delivery systems such as micro/nano particles and liposomes have the ability to protect and carry subunit antigens to mucosal inductive sites. We aim to formulate particulate delivery systems for a porin multivalent vaccine for mucosal administration.

Project outcomes

Highly pure Salmonella porins were produced and formulated with liposomes and chitosan. Porins-Dimethyldioctadecylammonium (DDA) or trehalose 6,6-dibehenate (TDB). DDA:TDB liposomes were produced in either Tris buffer or two different concentrations of sucrose (2.5 and 20 %). After preparation, liposomes prepared in Tris buffer were 800-900 nm, whilst those prepared in sucrose were smaller in size (500 – 700 nm). All three formulations were highly cationic (zeta potential 50-60 mV), with no significant difference. The porins composed liposomes effectively retained porins within the liposomal adjuvants offering protection and enhanced delivery as shown by biodistribution studies. To enhance their shelf-life, a freeze-dried formulation was developed. After freeze-drying, high aggregation was noted with formulations prepared in Tris (with a 5-fold increase in size), whereas those prepared in 2.5% and 20% sucrose had no significant increase in size for up to 2 weeks at room temperature. These formulations produce strong antibody responses after intra-muscular administration.

Porins loaded chitosan micro- and nanoparticles were prepared with suitable particle size and distribution, surface charge, loading efficiency, in addition porins-chitosan gel was formulated. The particles and gel were shown to be stable, protecting porins integrity. Biocompatibility, cellular uptake and macrophage activation of these formulations were demonstrated in J774A.1 macrophage cell line. In BALB/c mice immunised with porins+chitosan nanoparticles or porins+microparticles showed an increased antibody titres when compared to porins alone. To analyse the protective response, BALB/c mice were immunised intraperitoneally on day 0 and boosted on day 14 with saline, or porins or porins-chitosan formulations, animals were challenged with Salmonella Typhi on day 30, and the reduction of bacterial numbers in liver and spleen was determined 24h after infection. We found that porins-chitosan formulation was more efficient than porins alone to reduce bacterial numbers in liver, whereas in the spleen, both formulations induced similar reduction of bacterial numbers.

Taken together, our results showed that porins could be efficiently formulated in liposomes and chitosan to form micro- and nano-particles, these formulations are immunogenic and suitable to induce protection against Salmonella challenge in mice. These formulations could be used to produce room temperature vaccines for non-invasive immunisations that improve its global distribution and facilitate its use.

Dr Barry Moore 

Reader of Biophysical Chemistry
Pure & Applied Chemistry, University of Strathclyde (UK)

Collaborators:
Dr Minoarisoa Rajerison, Institut Pasteur Madagascar (Madagascar)

Dr Diane Williamson, DSTL (UK)

Summary

The most serious outbreak of plague in modern times occurred in Madagascar in 2018, with in excess of 2600 cases and an estimated case fatality rate of 8.9% (WHO 2018). In Madagascar as well as in other parts of the world, plague causes seasonal outbreaks, with risk of epidemic potential and transmission to new regions. These seasonal outbreaks are caused by flea-vectored transmission from wildlife reservoirs (principally rats).  There is no approved vaccine for plague and antibiotic therapy needs to be given early after exposure to infection, to be fully effective. Here we propose to test a sub-unit vaccine in a novel formulation for efficacy against a Malagasy strain of the causative bacterium, Yersinia pestis.  In a previous liquid formulation, this sub-unit vaccine has been shown to be efficacious in mice and macaques against the reference Y.pestis Co92 type strain and was also shown to be safe and immunogenic in a Phase 1 clinical trial (Williamson et al. 2005).  Here, we have reformulated the vaccine for distribution to an LMIC, as a stable, dry powder which is reconstituted just before use.  We will test the vaccine under laboratory conditions at the Institute Pasteur in Madagascar in rats derived from either plague-infected or non-infected areas of the island, prior to challenge with a circulating Malagasy strain of plague. The objective is to determine if this vaccine can induce immunity in the local rat population, preventing transmission to man. Production of a vaccine will mitigate development of AMR resistant strains of plague.

Project outcomes

Candidate plague vaccine formulations were produced in the UK and tested at the Institut Pasteur of Madagascar. The novel dry powder formulations were shipped to Madagascar at room temperature and tested using Malagasy wild rats from plague endemic areas or from plague free areas. Vaccine formulations were prepared by incorporating protein antigens within calcium phosphate coated microcrystals (CaP-PCMC) and formulations containing either separate F1, V antigens or an F1-V fusion protein were tested.

The study showed that all of the formulations produced an immune response in rats specific to the F1 antigen, following subcutaneous delivery of the vaccine and this typically increased following boosting. Notable differences were observed with the rats from plague endemic areas producing a stronger immune response even where first generation laboratory bred rats were used. On day 57 the rats were challenged with a high dose of a Malagasy strain of plague (107 cfu). Protection against challenge was observed in most rats that showed a strong antibody titre. This is important as it indicates that the selected antigens are effective against a Malagasy strain of plague and also that the CaP-PCMC formulation allows for room-temperature shipment of the vaccine.

Notably it was also observed that a single dose of vaccine was sufficient to provide protection for formulations containing either the individual antigens or the F1-V fusion. Thus, there is a clear opportunity to use this approach in the development of an emergency LMIC vaccine and this study provided important information about the factors that will need to be considered when selecting a lead vaccine candidate to take forward for clinical development.

This work was a collaboration between DSTL (antigen production), University of Strathclyde (vaccine formulation) and the Institut Pasteur of Madagascar (vaccine study).

Professor Marco Rinaldo Oggioni

Professor
Dept. Genetics and Genome Biology, University of Leicester (UK)

Collaborators:
Dr Mariagrazia Pizza, GSK Vaccines (Italy)

Professor Peter W Andrew, Dept. Respiratory Science, University of Leicester (UK)

Summary

The bacterium Escherichia coli normally inhabits our intestines but when it occurs in body sites outside the intestines, it can cause serious diseases. These extraintestinal pathogenic Escherichia coli (ExPEC) are important causes of morbidity and mortality, worldwide, being the number one cause of urinary tract infections and being the second most frequent cause of meningitis and sepsis in neonates. The extent of these infectious diseases plus the increasing occurrence of antibiotic resistance ExPEC means that possession of an effective vaccine is a public health priority. Unfortunately, effective protective vaccines against ExPEC are not available at present. This joint project of the University of Leicester with GSK Vaccines aims to make an important contribution to the solution of this deficiency. One of the hurdles in vaccine development is the difficulty in doing large scale vaccine efficacy tests in animals or humans. An excepted means to overcome this issue is to use straightforward in vitro tests that reflect and predict efficacy in vivo, in other words are correlates of protection. Sadly, in the case of ExPEC, the classical opsonophagocytic or bactericidal assays do not give correlates of protection, raising the need for novel tests to prioritise vaccine antigens during preclinical screening. We have recently discovered that bacterial replication within splenic macrophages precedes and correlates to invasive sepsis disease. Using samples of spleens in culture, in this project we will test if prevention this splenic phase of ExPEC infection is a correlate of protection and can be used for vaccine antigen screening. 

Project outcomes

Our research hypothesis was that we would be able to detect significantly different bacterial organ counts in early infection stage animals few hours after the challenge in mice vaccinated with protective antigens, both by passive or active immunization. The different counts were then to be traced back to differences in interaction of the bacteria with tissue macrophages.

The experiments provide a clear response to our hypothesis. Unfortunately, for our research perspective, our initial hypothesis was not supported by the data.

Dr Christine S Rollier

Associate Professor
Preclinical and Novel Vaccine Development team
Oxford Vaccine Group, Department of Paediatrics, University of Oxford (UK)

Collaborators:
Dr V Krishna Mohan, Bharat Biotech International Ltd (India)

Prof Andrew J Pollard, Oxford Vaccine Group, Department of Paediatrics, University of Oxford and Children’s Hospital, Oxford (UK)

Dr Christina Dold, Oxford Vaccine Group, Department of Paediatrics, University of Oxford (UK)

Summary

Typhoid and paratyphoid fever together cause a high burden of disease in low-and-middle-income countries, responsible for up to 21.6 million cases and 216,510 deaths annually. These pathogens are developing resistance to antibiotics, to the point that some infections cannot be treated anymore. Moreover, where enteric fever is endemic, health care providers prescribe antibiotics unnecessarily for any febrile illness, and this practice contributes to the dangerous rise in antibiotic resistance. In this project, we propose to create a novel vaccine against both typhoid and paratyphoid fever that has a realistic path to clinical development and commercialisation, based on an existing vaccine that we improve by combination with a novel component.

Conjugate vaccines against typhoid fever (TCV) have been developed and are being produced by several companies, and will be used to prevent typhoid. However there remains no vaccine to prevent the substantial paratyphoid disease burden in Asia. Importantly, high cost is a limiting factor for wide use of vaccines in low-income countries, therefore a bivalent vaccine targeting both infections would be a particularly suited approach to improving health in the affected areas. Developing a novel bivalent vaccine that incorporates the TCV provides a realistic path to clinical development as compared to a new vaccine composition that has not proven efficacy against typhoid fever yet. We have developed a vaccine against paratyphoid in a previous project, and in this project we propose to combine the new paratyphoid vaccine with a TCV.

Project outcomes

Typhoid and paratyphoid fever together cause a high burden of disease in low-and-middle-income countries, responsible for up to 21.6 million cases and 216,510 deaths annually. Conjugate vaccines against typhoid fever (TCV) have been developed and are being produced by several companies, and will be used to prevent typhoid. However there remains no vaccine to prevent the substantial paratyphoid disease burden in Asia. Importantly, high cost is a limiting factor for wide use of vaccines in low-income countries, therefore a bivalent vaccine targeting both infections would be a particularly suited approach to improving health in the affected areas. We have developed a vaccine against paratyphoid in a previous project, but it is not known if our novel paratyphoid vaccine, based on a viral-vectored formulation, can be mixed or used at the same time as a conjugate vaccine.

In this project we explored that possibility. We combined the new paratyphoid vaccine with a TCV, and, interestingly, demonstrated that both retain their capacity to induce the desired responses. This is an encouraging finding that may open new possibilities in the field of vaccine development against enteric fever but also other diseases.

Professor Robin Shattock

Professor of Mucosal Infection and Immunity
Imperial College London (UK)

Collaborators:
Dr Francesca Micoli, GSK Vaccines Institute for Global Health (Italy)

Dr Anjam Khan, Centre for Bacterial Cell Biology, Institute for Cell and Molecular Biosciences, Newcastle University (UK)

Dr Mariagrazia Pizza, GSK Vaccines (Italy)

Summary

Background

Bacterial diseases are a grave threat for humankind causing approximately six million deaths per year. Antibiotic resistance is increasing. Hygiene measures are failing. Global travel makes disease control increasingly difficult. Enterotoxigenic Escherichia coli (ETEC) and Shigella are enteropathogens that cause significant global mortality and morbidity, particularly in low- and middle-income countries (LMIC). Development of new and improved vaccines against diarrheal infections is a fundamental element of the strategy towards reducing deaths from diarrhea in children under 5 years of age. No vaccines are currently widely available for both residents and travellers to endemic areas.

Aim

The aim of the proposed project is to use Generalised Modules for Membrane Antigens (GMMA) as a means to generate a multivalent and low cost vaccine to protect humans against diarrheal diseases caused by Shigella and ETEC. GMMA are outer membrane vesicles naturally shed by Shigella and other Gram-negative bacteria genetically manipulated to increase blebbing and decrease reactogenicity. GMMA present multiple antigens in the context of a membrane and have optimal size for immune stimulation plus self-adjuvanting properties. These vesicles are easy and cheap to produce, are strongly immunogenic and protective. GMMA can also be genetically or chemically modified to present heterologous antigens from other pathogens, supporting the development of multivalent vaccines.

Work plan

Here, we will test the feasibility of using the GMMA approach for the development of a vaccine covering both Shigella and ETEC. As proof of concept we will use Shigella sonnei GMMA as scaffold and test different strategies for the delivery of LTK63, a not toxic derivative of the Heat-labile enterotoxin (LT) antigen of ETEC, together with the immunodominant O-Antigen (OAg) of Shigella sonnei. LTK63 is a powerful immunogen, non-toxic but maintaining all immunogenic and adjuvant properties of the wild-type LT toxin and is expected to play a Key role as strong mucosal immunogen and adjuvant. S. sonnei GMMA presenting LTK63 will be fully characterised and tested in robust and tractable preclinical animal models to investigate the immunogenicity and protective activity of these multivalent vaccine preparations.

This project will represent a valuable starting point for the development of a low cost and effective multivalent GMMA-based vaccine protecting against both Shigella and ETEC, that have a high incidence worldwide and coexist in many geographical areas, especially LMIC.

Project outcomes

Shigella and Enterotoxigenic E. coli (ETEC) have a high global incidence and are major causes of bacterial diarrhoea worldwide, especially in children under five years of age in low and middle income countries (LMICs). Various studies have indicated that antibodies generated against the O-Antigen (OAg) portion of Shigella lipopolysaccharides (LPS) correlate with protection from disease and also that the effects of ETEC infection can be mitigated by the action of antibodies against the bacterial toxins. We combined the ETEC toxin and Shigella OAg LPS to make a vaccine which aimed to target these pathogens, producing an immune response with the potential to reduce infection and disease.

We used a Shigella sonnei strain, that express the OAg and are already mutated to increase outer membrane blebbing and to make generalised modules for membrane antigens (GMMA). These are tiny virus sized non-infectious vesicles that are ideally suited to generate an immune response. We then engineered this Shigella strain to express the LTK63 toxoid antigen, that would either be expressed on the surface of GMMA or could be directly bound to their surface. This approach therefore aimed to generate immune responses to both bacteria in a single vaccination. We also looked at different routes of vaccination with the aim to induce both blood responses and mucosal responses, particularly the gut as the main site of entry that these bacteria use to get into the body.

The project successfully generated vaccines that expressed both the Shigella OAg and the ETEC LTK63 antigen and these were characterised and then used in animal studies to investigate the immune responses. The two components of the vaccine were used either embedded within the same GMMA vesicle or the LTK63 was physically chemically attached to the GMMA vesicle. Both worked very well with the physical linkage generating slightly higher levels of specific antibodies, however the differences were not significantly different. In these studies we also measured the effect of administering the vaccines to the mice via different immunisation routes. These animal studies demonstrated that the preparations were very potent, generating high levels of antibody responses that were functionally able to neutralize LT and had bacteriocidal activity against Shigella.

Due to the COVID pandemic the animal studies were delayed but by the end of the program the intended mouse studies that aimed to examine the effect of different routes had been completed, with the only remaining work being the assessment of the immune responses, the mucosal response and sera neutralization / bacteriocidal activity. These final analyses are in progress and will be completed within 3 months after the end date of the project, at which time the results will be published in a suitable peer reviewed journal.

Dr Matthew Siggins

Postdoctoral Research Associate
Imperial College London, Department of Medicine (UK)

Collaborators:
Professor Shiranee Sriskandan, Imperial College London, Department of Medicine (UK)

Professor Adam Finn, University of Bristol, Bristol Children’s Vaccine Centre (UK)

Summary

Antimicrobial resistance (AMR) represents a significant problem to health, particularly in low- and middle-income countries (LMIC). Vaccines are proven in combating AMR and reducing deaths from drug-resistant infections. However, irrespective of a multitude of antigen candidates and the development of novel adjuvants, efficacious vaccines that provide one-dose, broad and long-lived immunity remain elusive for many important AMR bacterial pathogens.

Advances in vaccine immunology have revealed that strong and lasting humoral and cell-mediated immunity is generated following sustained delivery of high levels of vaccine antigens to lymph nodes. Investigators have tried to achieve targeting of lymph nodes through use of hydrophilic nanoparticles. However, the properties that ensure efficient lymphatic delivery also hinder internalisation by antigen presenting cells within lymph nodes, meaning that these particles are poorly retained, hampering immunogenicity and causing systemic toxicity.

Recently, we demonstrated that Streptococcus pyogenes exhibits tropism for lymph nodes due to its hyaluronan capsule. We propose engineering the harmless probiotic bacterium Lactococcus lactis to express hyaluronan capsule and vaccine antigens against AMR bacteria to serve as a lymphatic-homing vaccine vector that persists in lymph nodes to provide sustained delivery of vaccine antigens in situ.

The project will assess the persistence and antigen production of our recombinant vaccine vector within lymph nodes and compare generated immune response with traditional vaccine approaches. These proof of concept murine studies represent an important first step for an approach that could deliver a flexible, low-cost vaccine vectors that can be manufactured in LMIC countries and provide long-lived humoral and cellular immunity.

Project outcomes

Antimicrobial resistance (AMR) represents a significant problem to health, particularly in low- and middle-income countries (LMIC). Vaccines are proven in combating AMR and reducing deaths from drug-resistant infections. However, irrespective of a multitude of antigen candidates and the development of novel adjuvants, efficacious vaccines that provide one-dose, broad and long-lived immunity remain elusive for many important AMR bacterial pathogens.

Advances in vaccine immunology have revealed that strong and lasting humoral and cell-mediated immunity is generated following sustained delivery of high levels of vaccine antigens to lymph nodes. Investigators have tried to achieve targeting of lymph nodes through use of hydrophilic nanoparticles. However, the properties that ensure efficient lymphatic delivery also hinder internalisation by antigen presenting cells within lymph nodes, meaning that these particles are poorly retained, hampering immunogenicity and causing systemic toxicity.

Recently, we demonstrated that Streptococcus pyogenes exhibits tropism for lymph nodes due to its hyaluronan capsule. We hypothesised that engineering a harmless probiotic bacterium such as Lactococcus lactis to express hyaluronan capsule and vaccine proteins could serve as a lymphatic-homing vaccine vector that persists in lymph nodes to provide sustained delivery of vaccine antigens in situ. An additional advantage of this strategy is that the low cost and ease of both development and production would allow LMIC to take a lead in vaccine manufacture.

Our BactiVac Catalyst Pump-Priming award funding allowed us to genetically modify L. lactis to express hyaluronan, as well as fluorescent proteins. Using a mouse model, we discovered that expression of this hyaluronan coat allowed L. lactis to persist inside the body longer than control L. lactis (which did not express hyaluronan), this enhanced persistence could be important for generating stronger immune responses. Importantly for safety, hyaluronan expression did not permit L. lactis to invade and drive infection, and the mice remained healthy as the vaccine vector was cleared over time.

As hypothesised, hyaluronan expression also directed lymph node-homing, with counts of hyaluronan expressing vaccine vector in lymph nodes significantly increased compared to control L. lactis vectors. While the COVID-19 pandemic has, for now, delayed experiments assessing whether lymph node-homing drives better immune responses; these proof of concept murine studies represent an important first step. The hyaluronan expressing bacterial vaccine vector approach could deliver flexible, low-cost vaccine vectors that can be manufactured in LMIC countries and might be capable of generating longer-lived humoral and cellular immunity.

Dr John Tregoning 

Senior Lecturer
Imperial College London (UK)

Collaborators:
Prof Paul Kellam, Kymab Ltd (UK)

Prof Stephen Baker, Medicine, Cambridge and OUCRU Vietnam (UK & Vietnam)

Miss Sophie Higham, Imperial College London (UK)

Dr Stephen Reece, Kymab Ltd (UK)

Dr Aisha Krishna, Kymab Ltd (UK)

Summary

Untreatable infections caused by antimicrobial resistant bacteria are one of the major healthcare threats facing mankind. Most of these fatal infections occur in low- and middle-income countries (LMICs), where drug resistant bacteria are more prevalent. We are aiming to prevent infections caused by the bacterium Acinetobacter baumannii, a common cause of infections within hospitalised individuals; the disease has a very high mortality rate in Southeast Asia. One approach to tackle the problem of drug resistant Acinetobacter baumannii is to develop vaccines to prevent infection. In order to develop vaccines we need to understand how our immune systems see the bacteria, and to specifically identify which parts of the bacteria can be targeted by antibodies to prevent infection.

Our project is a collaboration between two leading academic institutions, Imperial College and the University of Cambridge, researchers in Vietnam and the biotechnology company Kymab. We will screen bacterial isolates collected from hospitals in Vietnam to find novel antigens. In particular we will be using infection models to determine whether these new vaccines can protect against infection. This work will provide a platform for future work to control this pathogen.

Project outcomes

Lower respiratory tract infections (LRTIs) are the leading cause of infectious death globally. Affecting low to middle income countries most severely, major risk factors include HIV, low vaccination rates, overcrowding, and malnutrition. Pneumonia is a leading cause of death in children under 5 years old. Ventilator associated pneumonia (VAP) is defined as pneumonia that occurs over 48 hours after a patient has been mechanically ventilated. VAP is the most frequently occurring infection in intensive care units of hospitals and has the highest associated costs and mortality rates out of any infections acquired through hospitalisation.

The bacteria Acinetobacter baumannii is frequently responsible for causing hospital acquired pneumonia. A. baumannii is a Gram negative bacterium, which means it has an external protective membrane around it. This outer membrane provides protection against many antibiotics. A. baumannii have also acquired many genetic elements that give the bacterium additional mechanisms to both evade antibiotics and reduce their effectiveness. Infections with A. baumannii can be very problematic. Many different strains of A. baumannii are now able to survive treatment with many modern day antibiotics. A. baumannii are constantly evolving and developing new mechanisms to protect themselves against the already severely limited options. With the efficacy of antibiotic treatment quickly declining and the slow development of new antibiotics we must develop new methods to treat antibiotic resistant infections.

We have been using A. baumannii bacterial isolates that have been collected from hospitals in Vietnam to develop lower respiratory tract infection models using clinical strains of A. baumannii and develop protocols that can be used to assess the efficacy of the immune response. We were able to intranasally infect mice with a laboratory reference strain of A. baumannii, which resulted in low levels of colonisation and mild disease.

When we intranasally infected mice with clinical strains we saw increased levels of colonisation and more severe disease when compared to the laboratory reference strain. We then vaccinated mice using outer membrane vesicles (OMVs), which are spherical vesicles that are produced naturally by Gram negative bacteria and are made up of an inner cell membrane and a bacterial outer membrane. OMVs produced by the laboratory reference strain and clinical strains of A. baumannii were used to immunise mice in a prime boost approach before intranasal infection with the same strain of A. baumannii that each mouse was vaccinated against. We found that OMVs provide protection against infection in mice, increasing levels of OMV specific antibodies and reducing bacterial loads.

In conclusion we developed effective A. baumannii infection models with both clinical strains and a laboratory reference strain. We also showed that A. baumannii OMVs can be used to vaccinate mice to reduce bacterial burden and induce A. baumannii cross-reactive antibody responses. The development of this model provides a platform that can be used to test novel vaccines and therapeutics against multi drug resistant A. baumannii respiratory tract infection. We have also shown that OMVs have good potential to be used in the development of an A. baumannii vaccine.

Dr Sandra Van Puyvelde 

Postdoctoral Research Fellow
Vaccine and Infectious Diseases Institute (VAXINFECTIO)
University of Antwerp (Belgium)

Collaborators:
Dr Francesca Micoli, GSK Vaccines Institute for Global Health (Italy)

Professor Calman MacLennan, Jenner Institute, University of Oxford (UK)

Professor Jan Jacobs, Clinical Sciences Department, Institute of Tropical Medicine Antwerp (Belgium)

Dr Neil Ravenscroft, University of Cape Town (South Africa)

Dr Paola Cescutti, University of Trieste (Italy)

Professor Octavie Lunguya, National Institute for Biomedical Research (INRB), Kinshasa (Democratic Republic of the Congo)

Summary

In sub-Saharan Africa, invasive non-typhoidal Salmonella (iNTS) is the major cause of bacterial bloodstream infections among young children and disease management is jeopardised by increasing antimicrobial resistance (AMR). The O-antigen portion of Salmonella lipopolysaccharide (LPS) is recognised as key target antigen for protective immunity and O-antigen-based vaccines covering the main serovars Salmonella Typhimurium and Enteritidis are in development. Some of the vaccine candidates are about to enter phase 1 clinical trials; however, efficacy in Africa will not be tested for several years.

O-antigen structural variability can have an impact on the protective immunity of corresponding vaccines. Serotyping and genomic investigation of recent iNTS isolates from the Democratic Republic of the Congo (DRC) have shown increasing rates of iNTS isolates with variation in O-antigen structure. In particular, more than 45 % of the recent Salmonella Typhimurium isolates do not present O:5 specificity, associated to O-antigen O-acetylation.

In this project, we will analyse the genomic variation of O-antigen of Salmonella Typhimurium DRC isolates within the African context. The genomic basis of differences in O-antigenic structure will be proven by mutagenesis experiments. We will determine the O-antigen structure from a panel of Salmonella Typhimurium isolates recently collected in DRC, ascertaining the nature of the O-antigen genomic variations. The coverage of current O-antigen based vaccines against iNTS is likely to be impacted by the O-antigen structural variability, and this project will yield key insights on how to improve the current vaccines.

Project outcomes

In sub-Saharan Africa, infections by nontyphoidal Salmonella bacteria are one of the most important causes of childhood mortality. Whereas Salmonella is a well-known cause of gastro-intestinal diseases across the world, these bacteria can also cause severe infections of the otherwise sterile bloodstream. This disease predominantly affects malnourished children, or people with malaria and/or HIV-coinfections and is seen a lot among young children under five in sub-Saharan Africa. For up to 20 % of these children the infection is fatal.

Nontyphoidal Salmonella bloodstream infections can be treated with antibiotics, but we see increasing rates of antimicrobial resistance which might jeopardise effective treatment in the near future. For example, recently, an outbreak of extensively drug resistant nontyphoidal Salmonella was seen in DR Congo, which could only be treated with one remaining available antibiotic.

Therefore, preventive measures are highly needed, and a vaccine protecting against nontyphoidal Salmonella bloodstream infections would be most promising. No vaccine is available yet, but different vaccines are currently under development. The bacteria’s O-antigen compound, which is part of the bacterial envelope, is a key target to design these vaccines against.

In the DR Congo, bloodstream surveillance is ongoing since 2007 confirming the importance and high burden of nontyphoidal Salmonella infections in the region. Intriguingly however, an increasing number of Salmonella isolates show variation in their O-antigen structure. A novel, so called, O5 negative variant is increasing in numbers in favor of O5 positive variants. O5 negativity implies lack of acetylation of one particular abequose sugar compound in the O-antigen. Changes in the chemical structure of the O-antigen might impact coverage by a vaccine directed towards the O-antigen. Therefore, in our project, we aimed to understand this variation better.

In this BactiVac Catalyst Project SAL-O5, we used an interdisciplinary approach to study the variation of the O-antigen in the nontyphoidal Salmonella population. Our team included experts in (i) bacterial surveillance in DR Congo, (ii) the chemical structure of the O-antigen, (iii) genomics and population structure of nontyphoidal Salmonella and (iiii) immunologists using state-of-the-art methodologies in these fields.

We observed that the O5 negative variants appeared multiple times in the nontyphoidal Salmonella population, suggesting parallel evolution in the same direction. The chemical analyses confirmed the structural difference between O5 positive and negative isolates, but also showed that there is even further chemical variability of the O-antigen in these DR Congo isolates. It was found that acetylation on different sugar compounds could be variable, and that the O-antigen as a whole could be lost as well. All these chemical variants could be explained by changes observed in the genetic make-up of the Salmonella bacteria.

In summary, we observed substantial levels of O-antigen variation among the invasive Salmonella from bloodstream DR Congo. In our work, we could explain the genetic origin of this chemical variation. As the O-antigen compound is the key target for vaccines, this variation might have implications for coverage of future vaccines.