composition

Abstract

The present invention provides a lipid vesicle formed from a lipid bilayer comprising a sphingolipid, a sterol and a phospholipid, and compositions comprising the same. Methods of diagnosing a bacterial infection and / or detecting bacterial colonisation, comprising contacting a sample obtained from a subject with a lipid vesicle of the invention, or a composition of the invention are also provided. Kits comprising at least one swab and lipid vesicles of the invention, or a composition of the invention are also provided.

Description

[0001] Composition

[0002] This work has received funding from the European Union's Horizon 2020 research and innovation programme under the Marie SKLODOWSKA-CURIE grant agreement No 955664.

[0003] Field of Invention

[0004] The present invention relates to lipid vesicles that may be used in diagnosing infection, especially infection by bacteria such as Streptococcus agalactiae, commonly known as Group B Streptococcus (GBS).

[0005] Background to the Invention

[0006] Streptococcus agalactiae, or Group B Streptococcus (GBS) is a common Grampositive commensal bacterium which can cause serious disease in neonates, such as sepsis and meningitis (Manning et al., 2004; Stephanie J. Schrag et al., 2002). Transmission from mother to child either in utero or during birth can cause neonatal infection (Maisey et al., 2008). Studies conducted in the UK show that between 21.2-29.4% of all pregnant women are colonised with GBS when screened with recto-vaginal swabs (Baltimore et al., 2001; Daniels et al., 2011; Gopal Rao et al., 2017; Jones et al., 2006). Neonatal infection in the first six days, is known as early onset disease (EOD). Septicaemia due to EOD displays as minimal symptoms, such as irritability, lethargy, fever, and abnormal heartrate (Baltimore et al., 2001; Madrid et al., 2017). Neonatal infection in day 7-90 is known as late onset disease (LOD), and cause bacteriaemia and meningitis (Berardi et al., 2021).

[0007] The USA offer maternal screening during gestation week 35-37, where positive mothers are offered intrapartum antibiotic prophylaxis (IAP) (Hughes RG et al., 2017). GBS is a transient coloniser, and screening early in pregnancy can lead to false positives and negatives, a reason why other countries, such as the UK, have been hesitant to adopt screening.

[0008] Current detection systems for GBS include culture dependant methods; Gao et.al. conducted a study in 2021 testing the most common GBS detection systems. The gold standard detection method is using Lim broth subculture with subsequent plating on blood agar, other culture dependant methods include blood agar plate, liquid chromogenic medium and chromogenic agar plate, some of which require pre-enrichment steps. These tests take between 24-48h to conduct. Whilst assays such as the loop-mediated isothermal amplification (LAMP) assay without preenrichment is a far more time efficient test (Gao et al., 2021), it still requires technical skill and expertise to carry out and cannot be performed at the point-of- care.

[0009] Summary of the Invention

[0010] Transient colonisation by Streptococcus agalactiae, or Group B streptococcus (GBS) allows for false test results when testing mothers for vaginal carriage too early in pregnancy. Vaginal carriage of the bacteria can allow for transmission to the neonate during labour and can cause serious neonate infection, and potentially be fatal. Traditional culture methods are time consuming and do not allow for maternal testing during labour. The present inventors have developed a biosensor, utilising lipids with known interactions to GBS virulence factors. The biosensor uses lipid vesicles with a bilayer membrane encapsulating a signalling molecule, which is released when GBS is present, allowing for a visual and measurable change.

[0011] Accordingly, in a first aspect, the present invention provides a lipid vesicle formed from a lipid bilayer comprising a sphingolipid, a sterol and a phospholipid. Preferably the lipid vesicles encapsulate a signalling molecule, such as a dye. The dye may be a hydrophilic dye and / or a self-quenching dye.

[0012] The inventors have demonstrated that the lipid vesicles are specifically sensitive to the presence of GBS. A pilot study has demonstrated sensitivity of 83.3%, specificity of 85.7% and overall accuracy of 85.3%, with the test taking just 45 minutes to deliver a result. This accuracy is in line with current GBS testing and provides additional advantages due to its rapidity and the ability to conduct the test at the point-of-care, e.g., at a patient's bedside during labour.

[0013] Rapid as well as cost-effective and easy-to-use detection systems for GBS as described herein can allow for maternal screening intrapartum, and neonatal testing postpartum in both high income, and low income settings and can prevent transmission and infection to and in the neonate. Utilising a method that requires no prior scientific training can simplify detection and can allow for use at low income and rural locations.

[0014] Description

[0015] Sterols, such as cholesterol, and sphingolipids are major components of the human cell membrane. Sphingolipids compose mammalian cell membranes, and specifically sphingomyelin has shown to be targeted by GBS virulence factors (Bernheimer et al., 1979; van Meer et al., 2008). Sterols and sphingolipids can readily mix with each other, causing shielding of the non-polar sterol by the sphingolipid (Wurm et al., 2018). When utilising vesicles as a biosensor for bacterial detection, the catalyser for this reaction is believed to be bacterial virulence factors, such as CAMP-factors and 0-haemolysin, which are both pore forming toxins, and cause lysis of red blood cells (Ballard et al., 2021; Rosa-Fraile et al., 2014). The present inventors therefore determined that producing a liposome that has similar composition to the human cell membrane could allow for lipid vesicles that should be reactive to Group B Streptococcus (GBS).

[0016] The lipid vesicles are formed from a lipid bilayer comprising a sphingolipid, a sterol and a phospholipid.

[0017] The lipid vesicles may be about 50 to 800 nm in diameter, or about 100 to 600 nm in diameter, or about 100 to about 400 nm in diameter. Optionally the lipid vesicles may be about 200 nm in diameter. Very large vesicles, such as those with a diameter of lOOOnm or more, may become unstable due to their large structure.

[0018] The lipid vesicles may comprise about 20-80 mol% or about 30-70 mol% or about 40-60 mol% sphingolipid. Optionally, the lipid vesicles may comprise about 50 mol% sphingolipid or about 50-60 mol% or about 50-70% sphingolipid.

[0019] Preferably the sphingolipid is sphingomyelin. Sphingomyelin consists of a phosphocholine head group, a sphingosine, and a fatty acid and can be classified as a sphingophospholipid. The sphingomyelin may be naturally derived or artificially derived. Preferably, the sphingomyelin is naturally derived and may be of avian, bovine or porcine origin. Preferably the sphingomyelin is brain derived sphingomyelin. More preferably the sphingomyelin is porcine brain derived sphingomyelin. The sphingomyelin may have the following structure:

[0020] The lipid vesicles may comprise about 10-40 mol% or about 10-30 mol% or about 15-25 mol% sterol. Optionally, the lipid vesicles may comprise about 20 mol% sterol or about 20-25 mol% or about 20-30 mol% sterol.

[0021] Preferably the sterol is cholesterol. Cholesterol is a major component of lipids found in human cell membranes. It is known to increase lipid-packing density, maintain fluidity and lower membrane permeability. Cholesterol can also have a chemoprotective function in lipid membranes.

[0022] The lipid vesicle may comprise about 0.5-15 mol% or about 1-10 mol% or about 3-7 mol% phospholipid. Optionally, the lipid vesicle may comprise about 5 mol% phospholipid, or about 5-7 mol% phospholipid or about 5-10 mol% phospholipid.

[0023] Preferably the phospholipid is l,2-dipalmitoyl-sn-glycero-3-phosphocholine (DPPC). DPPC is a zwitterionic glycerophospholipid that is commonly used in the generation of micelles and liposomes.

[0024] The lipid vesicle may further comprise 10,12-tricosadiynoic acid (TCDA). TCDA is a long chain fatty acid, which has been shown to provide a stabilising effect on liposomes.

[0025] The lipid vesicle may comprise about 10-45 mol% or 15-40 mol% or about 20-30 mol% TCDA. Optionally, the lipid vesicle may comprise about 25 mol% TCDA or about 25-30 mol% or about 25-40 mol % TCDA.

[0026] Concentrations are preferably % mol by volume. The lipid vesicle may encapsulate a signalling molecule. Any appropriate signalling molecule may be used. The signalling molecule is used to indicate rupture or lysis of the lipid vesicle, the signal changing or display of the signal changing when the vesicle is ruptured or lysed. For example, the signalling molecule may be a dye or otherwise visualisable signalling molecule that can only be visualised when the vesicle is ruptured. The signal may be as simple as a colour being visible when vesicles are ruptured, such dyes are well known. Or it may be, for example, a fluorescent signal that requires illumination with UV light to be visible. Suitable dyes may be hydrophilic and / or self-quenching. Preferably the signalling molecule is a fluorescent dye, such as one or more of fluorescein, 5(6)-carboxyfluorescein and sulforhodamine B.

[0027] Other possible signalling molecules include organic and inorganic dyes such as pH sensitive dyes such as crystal violet, bromothymol blue, methyl orange, calcein and other pH or concentration sensitive fluorometric dyes. Alternatively, other chemical signalling molecules, such as redox signalling molecules, may be used.

[0028] Dye encapsulated within the lipid vesicle may be at a concentration of about 1- 100 mM.

[0029] The lipid vesicles may additionally or alternatively encapsulate one or more other molecules, especially active molecules, such as antibiotics, antimicrobials, bacteriophages or other biologically originating antimicrobials, and other biologically active agents. Preferably, the additional or alternative one or more other encapsulated molecules are not membrane active.

[0030] In some embodiments the lipid vesicle consists essentially of sphingolipid, cholesterol, DPPC and TCDA. In other words, the lipid vesicle can be formed from these four components with no other components being present in the lipid bilayer. Other substances, such as a signalling molecule, may still be encapsulated within the lipid vesicle as described above. The lipid vesicle may consist essentially of about 50 mol% sphingomyelin, about 20 mol% cholesterol, about 25 mol% TCDA and about 5 mol% DPPC.

[0031] The lipid vesicles are preferably sensitive to the presence of bacteria, meaning that the lipid vesicles may be lysed or ruptured by the presence of the bacteria. As the skilled person will appreciate, bacteria and especially pathogenic bacteria, release virulence factors such as toxins, which can cause tissue damage. They can also secrete enzymes and toxins that damage cells and connective matrix materials of host tissues. Many of these toxins and enzymes act as cell membrane-lysing and / or cell membrane permeating agents. The presence of specific virulence factors can therefore lead to lysis or rupture of lipid vesicles. However, the sensitivity and specificity of lipid vesicles to different types of virulence factors is dependent on the composition of the lipid bilayer and must also be balanced with a need for stability. The lipid vesicles are preferably sensitive the presence of Group B Streptococcus (GBS) bacteria, such as Streptococcus agalactiae. Without being bound by theory, the present inventors believe that the lipid vesicles are lysed by the virulence factors p-haemolysin and CAMP-factor, which are both pore forming toxins released by GBS bacteria.

[0032] The vesicles can be produced by dissolving a desired lipid mixture in a solvent, such as chloroform. The dissolved lipid mixture is then dried, forming a phospholipid film. The phospholipid film can then be dissolved in a dye solution, followed by heating (e.g., heating to about 50°C in a water bath for about 10 minutes) and chilling to frozen (e.g., using liquid nitrogen) steps. The vesicles can then be formed by extrusion using a liposome extruder and can be size separated, e.g., using a gel column. Vesicle extrusion may comprise extruding the vesicles through one or more porous membranes (e.g., a porous polycarbonate membrane). For example, vesicle extrusion may include extruding the vesicles through a 200nm pore diameter porous membrane.

[0033] A composition comprising the lipid vesicles is also provided. The composition may comprise vesicles suspended in saline or in a buffer. The buffer may be a phosphate buffer, such as phosphate buffered saline (PBS), or a bicarbonate buffer or a Good's buffer. For example, the buffer can be a piperazine containing Good's buffer, such as one or more of 4-(2-hydroxyethyl)-l- piperazineethanesulfonic acid (HEPES), piperazine-N,N'-bis(2-ethanesulfonic acid) (PIPES), piperazine-l,4-bis(2-hydroxy-3-propanesulfonic acid), dihydrate (POPSO) and 3-[4-(2-Hydroxyethyl)piperazin-l-yl]propane-l-sulfonic acid (EPPS). Preferably the buffer is HEPES.

[0034] The composition is preferably sterile. The lipid vesicles or composition may be used in therapy. For example, the lipid vesicles or composition may be used in the diagnosis of an infection, preferably a bacterial infection. The lipid vesicles or composition may also be used in detecting bacterial colonisation. The bacterial infection or colonisation may be a Group B Streptococcus infection or colonisation. Bacterial colonisation means that bacteria may be present on or in a host but are not causing disease or generating an immune response. A bacterial infection typically means that the presence of bacteria is causing disease and / or an immune response in the host.

[0035] The present invention also provides a method of diagnosing a bacterial infection and / or detecting bacterial colonisation, the method comprising: contacting a sample obtained from a subject with a lipid vesicle encapsulating a signalling molecule or a composition comprising the lipid vesicles; and monitoring for release of the signalling molecule, wherein release of the signalling molecule indicates an infection and / or bacterial colonisation.

[0036] Contacting the sample with the lipid vesicle or composition typically comprises incubating the sample with the lipid vesicle or composition comprising the lipid vesicles at a temperature of about 30°C to about 60°C, preferably about 46°C.

[0037] The lipid vesicles may be at a volume of about 0.1-200 pl or about 50-150 pl or about 100 pl.

[0038] The lipid vesicle or composition comprising the lipid vesicles may be diluted with a water-soluble solution, such as saline or buffer. The buffer may be a phosphate buffer (such as phosphate buffered saline) or a bicarbonate buffer or a Good's buffer. For example, the buffer can be a piperazine containing Good's buffer, such as one or more of 4-(2-hydroxyethyl)-l-piperazineethanesulfonic acid (HEPES), piperazine-N,N'-bis(2-ethanesulfonic acid) (PIPES), piperazine-l,4-bis(2-hydroxy- 3-propanesulfonic acid), dihydrate (POPSO) and 3-[4-(2-Hydroxyethyl)piperazin- l-yl]propane-l-sulfonic acid (EPPS). Preferably the buffer is HEPES.

[0039] Optionally, the lipid vesicle or composition comprising the lipid vesicles may be diluted with about 0.1-1000 pl of the water-soluble solution. The lipid vesicle or composition may be diluted with the water-soluble solution at a ratio of 3: 1 or 6: 1 or 7: 1 or 8: 1. A signal from the signalling molecule is a sign or indicator that the lipid vesicle has been ruptured or lysed and the signalling molecule released. For example, it could be a colour or fluorescence change, or electrochemical signal. Release of the signalling molecule preferably provides a colour change that can be observed visually.

[0040] Preferably the bacterial infection is a Group B Streptococcus (GBS) infection, such as a Streptococcus agalactiae infection.

[0041] The time from contacting the sample with a lipid vesicle to detecting release of the signalling molecule may be 90 minutes or less, or 60 minutes or less, or 45 minutes or less. The minimum time from contacting the sample may be 5 minutes or more, or 10 minutes or more or 20 minutes or more.

[0042] The sample may be any biological sample obtained from a mammal including humans. In embodiments of the invention the sample may be a sample of bodily fluid, such as urine, mucus or faecal matter. The sample may be obtained from a bodily orifice, such as the nose, mouth, throat, ears, vagina, urethra or anus. Additionally or alternatively, the sample may be obtained from the skin, such as the epidermis or from mucosa. Preferably the sample is obtained from the nose and / or mouth and / or skin of an infant or from the vagina of a pregnant individual.

[0043] The sample obtained from the subject may be a swab. The swab may be an oral swab, a nasal swab, a skin swab or a vaginal swab. For example, when the subject is a pregnant individual the sample may be a vaginal swab. Alternatively, when the subject is an infant the sample may be one or more of an oral swab and a nasal swab.

[0044] The subject may be a pregnant individual. The sample may be obtained before onset of labour, during labour and / or up to 24 hours after giving birth. Alternatively, the subject may be an infant. The infant may from birth up to three months of age.

[0045] In embodiments of the invention the method may further comprise administering a suitable treatment for the bacterial infection. In the event that a bacterial infection is identified in a pregnant individual the treatment may be administered to said individual and / or the treatment may be administered to the infant. Optionally, the treatment may be an antibiotic treatment. Additionally, or alternatively, the treatment may be a prophylactic treatment. For example, the treatment may be an intrapartum antibiotic prophylaxis.

[0046] The present invention additionally provides a kit comprising at least one swab and lipid vesicles or composition comprising vesicles as described herein. Typically, the at least one swab and the lipid vesicles or composition comprising vesicles are packaged together to form the kit.

[0047] The kit may further comprise instructions for use. Preferably the kit is suitable for use both in the home and in a clinical setting. Typically, the kit is portable and / or sterile.

[0048] The swabs may be conventionally known and commercially available swabs, such as flocked swabs or cotton wool swabs. Flocked swabs are typically comprise a brush-like tip formed of artificial fibres, such as nylon. Preferably the swab is a flocked swab. The present inventors have observed that flocked swabs can provide increased uptake of sample compared to cotton swabs.

[0049] The kit may further comprise one or more active therapeutic agents for the treatment of a bacterial infection or colonisation detected by the kit. For example, the one or more active therapeutic agents may be an antibiotic.

[0050] Brief Description of the Drawings

[0051] The invention will now be described in detail, by way of example only, with reference to the figures.

[0052] Figure 1. Shows a schematic of a biosensor mode of action. Vaginal swab from mother, or neonatal swab taken either in the nose or pharynx is introduced into vesicle solution. After 45 minutes, GBS virulence factors have caused lysis of vesicles allowing for release of fluorescent 5(6)-carboxyfluorescein.

[0053] Figure 2. (A) Shows the fluorescence released by vesicles lysed by GBS COH1 in culture (green) and the absorbance (OD600nm) of the bacteria at the same time points (red) during planktonic growth. Error bars show the standard error of the mean (SEM). (B) Shows normalised data, AFU (absolute fluorescence units) over 1000 CFU for single species biofilm, error bars show standard deviation GBS strains 523, 6073, KX0150, NPX0066, NPX0090 and SGX0300 as well as E. faecalis are clinical strains. (C) Shows fluorescence after incubation under UV light .

[0054] Figure 3. (A) Biosensor fluorescence response for 58 participants in study. Green shows true GBS negative samples (using ECM) and red shows GBS positive samples (using ECM). The threshold (y=4.90) is calculated based on the mean of the average y-value for the positive and negative samples. Values over the threshold are deemed GBS positive while samples under the threshold are deemed GBS negative using the biosensor test. (B) Images of donated swabs incubated with vesicles under UV light are shown.

[0055] Figure 4. Shows a Standards for Reporting of Diagnostic Accuracy (STARD) flow diagram, detailing participation, and results from donated samples.

[0056] Figure 5. Shows AFU relative to negative control for GBS type strains (18RS21, 2603V / R, 515, ATCC 12386, COH1, H36B, NCTC8181) and one clinical strain from St George's Hospital, London, SGX0083 in red, along with C. albicans ATCC 60193 (green), E. coli DH5o (orange) and P. aeruginosa PAO1 (blue). Biofilm membranes have been incubated with either 43 / 20, 50 / 25 or 50 / 20 vesicles for 45 min at 46°C.

[0057] Figure 6a-b. Shows lipid vesicles at 200nm using TEM. Vesicles of composition 50% sphingomyelin, 20% cholesterol, 25% TCDA and 5% DPPC were used. Vesicles stained with uranyl acetate on a glow discharged, carbon-formvar grid, and was imaged with a Jeol 2100plus TEM using 200kV acceleration voltage.

[0058] Examples

[0059] Materials and methods:

[0060] Phospholipid vesicle preparation:

[0061] Phospholipid vesicles were produced using extrusion with an encapsulated selfquenched 5,6-carboxyfluorescein dye, formulated as thus, brain sphingomyelin (SM, porcine) (860062P-1G, Avanti Polar Lipids), l,2-dipalmitoyl-sn-glycero-3- phosphocholine (DPPC) (850355P, Avanti Polar Lipids), 1,2-dipalmitoyl-sn- glycero-3-phosphoethanolamine (DPPE) (850705P, Avanti Polar Lipids), cholesterol (Cho) (C8667, Sigma Aldrich) and 10,12-tricosadiynoic acid (TCDA) (91445, Sigma Aldrich) were used in varying concentrations to make lipid vesicles (Table I).

[0062] Table I Lipid composition of vesicles (%)

[0063] Name Sphingomyelin Cholesterol TCDA DPPC DPPE

[0064] 43 / 20 43 20 25 10 2

[0065] 50 / 20 50 20 25 5 0

[0066] 50 / 25 50 25 20 5 0

[0067] Chloroform was added to phospholipid mixture and vortexed to fully dissolve phospholipids. The phospholipid and fatty acid mixture was dried completely to form a phospholipid film. Carboxyfluorescein was prepared by dissolving 1.877g 5(6)-carboxyfluorescein, 58mg NaCI, 540mg NaOH and 238mg HEPES (4-(2- hydroxyethyl)-l-piperazineethanesulfonic acid), in 100ml MiliQ water, and sterilised Phospholipid film was dissolved in 10ml dye solution by vortex. Dissolved phospholipids were heated to 50°C in a water bath for 10 minutes, and then chilled in LN2 until completely frozen.

[0068] Liposofast extruder (Avestin) was used to produce 200nm phospholipid vesicles. Vesicles were separated by size using NAP-25 gel columns, vesicles were loaded to each column, and HEPES buffer was added to separate vesicles. HEPES buffer was made by dissolving 6.24g NaCI, 2.382g HEPES and 224mg NaOH in IL MiliQ water and sterilising by heating to 121°C for 15 min.

[0069] In vitro single species biofilm:

[0070] To test the specificity of the vesicles to GBS, in vitro biofilms were incubated with vesicle solution. As this would more closely represent GBS growing in the vaginal tract than planktonic cells. Clinical samples of GBS, 523, 8071, KX0150, NPX0066, NPX0090, SGX0300 collected at St George's Hospital, London, Candida albicans ATCC 60193, Enterococcus faecalis (clinical strain from St George's Hospital, London), Staphylococcus epidermidis RP62A were all grown as single species in vitro biofilms. Cellulose membranes cut to 6mm disc were grown up with liquid cultures of each bacterium in Todd-Hewitt Broth supplemented with 1% D-glucose at 37°C, 120 rotations per minute (RPM) for 24h. After incubation, membranes were aseptically removed, washed in IX PBS and placed on sterile 200nm pore membranes on Colombia Blood Agar (CBA) supplemented with 1% D-glucose and incubated at 37°C for 24h. After a total of 48h incubation, vesicles were prepared in a 3: 1 ratio with HEPES buffer, to a total volume of lOOpI, and biofilm membranes were added to vesicle solution, and incubated for 46 min at 46°C. After incubation, vesicles were further diluted with HEPES and fluorescence intensity was measured using the SpectroOmega (BMG Labtech) at 520nm emission, 485nm excitation. Biofilm membrane of each species was added to 2ml IX PBS and sonicated for 5min to strip membrane and was then serially diluted and plated using the Miles Misra method on CBA plates to determine the colony forming units (CFU) per membrane.

[0071] Vesicle fluorescence response to GBS:

[0072] To measure the vesicle response as a result to planktonic GBS growth 2-3 colonies of GBS strain COH1 was suspended in 6ml THB with 1% glucose, and vortexed. Vesicle solution was prepared as stated previously, and mixed 1: 1 with prepared overnight, 150pl was added to a U-bottomed 96-well plate in triplicate. Using the SpectroOmega (BMG Labtech), fluorescence was measured every 5 min for 16h at 37°C using same parameters as before, and the absorbance was measured in the same fashion at 600nm, with shaking occurring between measurements to prevent pellet formation. Vesicles with media served as negative control for fluorescence read, and media only served as negative control for absorbance.

[0073] Ex vivo testing using anonymous donated swabs:

[0074] To obtain data on the utility of the test in the community, a healthy volunteer study was conducted at the University of Bath, with participants invited to donate two lower vaginal swabs. Purpose of this study was to obtain preliminary sensitivity, specificity and accuracy data of the test when compared to the Enriched Culture Media (ECM) test used as a current gold standard. Ethical approval for using donated vaginal swabs from anonymous participants was approved by the Research Ethics Approval Committee for Health (REACH), at the University of Bath, UK, REACH No. EP 22 072. Individuals who are biologically female, over 18 years, currently not pregnant, and physically able to swab themselves and give informed consent were recruited, n = 58. Participants were allowed to participate more than once, provided it was at least 30 days between donations. Participants donated two swabs (MW821, MWE), each inserted approx. 2-3cm into the vaginal canal. Participants were asked to place their swabs in sterile 15ml falcon tubes and place their samples in a designated fridge at 4°C, to remain anonymity, swabs were collected at certain times during the day (mean time before testing, 102 min). One swab was used for traditional enriched culture medium detection, using overnight culture in Trans-Vag Broth (TVB), and culture on CBA, GBS chromogenic agar as well as Aesculin agar for detection of Enterococcus spp. Colonies displaying orange colonies on GBS chromogenic agar, and were GBS latex positive using the Thermo ScientificTM Streptococcal Grouping Kit (conducted according to manufacturer's instructions), were determined to be GBS positive samples. The other swab was immersed in lOOpI vesicle solution, incubated for 45min at 46°C, and was after incubation diluted with 400pl HEPES, and the fluorescence intensity was measured as previously stated. Test specificity and sensitivity was based on a threshold set relative to the fluorescence intensity based on the mean of the average of the positive and negative samples and was compared to the microbiological analysis.

[0075] Statistical analysis:

[0076] Statistical analysis for standard deviation of the mean of samples has been calculated using Graphpad version 10.0.0. Statistical analysis of sensitivity, specificity, accuracy, positive predictive value, negative predictive value, falsepositive rate, and false-negative rate has been calculated using standard equations for each (detailed in Table II).

[0077] Results and discussion:

[0078] Laboratory validation of test efficacy:

[0079] Preliminary work used a lower concentration of sphingomyelin, and experiments comparing lower sphingomyelin content, to a higher one showed increased specificity for GBS when vesicles contained a higher sphingomyelin content (50%). Vesicle integrity was also observed to be more reliable at higher temperatures (46°C), when there was a higher sphingomyelin and TODA content (Figure 5). Henceforth, a concentration of 50% sphingomyelin, 20% cholesterol, 25% TCDA and 5% DPPC was used, this is referred to as 50 / 20 vesicles. To determine size and shape of vesicles TEM was performed (Figure 6a-b), showing uniform size, measuring at 200nm in diameter.

[0080] Preparing an overnight with GBS COH1 to be grown up along with vesicles whilst simultaneously measure the fluorescence and absorbance, allowed for a vesicle response curve to be formed. After 7h of incubation the fluorescence starts to increase for vesicles that have been incubated with COH1, and around 13h a plateau starts (Figure 2A). The absorbance increases at 4h and reaches a plateau at 8h. The delayed response in fluorescence could indicate that a system such as quorum sensing is responsible for the release of virulence factors and subsequent lysis of the vesicles. The low sensitivity shown in Figure 2A for GBS to lyse the vesicles is when the bacteria is in a planktonic culture, and not in a biofilm, and does not represent how GBS grows in the vaginal tract.

[0081] Determination of sensitivity of vesicles to GBS as well as C. albicans, E. faecalis and S. epidermidis was studied using a novel biofilm method. Incubation of vesicles along with single species biofilms of each strain allowed for comparative analysis of fluorescence intensity in relation to microbe load per membrane (Figure 2B). C. albicans, E. faecalis and S. epidermidis were all analysed along GBS to determine sensitivity of vesicles to likely contaminants present in samples. As they are likely to be present either in the vagina or on the skin. All six clinical samples of GBS showed high fluorescence when normalised to 1000 CFU, varying between 2-8 AFU / 1000 CFU, whilst C. albicans, E. faecalis and S. epidermidis was all below 0.5 AFU / 1000 CFU. When visualising the vesicles after incubation with biofilm membranes, all show lysis to certain extent (Figure 2C). Normalising the fluorescence based on microbe load allows to determine specificity of vesicles.

[0082] Pilot study using healthy human volunteers

[0083] To establish that the vesicles are sensitive and specific to GBS, donated human vaginal swabs were tested on the vesicles. Ethics approval for human vaginal samples was approved by the REACH committee at the University of Bath (No EP 22 072). Participants recruited were biologically female, over 18 years old, physically able to swab themselves and able to consent. Study group was N = 60, a total of which n = 58 met the criteria. Participants were allowed to participate more than once but were asked to leave 30 days in-between donations. A threshold was set at y=4.90, based on the mean of the average y-values of the positive and negative samples (6.32 and 3.48 respectively). Depending on the relative value of each sample to the negative control, values over the threshold indicate a positive GBS sample using the biosensor, and any value below indicate a negative GBS sample (Figure 3A). With additional testing, the threshold may fluctuate. Red bars in Figure 3a shows true GBS positive samples using the ECM test, and samples in green show true negative samples. After incubation, swabs were observed under UV light (Figure 3B) and change in fluorescence intensity could visually be observed between the swab shown later to contain GBS, compared to swabs shown to lack GBS.

[0084] A total of 58 participants donated swabs. Two (n = 2) were excluded from the study due to living too far away to deliver their samples to the lab anonymously. The overall success rate of participation in the study was at 96.6%, GBS carriage rate was determined to be 17.2%, which is on par with literature values for carriage rates of pregnant individuals (Russell et al., 2017). Using the biosensor point-of-care (POC) test, 16 samples were determined to be GBS positive, and 42 samples to be GBS negative. Using the gold standard ECM test, 10 samples were determined to be GBS positive, and 48 samples were determined to be GBS negative. Utilising this data allowed for evaluating the sensitivity and specificity of the test and was established to be 83.3% and 85.7% respectively. Overall accuracy of the study was 85.3% (Fig. 4 and Table II).

[0085] The positive predictive rate was determined to be 55.5% and the negative predictive value to be 96.0% (Table II). This shows that the POC test has a high probability of correctly predict GBS negative carriers. The false positive rate was determined to be 14.3% and the false negative rate was determined to be 16.7%.

[0086] Table II: Shows statistical type, equation, and the percentage of each. TP = true positive (10), TN = true negative (48), FP = false positive (8) and FN = false negative (2).

[0087] Statistic type Equation Percentage

[0088] Sensitivity TP / (TP+FN) 83.3% (10 / 12)

[0089] Specificity TN / (FP+TN) 85.7% (48 / 56)

[0090] Positive predictive value TP / (TP+FP) 55.5% (10 / 18)

[0091] Negative predictive value TN / (FN+TN) 96.0% (48 / 50) False-positive rate FP / (FP+TN) 14.3% (8 / 56)

[0092] False-negative rate FN / FN+TP) 16.7% (2 / 12) Accuracy (TP+TN) / (TP+TN + FP+FN) 85.3% (58 / 68)

[0093] Lipid vesicles have comparable efficacy to current detection methods

[0094] Liposomes have shown potential in many aspects of science, such as utilising them as a mode of drug delivery. Previous work by our group have shown vesicles made to specifically be lysed by wound infections caused by common wound pathogens. Utilising known interactions between bacteria and specific components of human cell membranes, allows to produce specific liposomes that are uniquely designed to be targeted by specific pathogens. Sphingomyelin is a lipid known for interacting with GBS virulence factors. The exact mechanism of action of lysis of our liposomes is still unknown. The working hypothesis is that both pore forming toxins work on the lysis of these vesicles.

[0095] Attempting to determine the vesicle lysis response in relation to bacterial load was shown by allowing the bacteria to be cultured in THB with the vesicles. This allowed for an understanding in the correlation between the fluorescence intensity and the bacterial optical density. As seen in Fig. 2A, the OD increases before the fluorescence is increased. Why this delay occurs is something that needs to be further explored but could be explained by a quorum sensing mechanism.

[0096] Making a lipid vesicle that represents the human cell membrane would be expected to have cross reactivity with other bacteria. When utilising a novel in vitro biofilm model, where the biofilm was allowed to form on a cellulose membrane, it was noted that other bacteria did cause lysis of the vesicles, but only at much higher bacterial loads. Data had shown that E. faecalis caused lysis of the vesicles, and when testing human samples this was noted as one of the major limitations. When conducting the human study, the enriched media test was expanded to culture the swabs on Aesculin agar to see if Enterococci species were present. The majority of all samples contained Enterococci, and this was later confirmed by MALDI-TOF analysis. When looking at the results from the donated vaginal swabs, it can be noted that Enterococci spp. did not have this unwanted effect. Using donated lower vaginal swabs allowed for establishing a threshold relative to the mean fluorescence intensity of the average value of the positive and negative samples. Samples above the threshold indicate GBS positive samples using the POC test and samples below indicate GBS negative samples using the POC test. Again, establishing the exact mode of action between vesicles and GBS would allow for fine tuning the vesicles, and allow for the uncertainty between a positive and negative swab to be minimised, and for the threshold to move. Overall, this novel POC test has a high specificity, sensitivity, and accuracy, directly comparable to other methods available for GBS testing (Gao et al., 2021). When looking at data presented by Gao et.al. in 2021 and comparing the sensitivity, specificity and accuracy with other methods which do not require a preenrichment step, our test produced similar results as presented in that study (Table III). Incubation of the test is limited to 45 minutes, which results in a similar turnaround time as the LAMP assay. The limitation of the LAMP assay is that it cannot be done bedside and requires high level equipment and trained personnel. The biosensor test can be done at bedside, without any specific training, and with limited materials. This allows for rapid detection of carriage and is when compared to standard techniques overwhelmingly more time saving. Improving on techniques would allow for this test to be used in low-income settings and in small practices, where mothers could routinely be screened intrapartum, to detect mothers and neonates at risk when it matters the most. Knowing the maternal carriage status at the point of labour would further allow for a more informed choice in prescribing antibiotics to prevent infection to mother and child and would enhance antibiotic stewardship.

[0097] >

[0098] Methods Sensitivity Specificity Accuracy Turnaround Can be

[0099] (without time (h) done pre- bedside enrichment)

[0100] Blood agar 81.5% 100% 98.3% 24-48 No plate

[0101] GBS antigen 18.5% 89.4% 83.2% 0.5-1 No detection (Weimi)

[0102] GBS antigen 21.7% 74% 67.2% 0.5-1 No detection (Huaao)

[0103] Liquid 71.1% 98.1% 96.0% 24-48 No chromogenic medium

[0104] LAMP 100% 94.0% 94.9% 0.5-1 No

[0105] (HiberGene)

[0106] Vesicle 83.3% 85.7% 85.3% 1 Yes assay (PoC) Table III: Shows data on sensitivity, specificity, accuracy, and turnaround time for different GBS detection methods without pre-enrichment steps, adapted from Gao et.al, with the addition of data on POC test and the possibility to conduct the different tests at bedside.

[0107] Conclusion:

[0108] We have presented the development and testing of a novel biosensor composed of phospholipid vesicles reactive to GBS virulence factors for detecting vaginal carriage in a non-pregnant population. The test shows high sensitivity, specificity and accuracy, and is directly comparable to traditional methods for GBS detection, with the benefit of being able to be conducted at bedside. Vaginal swabs will continue to be tested on the biosensor test to continuously evaluate the efficacy, and testing using swabs from a pregnant individuals will be done to determine the efficacy in this population.

[0109] References

[0110] Ballard, M. B., Mercado-Evans, V., Marunde, M. G., Nwanosike, H., Zulk, J., & Patras, K. A. (2021). Group B Streptococcus CAMP Factor Does Not Contribute to Interactions with the Vaginal Epithelium and Is Dispensable for Vaginal Colonization in Mice, https: / / journals.asm.org / journal / spectrum

[0111] Baltimore, R. S., Huie, S. M., Meek, J. I., Schuchat, A., & O'brien, K. L. (2001). Early-Onset Neonatal Sepsis in the Era of Group B Streptococcal Prevention. Pediatrics, 108, 1094-1098. www.aappublications.org / news

[0112] Berardi, A., Trevisani, V., Di Caprio, A., Bua, J., China, M., Perrone, B., Pagano, R., Lucaccioni, L., Fanaro, S., lughetti, L., Lugli, L., & Creti, R. (2021). Understanding factors in group b streptococcus late-onset disease. In Infection and Drug Resistance (Vol. 14, pp. 3207-3218). Dove Medical Press Ltd. https: / / doi.org / 10.2147 / IDR.S291511

[0113] Bernheimer, A. W., Linder, R., & Avigad, L. S. (1979). Nature and Mechanism of Action of the CAMP Protein of Group B Streptococci. In INFECTION AND IMMUNITY (Vol. 23, Issue 3). https: / / journals.asm.org / journal / iai Daniels, J., Gray, J., Pattison, H., Gray, R., Hills, R., & Khan, K. (2011). Intrapartum tests for group B streptococcus: Accuracy and acceptability of screening. BJOG: An International Journal of Obstetrics and Gynaecology, 118(2), 257-265. https: / / doi.Org / 10.llll / j.1471-0528.2010.02725.x

[0114] Gao, K., Deng, Q., Huang, L., Chang, C. Y., Zhong, H., Xie, Y., Guan, X., & Liu, H. (2021). Diagnostic Performance of Various Methodologies for Group B Streptococcus Screening in Pregnant Woman in China. Frontiers in Cellular and Infection Microbiology, 11. https: / / doi.org / 10.3389 / fcimb.2021.651968

[0115] Gopal Rao, G., Nartey, G., Mcaree, T., Reilly, A. O. ', Hiles, S., Lee, T., Wallace, S., Batura, R., Khanna, P., Abbas, H., Tilsed, C., Nicholl, R., Lamagni, T., & Bassett, P. (2017). Outcome of a screening programme for the prevention of neonatal invasive early-onset group B Streptococcus infection in a UK maternity unit: an observational study. Open, 7, 14634. https: / / doi.org / 10.1136 / bmjopen- 2016

[0116] Hughes RG, Brocklehurst P, Steer PJ, Heath P, & Stenson BM. (2017). Prevention of Early-onset Neonatal Group B Streptococcal Disease: Green-top Guideline No. 36. BJOG: An International Journal of Obstetrics and Gynaecology, 124(12), e280-e305. https: / / doi.org / 10.llll / 1471-0528.14821

[0117] Jones, N., Oliver, K., Jones, Y., Haines, A., & Crook, D. (2006). Carriage of group B streptococcus in pregnant women from Oxford, UK. Journal of Clinical Pathology, 59(4), 363-366. https: / / doi.org / 10.1136 / jcp.2005.029058

[0118] Madrid, L., Seale, A. C., Kohli-Lynch, M., Edmond, K. M., Lawn, J. E., Heath, P. T., Madhi, S. A., Baker, C. J., Bartlett, L., Cutland, C., Gravett, M. G., Ip, M., Le Doare, K., Rubens, C. E., Saha, S. K., Sobanjo-Ter Meulen, A., Vekemans, J., & Schrag, S. (2017). Infant Group B Streptococcal Disease Incidence and Serotypes Worldwide: Systematic Review and Meta-analyses. Clinical Infectious Diseases, 65, S160-S172. https: / / doi.org / 10.1093 / cid / cix656

[0119] Maisey, H. C., Doran, K. S., & Nizet, V. (2008). Recent advances in understanding the molecular basis of group B Streptococcus virulence. Expert Reviews in Molecular Medicine, 10(27), 1-16. https: / / doi.org / 10.1017 / S1462399408000811 Manning, S. D., Neighbors, K., Tailman, P. A., Gillespie, B., Marrs, C. F., Borchardt, S. M., Baker, C. J., Pearlman, M. D., & Foxman, B. (2004). Prevalence of Group B Streptococcus Colonization and Potential for Transmission by Casual Contact in Healthy Young Men and Women. Clinical Infectious Diseases, 39, 380- 388. https: / / academic.oup.eom / cid / article / 39 / 3 / 380 / 352961

[0120] Rosa-Fraile, M., Dramsi, S., & Spellerberg, B. (2014). Group B streptococcal haemolysin and pigment, a tale of twins. FEMS Microbiology Review, 38, 932- 946. https: / / doi.org / 10. llll / 1574-6976.12071

[0121] Russell, N. J., Seale, A. C., O'Driscoll, M., O'Sullivan, C., Bianchi-Jassir, F., Gonzalez-Guarin, J., Lawn, J. E., Baker, C. J., Bartlett, L., Cutland, C., Gravett, M. G., Heath, P. T., Le Doare, K., Madhi, S. A., Rubens, C. E., Schrag, S., Sobanjo- Ter Meulen, A., Vekemans, J., Saha, S. K., & Ip, M. (2017). Maternal Colonization with Group B Streptococcus and Serotype Distribution Worldwide: Systematic Review and Meta-analyses. In Clinical Infectious Diseases (Vol. 65, pp. S100- Slll). Oxford University Press, https: / / doi.org / 10.1093 / cid / cix658

[0122] Stephanie J. Schrag, D. P., Elizabeth R. Zell, M. S., Ruth Lynfield, M. D., Aaron Roome, PH. D., Kathryn E. Arnold, M. D., Allen S. Craig, M. D., Lee H. Harrison, M. D., Arthur Reingold, M. D., Karen Stefonek, R. N., Glenda Smith, B. S., Melanie Gamble, M. P. H., & Anne Schuchat, M. D. (2002). A Population-Based Comparison of Strategies to Prevent Early-Onset Group B Streptococcal Disease in Neonates. N Engl J Med, 347(4), 233-239. https: / / doi.org / 10.1056 / NEJMoa020205 van Meer, G., Voelker, D. R., & Feigenson, G. W. (2008). Membrane Lipids: where they are and how they behave. Nature Reviews Molecular Cell Biology, 9, 112- 124. www.nature.com / reviews / molcellbio

[0123] Wurm, F. R., Landfester, K., Rideau, E., Dimova, R., & Schwille, P. (2018). Liposomes and polymersomes: a comparative review towards cell mimicking Liposomes and polymersomes: a comparative review towards cell mimicking. Chem. Soc. Rev, 47, 8572. https: / / doi.org / 10.1039 / c8cs00162f

Claims

Claims1. A lipid vesicle formed from a lipid bilayer comprising a sphingolipid, a sterol and a phospholipid.

2. The lipid vesicle of claim 1, wherein the lipid vesicle is about 50 to 800 nm in diameter.

3. The lipid vesicle of claim 1 or 2, wherein the lipid vesicle comprises 20-80 mol% sphingolipid.

4. The lipid vesicle of any of claims 1 to 3, wherein the lipid vesicle comprises about 50 mol% sphingolipid.

5. The lipid vesicle of any of claims 1 to 4, wherein the sphingolipid is sphingomyelin.

6. The lipid vesicle of any of claims 1 to 5, wherein the lipid vesicle comprises about 10-40 mol% sterol.

7. The lipid vesicle of any of claims 1 to 6, wherein the lipid vesicle comprises about 20 mol% sterol.

8. The lipid vesicle of any of claims 1 to 7, wherein sterol is cholesterol.

9. The lipid vesicle of any of claims 1 to 8, wherein the lipid vesicle comprises about 0.5-15 mol% phospholipid.

10. The lipid vesicle of any of claims 1 to 9, wherein the lipid vesicle comprises about 5 mol% phospholipid.

11. The lipid vesicle of any of claims 1 to 10, wherein the phospholipid is 1,2- dipalmitoyl-sn-glycero-3-phosphocholine (DPPC).

12. The lipid vesicle of any of claims 1 to 11, wherein the lipid vesicle further comprises 10,12-tricosadiynoic acid (TCDA).

13. The lipid vesicle of claim 12, wherein the lipid vesicle comprises 10-45 mol% TCDA.

14. The lipid vesicle of claim 12 or 13, wherein the lipid vesicle comprises about 25 mol% TCDA.

15. The lipid vesicle of any of claims 1 to 14, wherein the lipid vesicle encapsulates a signalling molecule.

16. The lipid vesicle of claim 15, wherein the signalling molecule is a dye.

17. The lipid vesicle of claim 16, wherein the lipid vesicle comprises the dye at a concentration of about 1-100 mM.

18. The lipid vesicle of claim 16 or 17, wherein the dye selected from one or more of fluorescein, 5(6)-carboxyfluorescein and sulforhodamine B.

19. The lipid vesicle of any of claims 1 to 17, wherein the lipid vesicle consists essentially of sphingolipid, cholesterol, DPPC and TCDA.

20. A composition comprising lipid vesicles of any of claims 1 to 19.

21. A lipid vesicle of any of claims 1 to 19 or a composition of claim 20 for use in therapy.

22. A lipid vesicle for use according to claim 21, wherein the lipid vesicle or composition is for use in the diagnosis of a bacterial infection and / or detection of bacterial colonisation.

23. A lipid vesicle for use according to claim 22, wherein the bacterial infection or bacterial colonisation is a Group B Streptococcus infection or colonisation.

24. A method of diagnosing a bacterial infection and / or detecting bacterial colonisation, the method comprising: contacting a sample obtained from a subject with a lipid vesicle of any of claims 1 to 19 or a composition of claim 20; andmonitoring for release of the signalling molecule, wherein release of the signalling molecule indicates an infection and / or bacterial colonisation.

25. The method of claim 24, wherein contacting the sample with the lipid vesicle or composition comprises incubating the sample with the lipid vesicle or composition at a temperature of about 30°C to about 60°C, preferably about 46°C.

26. The method of claim 24 or 25, wherein utilising the lipid vesicle or composition with a sample, where the lipid vesicle or composition is at a volume of about 0.1-200pl.

27. The method of any of claims 24 to 26, wherein utilising the lipid vesicle or composition with a sample where the lipid vesicle or composition is diluted with about 0.1-1000pl water soluble solution.

28. The method of any of claims 24 to 27, wherein release of the signalling molecule provides a visible colour change.

29. The method of any of claims 24 to 28, wherein the bacterial infection is a Group B Streptococcus infection.

30. The method of any of claims 24 to 29, wherein the time from contacting the sample with a lipid vesicle to detecting release of the signalling molecule is 90 minutes or less.

31. The method of any of claims 24 to 30, wherein the sample obtained from the patient is an epithelial or mucosal skin swab.

32. The method of claim 31, wherein the mucosal skin swab is a nasal, oral or vaginal swab.

33. A kit comprising (i) at least one swab, and (ii) lipid vesicles of any of claims 1 to 19 or a composition of claim 20.

Images