Method of endotoxin detection

Abstract

An in vitro method of detecting one or more endotoxins of one or more pathogens in a sample, comprising the steps of a. coating a lipid binding protein on a substrate or capturing a lipid binding protein on a capture molecule immobilized on a substrate, b. contacting the lipid binding protein, thus coated or captured, with the sample, c. detecting whether an endotoxin binds to the lipid binding protein, wherein the one or more endotoxins comprise a lipid A moiety and O- polysaccharide moiety, wherein the lipid binding protein is capable of binding the lipid A moiety of an endotoxin, and characterized in that the lipid binding protein is a mammalian protein and preferably is a murine or human protein.

Description

[0001] TITLE

[0002] METHOD OF ENDOTOXIN DETECTION

[0003] TECHNICAL FIELD

[0004] The present invention relates to methods of detecting endotoxins, quantitatively and qualitatively, and further of identifying specific strains of microorganisms (typing).

[0005] BACKGROUND OF THE INVENTION

[0006] Endotoxins are large molecules that are bacterial toxins found in the outer membrane of Gram-negative bacteria. The minimal structure of endotoxin is composed of glucosamine moieties substituted with a variable number of fatty acid chains and with Kdo (3-deoxy-D- manno-oct-2-ulosonic acid) (the Lipid A moiety) which can be diversly modified by different numbers of lipidic moieties, by phosphoryl groups, by ethanolamine groups, and other groups depending on the source bacterium. Further, Lipid A can be covalently modified by an inner oligosaccharide core, it can be further covalently modified by an outer oligosaccharide core to build Lipooligosaccharides (LOS) as found in Neisseria Meningitidis and Haemophilus influenzae (Moran et al, FEMS Immunology & Medical Microbiology 1996) and it can be further covalently extended by an O- polysaccharide moiety to form a lipopolysaccharide (LPS), derivatives all of which belong to the class of endotoxins and are outer membrane components of gram-negative bacteria (reviewed in Raetz and Withfield Annu. Rev. Biochem. 2002.

[0007] Endotoxins, such as lipid A, lipopolysaccharides and derivatives thereof are potent activators of the human immune system and pyrogenic, i.e., they may cause fever. In severe cases, LPSs can play a role in causing septic shock. In lower levels and over a longer time period, there is evidence that LPSs may play an important and harmful role in autoimmunity, obesity, depression, and cellular senescence.

[0008] It is thus important to be able to quantify lipopolysaccharides, even at very low levels, as a contaminant in pharmaceutical products, such as biopharmaceuticals, medicinal devices, source products and materials for manufacturing processes etc., as even the lowest levels may cause seriously adverse effects in recipients. In addition, for therapeutic applications, the detection and eventually the removal of endotoxins from the blood systems of a patient may be of benefit.

[0009] The standard assay for detecting presence of endotoxin is the Limulus Amebocyte Lysate (LAL) assay, utilizing blood from the Horseshoe crab (Limulus polyphemus). Very low levels of LPSs can cause coagulation of the limulus lysate due to a powerful amplification through an enzymatic cascade. However, due to the dwindling population of horseshoe crabs, and the fact that there are factors in test samples that interfere with the LAL assay (matrix effects), efforts have been made to develop alternative assays, with the most promising ones being ELISA tests using a recombinant version of a protein used in the LAL assay, horseshoe crab Factor C.

[0010] WO0127289 A2 discloses fragments of Factor C which show great potency in recognizing, binding to, neutralizing, and removing endotoxin. These molecules are described as being suitable for anti-microbial, anti-endotoxin, and anti-sepsis therapy. The ability of said fragments of Factor C to bind lipid A of an endotoxin was analyzed using an ELISA-based assay as well as via surface plasmon resonance experiments. Surface plasmon resonance similarly was carried out for recombinant proteins such as SSCrFC-sushi-1 ,2,3-GFP, SSCrFC-sushi-1GFP, and SSCrFC-sushi-3GFP, and confirmed their superior affinity for endotoxin.

[0011] US6719973 B1 discloses a recombinant fragment of Factor C, which show great potency in recognizing, binding to, neutralizing and removing endotoxin. These molecules are described as being suitable for anti-microbial, anti-endotoxin, and anti-sepsis therapy. The ability of SSCrFCES to bind lipid A was analyzed using an ELISA-based assay as well as surface plasmon resonance.

[0012] KR20180020040 A discloses aptamers that have endotoxin affinity and which are fixed on the electrode chip to provide a sensor for endotoxins.

[0013] WO2021260144 A1 discloses the use of an oligomeric protein as a binding agent for binding lipid A of a lipopolysaccharide (LPS), the oligomeric protein having a coiled coil structure comprising at least two monomer peptides, wherein each monomer peptide, which may be the same or different, is capable of forming an a-helix based on the alpha-helical coiled-coil structure that can be found in the yeast transcription factor GCN4. CN 103267745 B discloses a molecular imprinting technology to provide an endotoxin MIP- SPR chip and its preparation method and use.

[0014] LIS20133244224 A1 discloses a method for detection of endotoxin, comprising the steps incubating a sample with a bacteriophage tail protein, and subsequently detecting of endotoxins bound to bacteriophage tail proteins by means of spectroscopic methods, ELISA, chemical or enzymatic detection reaction of endotoxins or cleaved-off endotoxin components, surface plasmon resonance or by means of capacitance measurements. The bacteriophage tail protein may be provided with a His-tag which, via bivalent ions (zinc or nickel) or an antibody specific for it (Qiagen GmbH, Hilden), can bind to a carrier material.

[0015] EP1953227 A1 discloses a peptide exhibiting a lipopolysaccharide and / or lipid A binding activity and further describes verifying the activity via an ELISA test, or a method based on surface plasmon resonance (SPR).

[0016] EP 1450160 A1 discloses a method for measuring endotoxin (ET) that utilizes surface plasmon resonance (SPR) to facilitate the measurement of ET by automation without using expensive reagents. The method uses a sensor in which a substance, such as polymyxin B (PMX), that is capable of specifically adsorbing the ET and is immobilized on a SPR carrier. The used ET sensor can be re-used by putting it into contact with a regenerant that can elute ET trapped on the used ET sensor.

[0017] SUMMARY OF THE INVENTION

[0018] The present invention provides a method of detection, both qualitatively and quantitatively, of endotoxins present in a sample without the need for a reagent sourced from a horseshoe crab species while at the same time providing a method that is highly sensitive to small amounts of endotoxins. Furthermore, in an embodiment of the method according to the present invention, the method further provides a method of detection allowing to exclude or determine the microorganism(s) or bacteria from which the detected endotoxins emanated. Additionally, the method according to the according to the present invention provides a method that more closely emulates the detection of endotoxins in the human body, which is advantageous for evaluating the reaction to endotoxins in the medical context. The present invention thus provides a method of detection that is based on the actual in vivo interactions while at the same time providing a sensitive assay that is capable of detecting endotoxins at concentrations lower than 1 pg / ml. The applicants have found that the lower limit of quantification (LLOQ) ranged in between 0.08 pg / ml and 0.67 pg / ml.

[0019] It is a first object of the present invention to provide an in vitro method of detecting one or more endotoxins of one or more pathogens in a sample, comprising the steps of: a. coating a lipid binding protein directly on a substrate or capturing a lipid binding protein on a capture molecule directly immobilized on a substrate, b. contacting the lipid binding protein, thus coated or captured, with the sample, c. detecting whether an endotoxin binds to the lipid binding protein, wherein the one or more endotoxins comprise a lipid A moiety, or a lipid A moiety and a O- polysaccharide moiety, wherein the lipid binding protein is capable of binding the lipid A moiety of an endotoxin, and characterized in that the lipid binding protein is a mammalian protein, and preferably is a murine or human protein.

[0020] It is a second object of the present invention, to provide a device for use in the in vitro method of detecting one or more endotoxins in a sample according to the first object of the invention, comprising a substrate and a lipid binding protein coated on said substrate or a lipid binding protein captured on a capture molecule immobilized on said substrate.

[0021] Further embodiments of the invention are laid down in the dependent claims.

[0022] BRIEF DESCRIPTION OF THE DRAWINGS

[0023] Preferred embodiments are discussed in the following, and are described via figures meant to exemplify, without limitation, the present invention.

[0024] In the Figures:

[0025] Fig. 1 shows the dose-dependent signal with different concentrations of lipopolysaccharide on the surface plasmon resonance (SPR) chip using murine lipopolysaccharide-binding protein (LBP, C-terminal polyhistidine-tag, SEQ ID9) bound to an immobilized anti-polyhistidine antibody. At ca. 260 s, different concentrations (0, 10, 50, 100 pg / ml) of phenol-extracted lipopolysaccharide (LPS) of S. enterica Typhimurium (Sigma-Aldrich) were added, and resulted in a dose-dependent signal as well as highly stable LPS binding. A LPS dissociation step was initiated at around 380 s with SPR buffer devoid of LPS analyte.

[0026] Fig. 2 shows the dose-dependent signal with different concentrations of lipopolysaccharide on the surface plasmon resonance (SPR) chip using human bactericidal permeability-increasing protein (BPI, N-terminal or C-terminal polyhistidine-tag, SEQ ID4 and ID3), bound to the immobilized anti-polyhistidine antibody. At ca. 260 s, different concentrations (50, 100 pg / ml) of phenol- extracted lipopolysaccharide (LPS) of S. enterica Typhimurium (Sigma-Aldrich) were added, and resulted in a dose-dependent signal as well as stable LPS binding. LPS dissociation step at around 380 s with SPR buffer devoid of LPS analyte.

[0027] Fig. 3 shows the binding of 8 different lipopolysaccharides and three differently modified Lipid A moieties to murine lipopolysaccharide-binding protein (LBP, C- terminal polyhistidine-tag, SEQ ID9) bound to a surface plasmon resonance (SPR) chip. In Fig. 3A, the entire sensorgram including a binding event of lipid- binding protein (0 to about 150 s ) followed by endotoxin association and dissociation step is shown. In Fig. 4B a section of the sensorgram showing only endotoxin analyte binding and dissociation steps is shown. In the Figure 3, measurement curve 1 corresponds to E. coli EH 100 LPS mutant; measurement curve 2 corresponds to S. enterica Minnesota RE595 LPS mutant; measurement curve 3 corresponds to Kdo2-Lipid A; measurement curve 4 corresponds to: S. enterica Enteritidis LPS; measurement curve 5 correspond to: S. enterica Minnesota LPS; measurement curves 6-9 correspond to: S. enterica Typhosa LPS, E. coli O18A LPS, S. enterica Typhimurium LPS, E. coli 0157 LPS; measurement curve 10 corresponds to: S. enterica Abortus Equi LPS; measurement curve 11 corresponds to: E. coli 027 LPS.

[0028] Fig 4 shows the typing via surface plasmon resonance (SPR) chip of 7 lipopolysaccharides emanating from different pathogens via the O-antigen by using a bacteriophage tailspike protein with unique and high specificities for some of the O-polysaccharide within the different lipopolysaccharides. Figure 4A shows an example of an entire sensorgram with murine lipopolysaccharide- binding protein (LBP, C-terminal polyhistidine-tag, SEQ ID9), binding and dissociation events followed by binding and dissociation of different lipopolysaccharides, followed by specific binding of the tailspike protein mutant ORF169.1 D440A from EP75 bacteriophage (SEQ ID6) to a sub-set of the available O-polysaccharides. Binding is only observed for lipopolysaccharides where the tailspike protein ORF169.1 shows specificity for (non-dashed lines). Figure 4B shows the sensorgram starting at the tailspike protein binding event only. Three different tailspike protein concentrations were applied to demonstrate dose-dependent signal generation. Note that the tailspike protein ORF169.1 D440A from EP75 bacteriophage did not show any binding to E. coli O-antigens such as 0157 or 018A. Measurement curve 1 corresponds to S. enterica Abortus Equi LPS; Measurement curve 2 corresponds to S. enterica Typhimurium LPS; Measurement curve 3, 4 corresponds to S. enterica Enteritidis LPS and S. enterica Typhosa LPS. Measurement curve 5-7 corresponds E. coli 0157 LPS, of E. coli 018A LPS, of S. enterica Minnesota LPS.

[0029] Fig. 5 shows the typing via surface plasmon resonance (SPR) of 7 lipopolysaccharides (LPSs) emanating from different pathogens via the O- polysaccharides by using a bacteriophage tailspike protein with unique and high specificities for some of the O-antigens within the different lipopolysaccharides. Figure 5A shows an example of entire sensorgram with murine lipopolysaccharide-binding protein (LBP, C-terminal polyhistidine-tag, SEQ ID9) binding (0 - about 100 s) and dissociation events (from about 150 s) followed by binding (about 150 s to about 340 s) and dissociation (from about 340 s) of different lipopolysaccharides followed by specific binding of tailspike protein mutant ORF168 D511A from EP75 bacteriophage (SEQ ID5). Binding is only observed for LPSs where the tailspike protein ORF168 shows specificity for (non-dashed line). Figure 5B shows the sensorgram starting at the binding of the tailspike protein only. Three different tailspike protein concentrations were applied to demonstrate dose-dependent signal generation. ORF168 showed binding only to the O-antigen of the E. coli 0157 lipopolysaccharide but not to the lipopolysaccharides of other pathogens. Measurement curve 1 corresponds to E. coli 0157 LPS (labelled 1) , whereas measurement curves 2 -7 correspond to E. coli 018A LPS, E. enterica Typhimurium LPS, S. enterica Enteritidis LPS, S. enterica Minnesota LPS, S. enterica Typhosa LPS, S. enterica Abortus Equi LPS (labelled 2). Fig. 6 shows the dose-dependent signal at different concentrations of lipopolysaccharide using Enzyme-linked Immunosorbent Assay (ELISA) after coating the murine lipopolysaccharide-binding protein (LBP, C-terminal polyhistidine-tag, SEQ ID9) on the plate, subsequent binding of lipopolysaccharide from S. enterica Enteritidis, subsequent binding by S. enterica Enteritidis-specific tailspike protein mutant ORF169.1 D440A (SEQ ID6) and with signal development, employing tailspike-specific antibodies and secondary, HRP-labelled detection antibody. Controls show either coating of the tailspike protein directly (ring symbol, control of the detection system, positive control) and the setup devoid of tailspike protein addition (square symbol, control of non-specific signals of the detection system, negative control.)

[0030] DESCRIPTION OF PREFERRED EMBODIMENTS

[0031] It is a first object of the present invention to provide an in vitro method of detecting one or more endotoxins in a sample.

[0032] It is understood that the method of detection according to the first object of the present invention may be a method of detection that provides a qualitative result for a sample, in the sense that the result will at least indicate the presence or absence of one or more endotoxins in the sample. However, the method of detection according to the first object of the present invention may, in some embodiments, also provide a dose-dependent, or quantitative, result for a sample, in the sense that the result will indicate the presence or absence of one or more endotoxins in the sample, as well a relative or absolute direct indication of the quantity of one or more endotoxins in the sample. In other embodiments, the method of detection according to the first object of the present invention may, in addition allow for the typing one or more endotoxins in a sample, in the sense that the result will indicate to which particular bacterium the one or more endotoxins in the sample can be attributed. It is noted that also typing the one or more endotoxins in a sample can provide a relative or absolute, indirect indication of the quantity of one or more endotoxins in the sample.

[0033] The sample may be obtained via methods known on the art, and depending on the particular methodology and / or devices used, may be prepared accordingly prior to the method according to the first object of the present invention to make the analyte, i.e. the endotoxin, available to the particular ligand, i.e. the lipid binding protein, used in the methodology and / or devices.

[0034] The sample may be derived from biological samples such as blood samples, or other tissues, secretions or fluids of a mammal, for example a human. It is understood that in most cases, the biological samples may be further prepared to isolate and / or enrich the endotoxins present in the biological sample, via appropriate procedures. Alternatively, in some cases, the samples may be directly analysed. For instance, endotoxins may be isolated from a liquid biological sample via two-phase extraction into a phenolic phase.

[0035] The sample may be derived from source materials and solutions, such as starting materials and solutions, intermediate process materials or solutions (in-process control samples) or from final process samples produced in the biopharmaceutical sector such as active pharmaceutical ingredients or after formulation, biopharmaceutical products, for all materials, solutions, medicinal devices intended for the production of and the final products intended for human use. Thus the sample may be derived from a material used in, and / or for, the production of a biopharmaceutical product, such as antibodies.

[0036] In the context of the present invention, the term "endotoxin" refers to lipopolysaccharides (LPSs) that comprise at least a lipid A moiety, since the present invention relies on the interaction between a lipid A moiety of the endotoxin and a corresponding molecule that is capable of binding said lipid A moiety of the endotoxin. However, endotoxins may include other moieties in addition to a lipid A moiety, such as an inner or outer oligosaccharide core, and a O-polysaccharide moiety. Endotoxins are of microorganic, and in particular bacterial, origin and which are generally released upon disintegration of bacterial cells and will differ in terms of inner or outer oligosaccharide core and o-polysaccharide moiety, depending on the microorganism and / or the conditions that led to the release of the endotoxin. In most cases, a bacterial lipopolysaccharide will be composed of three parts: the O antigen, the core oligosaccharide, and lipid A.

[0037] The O polysaccharide moiety corresponds to a repetitive glycan polymer contained within the bacterial lipopolysaccharide is referred also to as O-antigen or O side-chain of the bacteria. The O polysaccharide is attached to the outer core oligosaccharide and comprises the outermost domain of the LPS molecule. It is important to note that the composition of the O polysaccharide varies not only from one species of bacteria to the next, but also from bacterial strain to bacterial strain; over 160 different O polysaccharide structures are known to be produced by different E. coli strains. The O polysaccharide is exposed on the very outer surface of the intact bacterial cell, and, as a consequence, is a target for recognition by host antibodies or by bacteriophage viruses that infect the bacteria.

[0038] The core domain contains an oligosaccharide component that attaches directly to lipid A and commonly contains sugars such as heptose and 3-Deoxy-D-manno-oct-2-ulosonic acid (also known as KDO, keto-deoxyoctulosonate). In most cases, the core domain of bacterial LPSs also contain non-carbohydrate components, such as phosphate, amino acids, and ethanolamine substituents.

[0039] The lipid A moiety is, in normal circumstances, a phosphorylated glucosamine disaccharide decorated with multiple fatty acids. These hydrophobic fatty acid chains anchor the lipopolysaccharides (LPS) into the bacterial outer membrane, and the rest of the LPS projects from the cell surface. The lipid A moiety is responsible for much of the toxicity of Gram-negative bacteria in host organisms. When bacterial cells are lysed by the immune system, fragments of membrane containing lipopolysaccharides (and thus, lipid A) are released into the circulation, causing fever, diarrhea, and possible fatal endotoxic shock (also called septic shock). The Lipid A moiety is a very conserved component across lipopolysaccharides emanating from different bacteria, but may nonetheless vary to a certain degree among bacterial species. Lipid A structure largely defines the degree and nature of the overall host immune activation.

[0040] In a preferred embodiment of the method according to the first object of the present invention, the one or more endotoxin may be an endotoxin of one or more gram-negative bacteria, or one or more strains of a gram-negative bacteria. It is understood that the term "one or more endotoxins" is meant to encompass both one endotoxin of one gram-negative bacterium, a plurality endotoxins which may be from one and the same gram-negative bacterial species or strain, a plurality endotoxins which may be from more than one species of gram-negative bacteria, and a plurality endotoxins which may be from more than one gram-negative bacterial strains.

[0041] The in vitro method of detecting one or more endotoxins in a sample according to a first object of the present invention comprises the steps of a. coating a lipid binding protein on a substrate or capturing a lipid binding protein on a capture molecule immobilized on a substrate. It is noted that the term "substrate" should be understood as referring to a solid support, such as for example a surface of a solid material, and not to an enzymatic substrate.

[0042] In a preferred embodiment of the method according to the first object of the present invention, the lipid binding protein may be coated directly on a suitable substrate. In order to directly coat the lipid binding protein on a substrate, the lipid binding protein may be non- covalently adsorbed to the substrate. For example, the lipid binding protein may be coated onto a solid surface of polymer, which offers hydrophobic patches to which the lipid binding protein may adsorb. In ELISA-type tests, the lipid binding protein is preferably coated directly on a suitable substrate, which may be a polymer material support such as polystyrene.

[0043] In a preferred embodiment of the method according to the first object of the present invention, the lipid binding protein may be captured on a capture molecule that is immobilized on a substrate. The capture molecule may be immobilized, i.e. coupled covalently or non-covalently, to a substrate surface, which may be functionalized to that purpose. Examples for covalently coupling the capture molecule to a substrate are known methods such as click chemistry, which are known to the person skilled in the art, or such as for coupling via Diels-Alder or azide-alkyne cycloaddition reactions, or covalent amine coupling by activation via EDC / NHS.

[0044] Examples for non-covalently coupling of the lipid binding protein to a substrate are for example biotin / avidin coupling. When the lipid binding protein is immobilized directly, either covalently or non-covalently, on a suitable substrate, the is no need for an intermediate molecule, such as for example an antibody. The substrate may be any suitable material such as polymer material. Examples of polymer materials are polystyrene, polyester, polyamide, polycarbonate, silicone, or polyolefin materials.

[0045] In a preferred embodiment of the method according to the first object of the present invention, the lipid binding protein is chosen from lipopolysaccharide binding proteins. It is understood that, in principle, any lipid binding protein that is capable of binding the lipid A moiety of an endotoxin may be used in the context of the present invention, and preferably the lipid binding protein is a protein that, in its native form, binds to the endotoxin to capture and retain the endotoxin. However, the present invention is not limited to lipid binding proteins in their native form, and also encompasses proteins that have been engineered such as to be capable of binding the lipid A moiety of an endotoxin and thereby hold the lipopolysaccharide. As an example, acyloxyacyl hydrolase may be used as lipid binding protein, when the hydrolytic activity is inactivated. It is understood that inactivation may be achieved, for example, by the substitution one or more catalytic site amino acid residues with a non-catalytic amino acid or by chemical inactivation of the one or more catalytic site amino acid residues.

[0046] Acyloxyacyl hydrolase, in its native state, is a lipase that selectively releases the secondary (acyloxyacyl-linked) fatty acyl chains from the hexaacyl lipid A moiety found in many endotoxins. The resulting tetraacyl lipopolysaccharide is non-stimulatory and can be a potent inhibitor of lipopolysaccharides sensing via the MD-2-Toll-like Receptor 4 (TLR4).

[0047] In a preferred embodiment of the method according to the first object of the present invention, the lipid binding protein is chosen from a lipase that is capable of binding the lipid A moiety of an endotoxin and hydrolyzing said endotoxin, and in which lipase one or more amino acid residues of its catalytic site is replaced with a non-catalytic amino acid residue or by chemical inactivation of the one or more catalytic site amino acid residues. For instance, a serine of its catalytic site may be replaced with a non-catalytic amino acid such as glycine, alanine, valine, leucine, isoleucine, or methionine. In a more preferred embodiment of the method according to the first object of the present invention, the lipid binding protein is a lipase such as acyloxyacyl hydrolase, in which the active site serine is replaced with a non-catalytic amino acid, or is chemically inactivated.

[0048] In a preferred embodiment of the method according to the first object of the present invention, the lipid binding protein is chosen from a mammalian, murine or human lipase, and preferably is a human lipase such as for example acyloxyacyl hydrolase.

[0049] In a preferred embodiment of the method according to the first object of the present invention, the lipid binding protein is human acyloxyacyl hydrolase, in particular acyloxyacyl hydrolase in which the active site serine is replaced with an amino acid having a hydrophobic side chain, in particular alanine, and most preferably has an amino acid sequence according to SEQ ID1.

[0050] In another preferred embodiment of the method according to the first object of the present invention, the lipid binding protein is murine acyloxyacyl hydrolase, in particular acyloxyacyl hydrolase in which the active site serine is replaced with an amino acid having a hydrophobic side chain, in particular alanine, and most preferably has an amino acid sequence according to SEQ ID7.

[0051] In yet another preferred embodiment of the method according to the first object of the present invention, the lipid binding protein is rabbit acyloxyacyl hydrolase, in particular acyloxyacyl hydrolase in which the active site serine is replaced with an amino acid having a hydrophobic side chain, in particular alanine, and most preferably has an amino acid sequence according to SEQ ID8.

[0052] In a preferred embodiment of the method according to the first object of the present invention, the lipid binding protein is chosen from bactericidal permeability-increasing protein (BPI). In general, bactericidal permeability-increasing protein (BPI) is not catalytically active and thus can be used "as-is", in its wildtype form. Nonetheless, the bactericidal permeability-increasing protein (BPI) may be modified such as to include a protein affinity tag such as a polyhistidine-tag, which protein affinity tag may preferably located on the C- or N-terminal side of the protein. The bactericidal permeability-increasing protein (BPI) is preferably a mammalian protein, in particular a human protein such as for example the bactericidal permeability-increasing protein (BPI) according to SEQ ID3 or SEQ ID4.

[0053] In a preferred embodiment of the method according to the first object of the present invention, the lipid binding protein is chosen lipopolysaccharide-binding protein (LBP). Lipopolysaccharide-binding protein is involved in the innate immunity and binds to the lipid A moiety of the endotoxin. It enhances the affinity of the endotoxin for CD14 an thereby enhancing the immune response towards endotoxins. It is in general not enzymatically active and can employed in its wild-type form. However, the lipopolysaccharide-binding protein (LBP) may be modified to include a tag such as a polyhistidine-tag, which protein affinity tag may preferably be located on the C- or N-terminal side of the protein. The lipopolysaccharide-binding protein (LBP) is preferably a mammalian protein, it is preferably a murine or human protein such as for example the lipopolysaccharide-binding protein (LBP) according to SEQ ID10 or SEQ ID9.

[0054] The method according to the first object of the present invention includes a step of coating a lipid binding protein on a substrate or capturing a lipid binding protein on a capture molecule immobilized on a substrate. The substrate serves the purpose of binding the lipid binding protein or the immobilized capture molecule. The substrate will depend on the methodology used to detect the binding of any endotoxins to the lipid binding protein. For example, when the methodology used is of the ELISA-type, the substrate may be a polymer plate or well, such as polystyrene, which polymer plate or well can be locally or entirely coated with the capture molecule, such as for example an antibody, on the substrate. The coating is generally carried out, as will be known to the person skilled in the art, by first applying a coating solution comprising the capture molecule onto the polymer plate or well and then incubating, normally at room temperature and overnight, to allow the capture molecule to be bound to the substrate. After rinsing with a rinsing buffer, a blocking solution is then applied to avoid unspecific interaction between the endotoxin and the substrate. When the methodology used is of the surface plasmon resonance-type, the substrate may be a solid material such as a gold membrane, that may be coated locally or entirely coated with a suitable polysaccharide such as for example dextran that facilitates immobilizing the capture molecule, such as for example an antibody, on the substrate. The ELISA plates and surface plasmon resonance sensor chips are commercially available, and the person skilled in the art will know how to select the appropriate substrate such as the ELISA microtiter plate or the SPR sensor chip to match the methodology.

[0055] In a preferred embodiment of the method according to the first object of the present invention, the lipid binding protein is captured on a capture molecule immobilized on a substrate or is directly coated onto a solid support. The capture molecule may preferably comprise an antigen-binding region that selectively binds a lipid binding protein having a region that is recognized and bound by the antigen-binding region of the capture molecule. It is understood that the capture molecule should be chosen such that the lipid binding protein retains its binding to the endotoxins while captured on the capture molecule. The binding, or dissociation, between a capture molecule and a lipid binding protein may be quantified, for example, via a surface plasmon resonance (SPR) experiment in which the binding or dissociation between the capture molecule and lipid binding protein is measured. To quantify binding, or dissociation, a sensorgram is recorded during which the capture molecule is immobilized on a surface plasmon resonance chip and is contacted with a solution of lipid binding protein having a concentration of 5 pg / ml. Then, the solution of lipid binding protein having a concentration of 20 pg / ml is exchanged for a solution that is free of lipid binding protein but is otherwise identical in composition, to allow the release of lipid binding protein into the solution that is free of lipid binding protein. Then, the change in relative response is recorded over a period of 10 mins. If the lipid binding protein is tightly bound by the capture molecule, the decrease of relative response in the sensorgram will be small, and if the lipid binding protein is loosely bound by the capture molecule immobilized on the solid support, the decrease of relative response in the sensorgram will be high.

[0056] It was found that lipid binding proteins bound tightly to the capture molecule with a very small decrease of relative response in the sensorgram showing a highly stable binding interaction and exhibiting a constant decrease of relative response in the above surface plasmon resonance (SPR) measuring method of 0.54 % s’1or less, preferably of 0.06 % s’1or less, and more preferably of 0.02 % s’1or less, when captured on the capture molecule, were advantageous in the present invention.

[0057] The binding, or dissociation, between a lipid binding protein and an endotoxin may be quantified, for example, via a surface plasmon resonance (SPR) experiment in which the binding or dissociation between the lipid binding protein and an endotoxin is measured. To quantify binding, or dissociation, a sensorgram is recorded during which the lipid binding protein is captured on a capture molecule immobilized on a surface plasmon resonance chip and is contacted with a solution of endotoxin having a concentration of 10 pg / ml until no further increase in binding between the capture molecule and the lipid binding protein is recorded. Then, the solution of endotoxin having a concentration of 10 pg / ml is exchanged for a solution that is free of endotoxin but is otherwise identical in composition to allow the release of endotoxin into the solution that is free of endotoxin. Then, the change in relative response is recorded over a period of 10 mins. If the endotoxin is tightly bound by the lipid binding protein, the decrease of relative response in the sensorgram will be small, and if the endotoxin is loosely bound by the lipid binding protein, the decrease of relative response in the sensorgram will be high.

[0058] It was found that lipid binding proteins exhibiting a constant decrease of relative response in the above surface plasmon resonance (SPR) measuring method of 0.6 % s’1or less, preferably of 0.3 % s’1or less, and more preferably of 0.12 % s’1or less, when binding the endotoxin, were advantageous in the present invention.

[0059] In order to enhance and control the capturing on the capture molecule, the capture molecule and the lipid binding protein may be a recombinant lipid binding protein, or may be a wildtype protein. In a preferred embodiment of the method according to the first object of the present invention, the lipid binding protein is recombinant and may comprise a protein affinity tag, such as for example a polyhistidine-tag (His-tag). The affinity tag may be a C-terminal or N- terminal affinity tag. For example, when the lipid binding protein comprises a polyhistidine- tag, the capture molecule comprises an anti-polyhistidine-tag binding region, and may be an anti-polyhistidine-tag antibody, for example. In another embodiment, the lipid binding protein comprises a biotin-tag at either the N-terminus or the C-terminus and the capture molecule may be a biotin-binding capture molecule and may be a biotin-binding antibody or may be a streptavidin, for example. In another embodiment, the lipid binding protein comprises a strep-tag at either the N-terminus or the C-terminus and the capture molecule may be a strep-tag-binding molecule and may be a strep-tag binding antibody or may be a streptavidin, for example. In a more preferred embodiment of the method according to the first object of the present invention, the lipid binding protein is lipopolysaccharide binding protein comprising a polyhistidine-tag, preferably a C-terminal or N-terminal polyhistidine- tag. When the lipid binding protein is recombinant and comprises an affinity tag, and the capture molecule comprises an affinity-tag binding region such as in an anti-polyhistidine- antibody, the lipid binding protein may be released from the capture molecule under mild conditions by a change of pH, salt concentration or the like. In another embodiment, when the lipid binding protein is recombinant and comprises a streptavidin-tag, and the capture molecule comprises strep-tag binding region such as in an anti-strep-antibody or a streptavidin, the lipid binding protein may be released from the capture molecule under mild conditions by adding competing strep-tag peptide or biotin or desthiobiotin in the buffer, or by changing the salt concentration or the like. This allows to release a lipid binding protein tightly binding an endotoxin from the capture molecule, and to elute the lipid binding protein - endotoxin complex while retaining the capture molecule immobilized on the substrate, ready to capture new lipid binding protein for the next round of endotoxin detection using new lipid binding protein. Because no harsh conditions are applied, the capture molecules do not progressively denature or hydrolyze and the process of regeneration may be repeated multiple times without affecting the robustness of the assay. This contrasts with the situation where the lipid binding protein - endotoxin complex is undone to release the endotoxin. While this approach also frees up the existing lipid binding protein for the next endotoxin detection cycle, the lipid binding protein - endotoxin complex must be exposed to harsh conditions to dissociate as the binding between them is quite strong and these harsh conditions denature, aggregate or hydrolytically damage the capture molecule and / or the lipid binding protein, ultimately leading to a higher use in SPR chips or other surfaces. The same is true for the case where a tailspike protein is immobilized directly on a surface, such as an SPR chip.

[0060] Because no harsh conditions are applied, the capture molecules do not progressively denature, aggregate or get damaged in any other sense and the process of regeneration may be repeated multiple times without affecting the robustness of the assay. This regeneration procedure avoids necessity of dissociation of the endotoxin from the lipid binding , which would require harsh conditions due to strong binding between the lipid binding protein and the endotoxin and will denature, aggregate or damage in any other sense the lipid binding protein including the capture molecule on the chip and will not make the bound proteins, either the lipid binding protein and I or the capture molecule, suitable for another endotoxin binding cycle, , ultimately leading to a higher use in SPR chips or other surfaces.

[0061] In the same embodiment of the method according to the first object of the present invention, regeneration of the SPR chip is equally achieved on the level of the capture molecule immobilized on the sensor chip when endotoxin is bound to the lipid binding protein and a tailspike protein is contacting the endotoxin and is binding to the O-polysaccharide. In the same embodiment of the method according to the first object of the present invention, regeneration is not required for the detection and quantification of endotoxins by ELISA method, because different concentrations of the sample containing the endotoxin are bound to coated lipid binding protein in different wells of the solid support at the same time (in parallel), in contrast to the SPR-based method where different concentrations of the sample containing the endotoxin are bound to the lipid binding protein in a serial manner.

[0062] When the lipid binding protein is recombinant, it may further also comprise a signal peptide. A signal peptide may facilitate the expression and isolation of lipid binding protein.

[0063] The method according to the first object of the present invention includes a step of contacting the lipid binding protein, thus coated or captured, with the sample.

[0064] Contacting the lipid binding protein with the sample is preferably carried out by providing the sample in a suitable form, such as for example a buffer solution comprising the endotoxin. The sample may thus be a liquid sample, and preferably is in the form of an aqueous buffer solution. Again, the exact composition of the aqueous solution will be determined by the methodology used to detect whether any endotoxin binds the lipid binding protein, as pH, salt concentrations, and such, must be adapted to the methodology used to ensure good results. The method according to the first object of the present invention includes a step of detecting whether an endotoxin binds to the lipid binding protein, wherein the lipid binding protein is capable of binding the lipid A moiety of an endotoxin.

[0065] Whether any endotoxin binds to the lipid binding protein may be detected via a suitable methodology.

[0066] Whether any endotoxin binds to the lipid binding protein may be detected, for example, via a change in optical density such as fluorescence, colorimetry, chemiluminescence or ioluminescence. For instance, when the step of determining if the lipid binding protein binds the endotoxin is carried out via enzyme-linked immunosorbent assay (ELISA), a presence or intensity of coloration is determined, if a qualitative or quantitative ELISA is carried out. Alternatively, whether any endotoxin binds to the lipid binding protein may be detected, for example, via a change in mass. For instance, when the step of determining if the lipid binding protein binds the endotoxin is carried out via surface plasmon resonance, the increase of mass due to the analyte binding the lipid binding protein is detected on the solid support, which may be on a SPR chip.

[0067] When the step of determining if the lipid binding protein binds the endotoxin is carried out via enzyme-linked immunosorbent assay (ELISA), the enzyme-linked immunosorbent assay (ELISA) may be a direct or indirect enzyme-linked immunosorbent assay (ELISA).

[0068] In the case where the lipid binding protein is captured on on a substrate or solid support, such as an ELISA microtiter plate, the enzyme-linked immunosorbent assay (ELISA) may be referred to as a sandwich enzyme-linked immunosorbent assay (ELISA), which may also be either direct or indirect. In general, when the step of determining if the lipid binding protein binds the endotoxin is carried out via enzyme-linked immunosorbent assay (ELISA), the endotoxin bound to the lipid binding protein is contacted with at least one further antibody molecule that selectively binds either the lipid A moiety, inner or outer core oligosaccharide or the O-polysaccharide moiety of an endotoxin, i.e. the o-polysaccharide. In a preferred embodiment of the method of the present invention, the at least one antibody molecule selectively binds the polysaccharide moiety of an endotoxin of one or more particular bacteria. This allows typing for a particular strain of pathogen, which may be important when determining whether a sample comprises a pathogen of clinical importance, such as for example disease-causing pathogens or multiply resistant pathogens. In a preferred embodiment of the method according to the first object of the present invention, the step of determining if the lipid binding protein binds the endotoxin is carried out via enzyme-linked immunosorbent assay (ELISA), and the endotoxin bound to the lipid binding protein is contacted with at least one primary antibody molecule that selectively binds the polysaccharide moiety of an endotoxin, i.e. the o-polysaccharide. It is understood that said primary antibody molecule may be coupled with an enzymatic protein moiety or with a fluorophore or chromogenic, fluorescent, bioluminescent or chemiluminescent moiety, or alternatively, that said primary antibody molecule may be contacted with a secondary antibody molecule with an enzymatic protein moiety that selectively binds the primary antibody molecule. In a preferred embodiment of the method of the present invention, the of at least one antibody molecule selectively binds the polysaccharide moiety of an endotoxin of one or more particular bacteria. In another preferred embodiment of the method according to the first object of the present invention, said primary antibody may be coupled with an optically active or reactive moiety, or alternatively, said primary antibody molecule may be contacted with a secondary antibody molecule with an optically active or reactive moiety that selectively binds the primary antibody molecule. Optically active or reactive moieties may be for, example, fluorescent or luminescent moieties. The optically active or reactive moieties may be protein moieties or non-protein moieties. Examples of fluorescent protein moieties are green fluorescent proteins, which are available in a range of colors. Examples of luminescent protein moieties are luciferins, which are available in a range of colors. This allows typing for a particular strain of pathogen, which may be important when determining whether a sample comprises a bacteria of clinical importance, such as for example disease-causing pathogens or multiple resistant pathogens.

[0069] In a preferred embodiment of the method according to the first object of the present invention, the step of determining if the lipid binding protein binds the endotoxin is carried out via enzyme-linked immunosorbent assay (ELISA), and the endotoxin bound to the lipid binding protein is contacted with at least one bacteriophage protein that selectively binds the polysaccharide moiety of an endotoxin, i.e. the o-polysaccharide, of one or more particular bacteria. In a preferred embodiment of the method according to the first object of the present invention, the bacteriophage protein comprises a tailspike protein or a depolymerase and comprises an enzymatically inactive form where the inactivation is achieved by exchanging catalytically active amino acid residues or by any other chemical means. Subsequently, the complex comprising the one or more endotoxin-bacteriophage protein complex is contacted with a primary antibody molecule that selectively binds the bacteriophage protein. It is understood that said primary antibody may be coupled with an enzymatic protein moiety, or alternatively, said primary antibody molecule may be contacted with a secondary antibody molecule with an enzymatic protein moiety that selectively binds the primary antibody molecule. In another preferred embodiment of the method according to the first object of the present invention, said primary antibody may be coupled with an optically active or reactive moiety, or alternatively, said primary antibody molecule may be contacted with a secondary antibody molecule with a optically active or reactive moiety that selectively binds the primary antibody molecule. Optically active or reactive moieties may be for, example, fluorophore or chromogenic, fluorescent, bioluminescent or chemiluminescent. The optically active or reactive moieties may be protein moieties such as horse radish peroxidase moieties or alkaline phosphatase, which induce a color change in a suitable substrate such as o-phenylenediamine dihydrochloride or p-nitrophenyl phosphate. Examples of fluorescent protein moieties are green fluorescent proteins, which are available in a range of colors. Examples of luminescent protein moieties are luciferins, which are available in a range of colors. This allows typing for a particular strain of pathogen, which may be important when determining whether a sample comprises a pathogen of clinical importance, such as for example disease-causing pathogens or multiple resistant pathogens.

[0070] In another embodiment of the method according to the first object of the present invention, the one or more molecules capable of selectively binding the O-polysaccharide moiety of an endotoxin of one or more particular bacteria are bacteriophage proteins, and may comprise bacteriophage tailspike proteins or depolymerase proteins or any other bacteriophage proteins selectively binding to O-polysaccharides within the endotoxins. In a preferred embodiment, the proteins capable of selectively binding the O-polysaccharides are bacteriophage tailspike proteins that can discriminate between o-polysaccharides of the endotoxins of host pathogens and non-host pathogens, which can levied in the context of the present invention. In a preferred embodiment, if the bacteriophage protein has an enzymatic activity, it will be inactivated by exchange of catalytically active amino acid residues or by any other chemical means

[0071] In a preferred embodiment of the method according to the first object of the present invention, the one molecule capable of selectively binding the O-polysaccharide moiety of an endotoxin of one particular bacterium is added. Thus, the method according to the first object of the present invention may be useful in confirming or refuting the suspected presence of an endotoxin of one particular bacterium in a sample. In another preferred embodiment of the method according to the first object of the present invention, a plurality of different molecules capable of selectively binding the O- polysaccharide moiety of endotoxins of a plurality of particular bacteria is added. Thus, the method according to the first object of the present invention may be useful in confirming or refuting the suspected presence of at least one endotoxin of a plurality of particular bacteria in a sample.

[0072] In a preferred embodiment of the method according to the first object of the present invention, the one or more molecules capable of selectively binding the O- polysaccharide moiety of an endotoxin of one or more particular pathogens are bacteriophage tailspike proteins. It is understood that in general, and in contrast to the lipid binding protein, the one or more molecules capable of selectively binding the O-polysaccharide moiety of an endotoxin, and in particular bacteriophage tailspike proteins, may be free of, or may comprise affinity tags and the like, provided they to not interfere with the binding between the one or more molecules capable of selectively binding the O- polysaccharide moiety of an endotoxin and the endotoxin.

[0073] In a preferred embodiment of the method according to the first object of the present invention, the one or more molecules capable of selectively binding the O- polysaccharide moiety of an endotoxin of one or more particular bacterium are bacteriophage tailspike proteins in which the hydrolytic activity against the O-polysaccharide moiety of the LPS (lipopolysaccharide) is inactivated, preferably by a point mutation in which the catalytic amino acid residue of the tailspike protein is exchanged for a non-catalytic amino acid residue or by post-translational chemical inactivation of the active residue. Some wildtype tailspike proteins exhibit hydrolytic activity against the O- polysaccharide moiety of the LPS (lipopolysaccharide), and while they may also be used in the context of the present invention, it is preferably to use bacteriophage tailspike proteins in which the hydrolytic activity against the O- polysaccharide moiety of the LPS (lipopolysaccharide) is inactivated, since the release of the fragment of the O- polysaccharide moiety upon hydrolysis may in some cases interfere with the method according to the present invention.

[0074] In a preferred embodiment of the method according to the first object of the present invention, the one or more endotoxins of one or more bacteria that may be detected, are endotoxins of Salmonella sp. such as Salmonella enterica, preferably chosen among Salmonella enterica serovar Enteritidis, Salmonella enterica serovar Abortus Equi, Salmonella enterica serovar Typhimurium, Salmonella enterica serovar Minnesota, S. Minnesota RE595, Salmonella enterica serovar Typhi; or of Escherichia sp. such as Escherichia coli, preferably chosen among Escherichia coli EH 100, Escherichia coli 018 A, Escherichia coli 0157, and Escherichia coli 027.

[0075] Accordingly, the one or more molecules capable of selectively binding the O- polysaccharide moiety of an endotoxin of one or more particular bacteria may be chosen among bacteriophage tailspike proteins that selectively bind to the O- polysaccharide moiety of an endotoxin of Salmonella sp. such as Salmonella enterica, preferably chosen among Salmonella enterica serovar Enteritidis, Salmonella enterica serovar Abortus Equi, Salmonella enterica serovar Typhimurium, Salmonella enterica serovar Minnesota, S. Minnesota RE595, Salmonella enterica serovar Typhi; or of Escherichia sp. such as Escherichia coli, preferably chosen among Escherichia coli EH 100, Escherichia coli 018 A, Escherichia coli 0157, and Escherichia coli 027.

[0076] When the step of determining if the lipid binding protein binds the endotoxin is carried out via surface plasmon resonance, the increase of mass due to the binding of an endotoxin to the lipid binding protein is detected.

[0077] In a preferred embodiment of the method according to the first object of the present invention, the step of determining if the lipid binding protein binds the endotoxin is carried out via surface plasmon resonance, and the endotoxin bound to the lipid binding protein is contacted with at least one primary antibody molecule that selectively binds the polysaccharide moiety of an endotoxin, i.e. the o-polysaccharide moiety. It is understood that in the case of surface plasmon resonance said primary antibody molecule may not need to be coupled a reporter moiety, or alternatively, said secondary antibody molecule may not need to be coupled to a reporter moiety, as detection is based directly on the change in mass. This allows typing for a particular strain of pathogen, which may be important when determining whether a sample comprises a pathogen of clinical importance, such as for example disease-causing bacteria or multiply resistant bacteria. Primary antibody molecule that selectively bind the polysaccharide moiety of an endotoxin, i.e. the o-antigen, are commercially available. Secondary antibody molecules that selectively bind the primary antibody molecule, are commercially available.

[0078] In a preferred embodiment of the method according to the first object of the present invention, the step of determining if the lipid binding protein binds the endotoxin is carried out via surface plasmon resonance, and the endotoxin bound to the lipid binding protein is contacted with at least one bacteriophage protein such as a tailspike protein that selectively binds the polysaccharide moiety of an endotoxin, i.e. the o-polysaccharide, of one or more particular bacteria. Subsequently, the complex comprising the one or more endotoxinbacteriophage protein complex may be contacted with a primary antibody molecule that selectively binds the bacteriophage protein. It is understood that in the case of surface plasmon resonance said primary antibody molecule may not need to be coupled a reporter moiety, or alternatively, said secondary antibody molecule may not need to be coupled to a reporter moiety, as detection is based directly on the change in mass. This allows typing for a particular strain of bacteria, which may be important when determining whether a sample comprises a bacteriumof clinical importance, such as for example disease-causing pathogens or multiply resistant pathogens. For example, the at least one bacteriophage protein that selectively binds the polysaccharide moiety of an endotoxin, i.e. the o- polysaccharide, of one or more particular bacteria may be one or more bacteriophage tail spike proteins, for example from Escherichia phages such as from Escherichia coli 0157 phage EP75, and may in particular be chosen from the gene product of ORF168 D511A , preferably according to SEQ ID5 and / or from the gene product of ORF169.1 D440A preferably according to SEQ ID6.

[0079] In a preferred embodiment of the method according to the first object of the present invention, the lipid binding protein is a lipid-binding mammalian protein, preferably is human bactericidal / permeability-increasing protein (BPI Uniprot: P17213), preferably having a sequence according to SEQ ID3 or SEQ ID4, but may originate from other mammalian organisms such as horse, pig, sheep, dog, cat, rat, rabbit, mouse, bovine or others, or may originate from BPI fold-containing protein families with similar 3D fold (human variant Uniprot: Q8N4F0) or other proteins with a similar fold and lipid binding properties from the tubular lipid-binding protein superfamily TULIP (Reviewed in Alva and Lupas Biochimica et Biophysica Acta 2016).

[0080] In a preferred embodiment of the method according to the first object of the present invention, the lipid binding protein is lipid binding protein is a lipid-binding mammalian glycoprotein, preferably is lipopolysaccharide binding protein such murine lipopolysaccharide binding protein (mLBP, Uniprot: Q61805), preferably having a sequence according to SEQ ID9 or SEQ ID10, but may originate from other mammalian organisms such as , horse, big, sheep, dog, cat, human, mouse, rat, rabbit, bovine or others, or may originate from LPS binding fold-containing protein families with similar 3D fold such as the tubular lipid-binding protein superfamily TULIP (Reviewed in Alva and Lupas Biochimica et Biophysica Acta 2016).

[0081] In a preferred embodiment of the method according to the first object of the present invention, is mammalian acyloxyacyl hydrolase, preferably human acyloxyacyl hydrolase (hAOAH Uniprot:P28039) having a sequence according to SEQ ID1 or SEQ ID2, or which may originate from other mammalian organisms such as mouse (mAOAH) (Uniprot: 035298) having a sequence according to SEQ ID7, rabbit (rAOAH Uniprot: 018823) having a sequence according to SEQ ID8, , rat, bovine, horse, big, sheep, dog, cat, bovine or others, or may originate from BPI fold-containing protein families with similar 3D fold (human variant Uniprot: Q8N4F0).

[0082] While endotoxins are found in gram-negative bacteria, the present invention is, in principle, also suitable for the detection of other bacteria such as gram-positive bacteria, archea or fungi. For instance, in the case of gram-positive bacteria, instead of one or more endotoxins, the method is amended to detect the binding of lipoteichoic acid to a ficolin, such as L- ficolin, which ficolin was either immobilized to a substrate or captured on a capture molecule immobilized on a substrate. In principle, the mammalian ficolin may originate from other mammalian organisms such as human, mouse, rat, rabbit, bovine or others. A suitable human plasma glycoprotein is L-ficolin (lipoteichoic acid binding protein, Uniprot: Q15485).

[0083] It is a second object of the present invention, to provide a device for use in the in vitro method of detecting one or more endotoxins in a sample according to the first object of the invention, comprising a substrate and a lipid binding protein coated on said substrate or a lipid binding protein captured on a capture molecule immobilized on said substrate.

[0084] In a preferred embodiment of the device according to the second object of the present invention, the lipid binding protein is coated on a substrate and wherein the device is an ELISA well microtiterplate, preferably an ELISA 96-well microplate.

[0085] In a preferred embodiment of the device according to the second object of the present invention, the lipid binding protein is captured on a capture molecule immobilized on said substrate and wherein the device is a surface plasmon resonance (SPR) sensor chip. EXAMPLES

[0086] Materials and Methods

[0087] Mammalian lipid-binding proteins AOAH, LBP and BPI, modified with either a N-terminal or a C-terminal polyhistidine-affinity tag (polyhistidine-tag) were produced in CHO cells (CHOgro® High Yield Expression System by Mirus) and purified using a His Trap HP column (Cytiva).

[0088] Tail spike proteins ORF168 D511A D251-I926 and ORF169. 1 D440A full-length active site mutants from E. co / / -specific bacteriophage EP75 ((https: / / www.ncbi.nlm.nih.gov / nuccore / 1377 595891 , Mierlo et al., Microbiol Resour Announc. 2019) modified by a N-terminal strep-tag followed by an enterokinase cleavage site were produced in E. coli BL21 (DE3) cells and purified using a StrepTrap XL pre-packed column (Cytiva) to yield tailspike proteins according to SEQ ID5 and SEQ ID6.

[0089] SPR Experiment

[0090] Surface plasmon resonance measurements were performed using a Biacore T200 instrument (Cytiva). A histidine-tag-specific antibody was immobilized on a CM5 sensor chip (Cytiva) by standard EDC / NHS amine coupling according to the manufacturers protocol to reach a signal of approximately 55’000 RU. After quenching of activated carboxyl groups, the lipid-binding proteins AOAH, BPI and LBP either carrying a N- or a C-terminal polyhistidine-tag were diluted to 5 pg / ml in SPR buffer (10 mM HEPES, 150 mM NaCI, 3 mM EDTA, 0.05 % (v / v) surfactant P20) and were captured onto the chip with a flow rate of 10 pl / min and a total contact time of 120 seconds (association of lipid-binding protein). In the following, a dissociation step was applied for a duration of up to 125 seconds at the same flow rate of 10 pl / min using SPR buffer (10 mM HEPES, 150 mM NaCI, 3 mM EDT, 0.05 % (v / v) surfactant P20) devoid of lipid-binding proteins. Essentially no signal loss in the dissociation step could be recorded, which is indicative of the strong binding between the histidine-tag-specific antibody and the lipid-binding proteins AOAH, BPI and LBP.

[0091] Then, different endotoxins (lipopolysaccharides (LPS) or lipid A) were bound to the captured lipid-binding proteins at a concentration of 50 pg / ml applying a contact time of 120 s with a flow rate of 10 pl / min, dissociation duration was 125 s.

[0092] Each different LPS concentration required a regeneration step applying 10 mM glycine-HCI pH 1 .5 with a flow rate of 30 pl I min and a contact time of a total of 30 s. Hence, regeneration took place at the level of the immobilized antibody. For each cycle / concentration of LPS, fresh lipid-binding protein was bound to the chip.

[0093] Examples of stable and dose-dependent signals generated using lipopolysaccharide (LPS) from S. enterica Typhimurium at 10, 50 and 100 pg / ml preceded by the capture step with murine lipopolysaccharide-binding protein (LBP, C-terminal polyhistidine-tag) are represented in Figure 1 .

[0094] Examples of stable and dose-dependent signals generated using lipopolysaccharide (LPS) from S. enterica Typhimurium at 10, 50 and 100 pg / ml preceded by the capture step with bactericidal permeability-increasing protein (BPI, N-terminal or C-terminal polyhistidine-tag) are represented in Figure 2.

[0095] Examples of stable and dose-dependent signal generated using 11 different lipopolysaccharides or Lipid A molecules from different bacteria binding to murine lipopolysaccharide-binding protein (LBP, C-terminal polyhistidine -tag). In Figure 3A on the left,: the entire sensorgram including binding event of lipid-binding protein followed by endotoxin association is depicted. In Figure 3B on the right, a sensorgram showing only endotoxin analyte binding and dissociation steps is depicted.

[0096] Subsequently, the typing aspect of the present invention is described. Typing of bound lipopolysaccharides was achieved by using O-polysaccharide-specific and recombinantly produced bacteriophage tailspike proteins (TSP) active site mutants, which were bound to the chip after association and dissociation step of lipopolysaccharides after obtaining a stable RU signal as shown in two examples in Figure 4 and Figure 5.

[0097] Generally, tailspike proteins were bound using at least five different concentrations ranging from 5 nM -360.6 nM protein concentration depending on binding affinities of the TSP for a particular O-antigen for the purpose of binding kinetics and quantification.

[0098] Figures 4 and 5 only show three tailspike protein concentrations each, for the purpose of bacterial typing to demonstrate that binding interaction is dose-dependent and hence is specific for the O-polysaccharide. Contact time for the association was 120 s, for the dissociation 60 s, flow rate of 10 pl / min was applied. SPR signals were only obtained for combinations where the tailspike protein originated from a bacteriophage with specificity for the particular bacterium comprising the particular O-polysaccharide within its lipopolysaccharide. Hence, ORF169.1 belongs to bacteriophage EP75 which has known specificity for either E. coli serotype 0157 or various Salmonella strains, whereas the particular tailspike protein ORF169.1 provides the salmonella specificity within this bacteriophage and has no specificity for E. coli 0157 O-polysaccharide. Another tailspike protein ORF168 within EP75 on the other hand, provides the E. coli 0157 specificity but does not feature any salmonella specificity (Witte et al., Computational and Structural Biotechnology Journal, 2021). Hence, binding is clearly only observed in accordance to specificities of the particular tailspike proteins for a specific O-polysaccharide. Four identical regeneration steps were applied, each with 10 mM Glycine-HCI pH 1 .5, contact time for 60 s, flow rate was 30 pl I min.

[0099] In particular, Figure 4 shows typing of different salmonella O- polysaccharide by employing the ORF169.1 D440A tailspike protein according to SEQ ID6 which yields a signal only for four salmonella strains known to be enzymatic substrate for this tailspike protein ( i.e. S. enterica Typhosa, S. enterica Typhimurium, S. enterica Enteritidis and S. enterica Abortus Equi), but it does not show any binding to S. enterica Minnesota and E. coli 0157 or E. coli 018A O- polysaccharide of the endotoxins which are known to be no enzymatic substrate for this tailspike protein since wildtype protein ORF169.1 does not hydrolyse the latter two O- polysaccharide and the intact phage is not specific for those hosts either. Further, the tailspike protein that is the gene product of ORF169.1 based on EP75 bacteriophage tailspike protein according to SEQ ID6 is inactivated with respect to the hydrolytic activity for S. enterica Typhosa, S. enterica Typhimurium, S. enterica Enteritidis and S. enterica Abortus Equi O-polysacchairdes via point mutation D440A. The experiment corresponding to the sensorgram in Figure 4A and 4B was set up with murine lipopolysaccharide-binding protein (LBP, C-terminal polyhistidine-tag) binding different lipopolysaccharides emanating from different bacteria, followed by specific binding of tailspike protein that is the gene product of ORF169.1 D440A based on EP75 bacteriophage tailspike protein according to SEQ ID6. Binding is only observed for lipopolysaccharides emanating from bacteria where the tailspike protein ORF169.1 shows specificity for (non-dashed lines). Figure 4B shows the part of the sensorgram starting at the tailspike binding event. As can be seen, using three increasing tailspike protein concentrations lead to three signals of increasing Rll signal levels, thereby demonstrating dose-dependent signal generation. Tailspike protein that is the gene product of ORF169.1 D440A based on EP75 bacteriophage tailspike protein according to SEQ ID6 did does not show any binding, and thus no change in signal, to E. coli O-polysaccharide such as 0157 or 018A.

[0100] In contrast, Figure 5 shows typing of E. coli O- polysaccharide by employing the ORF168 tailspike protein according to SEQ ID5, which yields a signal only for E. coli 0157, but it does not show any binding to S. enterica Typhosa, S. enterica Typhimurium, S. enterica Enteritidis, S. enterica Abortus Equi, S. enterica Minnesota or E. coli 018A O-antigens (Witte et al 2021 and own unpublished data). The tailspike protein that is the gene product of ORF168 based on EP75 bacteriophage tailspike protein according to SEQ ID5 is inactivated with respect to the hydrolytic activity via point mutation D511A. The experiment corresponding to the sensorgram in Figure 5A and 5B was set up with murine lipopolysaccharide-binding protein(LBP, C-terminal polyhistidine-tag) binding and dissociation events followed by binding and dissociation of different lipopolysaccharides emanating from different bacteria followed by specific binding of tailspike protein that is the gene product of ORF168 D511A based on EP75 bacteriophage tailspike protein according to SEQ ID5. Binding is only observed for LPS where the tailspike protein ORF168 shows specificity for (non-dashed line). Figure 5B shows the sensorgram starting at the binding of the tailspike protein only. Three increasing tailspike protein concentrations were applied to demonstrate dose-dependent and thereby specific signal generation. ORF168 showed binding only to E. coli 0157 O- polysaccharide but not to the other LPS tested, es expected from in vitro activity tests with this tailspike protein.

[0101] ELISA Experiment

[0102] Maxisorb 96 well microtiter plates (Sigma #M9410-1CS) were coated overnight, 4 °C with 100 pl of 50 pg I ml lipopolysaccharide-binding protein (LBP C-terminal polyhistidine-tag) which was diluted in coating buffer of 50 mM sodium carbonate, pH 9.6. The plate was washed three times with wash buffer (PBS, 0.05 % Tween 20) for 5 min, room temperature, and subsequently blocked with 200 pl I well blocking buffer for 1 h room temperature (0.1 M sodium phosphate, 0.1 NaCI, 0.1 % gelatine, 0.05 % Tween pH 7.4). The plate was washed again three times with wash buffer, followed by addition of 100 pl I well of lipopolysaccharide (derived from S. enterica Enteritidis) diluted in blocking buffer at concentration ranging from 0.1 to 50 pg I ml, and incubation was performed for 1 h at 30 °C. The plate was washed three times in wash buffer and 100 pl / well of tailspike protein (ORF169.1 D440A) diluted to 50 pg / ml diluted in coating buffer was added to each well with an incubation of 1 h 30 °C, 300 rpm. The plate was washed again three times in wash buffer. 100 pl / well of polyclonal rabbit anti-ORF169.1 IgG antibody was added at a 1 :500 dilution in blocking buffer to the wells, incubation was performed for 2 h at room temperature. The plate was washed again three times using wash buffer and 100 pl / well of secondary HRP-labelled goat anti-rabbit was added to the wells at 1 :2500 in blocking buffer and incubated for 1 h at room temperature. The plate was washed five times in wash buffer and detection was performed using OPD applying 100 pl / well, development was performed for max. 2 min at room temperature. Reaction was stopped using 100 pl / well of 2.25 M H2SO4. Plate was read at 492 nm - 405 nm in an Infinite 200 Pro (Tecan). Results are shown in Figure 6. As can be seen, a dose-dependent signal for LPS (S. enterica Enteritidis) is observed, as increasing concentrations of LPS lead to increased signal, when using a specific tailspike protein mutant (ORF169.1 D440A, SEQ ID6). It is thus apparent that also an ELISA setup can be employed to either determine the presence of an endotoxin, both qualitatively and quantitatively, and / or in addition to identify the bacterium from which the endotoxin emanates by using tailspike proteins that are highly specific binders for the O-polysaccharide of the endotoxin of a pathogen being investigated.

Claims

CLAIMS1 . An in vitro method of detecting one or more endotoxins of one or more pathogens in a sample, comprising the steps of a. coating a lipid binding protein on a substrate or capturing a lipid binding protein on a capture molecule immobilized on a substrate, b. contacting the lipid binding protein, thus coated or captured, with the sample, c. detecting whether an endotoxin binds to the lipid binding protein, wherein the one or more endotoxins comprise a lipid A moiety and O- polysaccharide moiety, wherein the lipid binding protein is capable of binding the lipid A moiety of an endotoxin, and characterized in that the lipid binding protein is a mammalian protein and preferably is a murine or human protein.

2. The in vitro method of detecting endotoxins according to claim 1 , further comprising the steps of, after step c., d. contacting any endotoxin bound to the lipid binding protein with one or more molecules capable of selectively binding the O- polysaccharide moiety of one or more endotoxins of one or more bacteria, wherein said molecules preferably comprise a bacteriophage tailspike protein or comprise an antigen-binding region, and e. detecting whether any of said one or more molecules capable of selectively binding the O- polysaccharide moiety of one or more endotoxins of one or more bacteria binds to any endotoxin bound to the lipid binding protein.

3. The in vitro method of detecting endotoxins according to claim 1 or 2, wherein the lipid binding protein is lipopolysaccharide binding protein (LBP), is bactericidal / permeability-increasing protein (BPI) or is acyloxyacyl hydrolase in which the catalytic amino acid residue of the acyloxyacyl hydrolase (AOAH) is exchanged for a non-catalytic amino acid residue.

4. The in vitro method of detecting endotoxins according to claim 2 or 3, wherein the one or more molecules capable of selectively binding the O- polysaccharide moiety of one or more endotoxin of one or more particular bacteria comprise a reporter moiety, preferably a fluorescent or luminescent protein moiety, or an enzymatic protein moiety such as horse radish peroxidase or alkaline phosphatase, preferably capable of generating a chromogenic or chemiluminescent product.

5. The in vitro method of detecting endotoxins according to any one of the claims 2 to 4, wherein in the one or more molecules capable of selectively binding the O-polysaccharide moiety of an endotoxin of one or more particular pathogens are bacteriophage tailspike proteins, preferably are bacteriophage tailspike proteins in which the catalytic amino acid residue of the tailspike protein is exchanged for a non-catalytic amino acid residue, and more preferably are bacteriophage tailspike proteins according to SEQ ID 5 or SEQ ID6.

6. The in vitro method of detecting endotoxins according to any one of the claims 2 to 4, wherein in the one or more molecules capable of selectively binding the O-polysaccharide moiety of an endotoxin of one or more particular pathogens are antibody molecules comprising an antigenbinding region.

7. The in vitro method of detecting endotoxins according to any one of the preceding claims, wherein the lipid binding protein comprises an affinity tag, preferably a polyhistidine-tag, and / or the capture molecule immobilized on a substrate comprises an anti-polyhistidine-tag binding region, and / or wherein the capture molecule is preferably an antibody, such as an anti-His antibody or an antibody capable of binding wildtype lipid binding protein.

8. The in vitro method of detecting endotoxins according to any one of the preceding claims, wherein detecting whether any endotoxin binds to the lipid binding protein is carried out via enzyme-linked immunosorbent assay (ELISA) and wherein the lipid binding protein is coated on thesubstrate.

9. The in vitro method of detecting endotoxins according to any one of the claims 1 to 7, wherein detecting whether any endotoxin binds to the lipid binding protein is carried out via surface plasmon resonance (SPR) and wherein the lipid binding protein is captured on a capture molecule immobilized on the substrate.

10. The in vitro method of detecting endotoxins according to any one of claims 1 to 7 or 9, wherein the method further includes the step of releasing the lipid binding protein binding any endotoxins from the capture molecule immobilized on a substrate, to regenerate the capture molecule immobilized on a substrate.11 . The in vitro method of detecting endotoxins according to any one of the preceding claims, wherein the lipid binding protein is human acyloxyacyl hydrolase (AOAH) according to SEQ ID1 or SEQ ID2; or human bactericidal permeability-increasing protein (BPI) according to SEQ ID3 or SEQ ID4; or mouse acyloxyacyl hydrolase (AOAH) having a sequence according to SEQ ID7; or rabbit acyloxyacyl hydrolase having a sequence according to SEQ ID8, or murine lipopolysaccharide-binding protein having a sequence according to SEQ ID9 or SEQ ID10.

12. A device for use in the in vitro method of detecting one or more endotoxins in a sample according to any one of the preceding claims, comprising a substrate and a lipid binding protein coated on said substrate or a lipid binding protein captured on a capture molecule immobilized on said substrate.

13. The device according to claim 12, wherein the lipid binding protein is coated on a substrate and wherein the device is an ELISA well microplate, preferably an ELISA 96-well microplate, which ELISA well microplate is preferably configured to be read in a microplate reader apparatus.

14. The device according to claim 12, wherein the lipid binding protein iscaptured on a capture molecule immobilized on said substrate and wherein the device is an SPR chip, which SPR chip is preferably configured to be read in a surface plasmon resonance (SPR) apparatus.

Images