Patent Application
20100092968
The present invention relates to polypeptides comprising a enzymatically non-active cell wall binding domain of an endolysin or another cell wall lysing enzyme, and a sequence according to SEQ ID NO: 1 or derivatives thereof, wherein the polypeptide besides the cell wall binding domain comprises no further domains of an endolysin or of another cell wall lysing enzyme, as well as means for their preparation. The present invention further relates to methods for binding, enriching, removing from a sample, capturing and/or detecting bacteria, particularly gram positive bacteria.
Gram positive bacteria are wildly spread in the environment. They can be found in samples like soil, water, plant material, feces, but also in humans and animals. A whole series of pathogen germs, e.g. of the species listeria, bacillus, clostridium, staphylococcus, streptococcus, enterococcus, micrococcus or mycobacteria is particularly relevant in the food sector as well as in prevention, diagnostics and therapy of infection diseases in humans and animals.
Bacteria of the group Bacillus cereus represent microorganisms of major economic and medical importance as well as prominent relevance in the sector of bioterrorism. The bacteria are closely related to each other within a group; a large amount of which are sequenced (Rasko et al, FEMS Microbiol. Reviews, 2005, 29, 303-329). They are wildly spread in nature and present in different kinds of food, mostly of plant origin. They are aerobic living, moving, rod shaped gram-positive bacteria. Due to their resistant endospores, they are able to survive different methods, used to cure food, e.g. like drying or heating. Frequently contaminated food is mainly starch containing food, cereals, rice, spices, vegetables and ready-to-eat products. Meat can be contaminated by using contaminated spices. Milk products are frequently contaminated, because the spores survive pasteurisation and unrestricted proliferation is subsequently possible. The Bacillus cereus group consists of six closely related species and includes besides the food germ Bacillus cereus the extremely dangerous human pathogen germ Bacillus anthracis, which is enormously relevant in the sector of bioterrorism, the insect pathogen Bacillus thuringiensis (widely spread by microbiological insecticides on the basis of B. thuringiensis), the rhizoid-like Bacillus mycoides as well as Bacillus pseudomycoides and Bacillus weihenstephanensis.
Listeria are human and animal pathogen bacteria, frequently present in food, particularly in fish, meat and milk products. The genus listeria comprises six different species with 16 different serotypes. Although only a small portion of the food related diseases is caused by listeria (about 1% in USA), almost 30% of the annually fatal diseases, caused by food pathogens, are cause by this germ. Affected are mainly immune suppressed persons, e.g. older people, diabetes patients, cancer patients and/or AIDS patients. Pregnant women and the yet unborn child represent 25% of all cases of listeriosis patients.
Staphylococcus and enterococcus are currently the most problematic pathogens associated with infection diseases, since they increasingly develop multi resistant germs (e.g. MRSA—multi resistant Staphylococcus aureus and VRE—vancomycin resistant Enterococcus), leading to dramatic developments, i.e. bad disease prognosis as well as an explosion of the costs, mainly in stationary health care.
Potential pathogens are frequently present in very small cell numbers in the infection state and in food area and additionally accompanied by an interfering background flora. On the one hand methods are needed for efficiently removing relevant gram-positive germs and on the other hand methods for selectively enriching these germs, e.g. to achieve a sensitive detection.
Criteria for a good detection method are sensitivity, rapidity, reliability as well as a simple and cost efficient application.
Initially, there is basically always the enrichment of the organisms which should be detected. Generally this is achieved in conventional methods in a multi-step process. A primary enrichment mostly occurs with non- or low-selective liquid nutrient media. Subsequently a selective secondary enrichment occurs, followed by a primary isolation which frequently takes place on selective agar. Single colonies are enriched again in subcultures and subsequently detected with diverse detection methods. These conventional methods take a very long time until the positive detection of a germ. The standard detection times for listeria are more than 4-7 days according, to ISO 11290-1:1996/FDAM 1:2004(E) and 4-7 days according to FDA and USDA/FSIS. The enrichment of bacteria of the Bacillus cereus group takes 1.5 to 2.5 days according to standard methods (USFDA-method, chapter 14; ISO—7932:2004), the subsequent detection 2 days and a more detailed discrimination within the Bacillus cereus group again 2-3 days.
In food industry the detection time is an important sector concerning the short shelf-live of some kind of foods and the cost intensive storage, which is necessary until it has been made sure that the sample is not contaminated. Moreover, cost intensive product recalls can consistently be observed, if contaminated goods are delivered ahead of schedule prior to the receipt of the control results. In health care long detection times are also problematic, since appropriate specific treatment methods on a safe basis can also be performed not until the identification of the pathogen germ.
As faster alternatives compared to conventional enrichment and detection methods, antibody based methods were frequently used (e.g. U.S. Pat. No. 6,699,679, US 2004/0197833, UA 2006/0211061, Fluit et al., 1993, Appl Environ Microbiol., 59, 1289-1293, Jung et al., 2003, J Food Prot., 66, 1283-1287, Hawkes et al., 2004, Biosens. Bioelectron., 19, 1021-1028). The application of sugar binding lectines as receptors for the carbohydrate portion of the bacterial surface was also considered. However, these are mostly too unspecific to select certain bacteria from a mixed culture and frequently exhibit agglutination problems based on their multimeric binding properties. The recovery rate of bacteria was also relatively poor using antibody based methods, the rate was in a range from 5% to 25%. Further disadvantages of immunomagnetic separation methods (IMS) are besides the poor recovery rate, insufficient sensitivity at low contamination rates, cross reactions with other cells and frequently agglutination problems with antibody coated beads. Additionally it is relatively difficult to obtain antibodies against intact bacteria. Although these methods are very promising in case of pure cultures, they exhibit significant problems in case of mixed cultures or complex matrices such as foods.
As alternative to antibodies the use of cell wall binding domain (CBDs) from bacteriophages peptidoglycan hydrolases for binding of gram-positive bacteria are known in the state of the art (Loessner et al., 2002, Mol. Microbiol., 44, 335-349). EP 1147419 and WO2004/088321 use CBDs for the detection of cells, wherein the CBDs are bound to a solid phase and generally carry a marker.
EP 1399551 describes a method for the selective purification of gram-negative bacteria cells or cell fragments using bacteriophage capsid proteins or bacteriophage tail proteins. In this case the bacteria are bound in a 2-step-method, first binding of the binding molecules to the target cells occurs, followed by the immobilisation of the complexes to solid carriers. The immobilisation occurs by coupling the bacteriophage tail proteins with the help of his-tag, biotin or strep-tag to functionalized surfaces of solid carriers. Especially for this 2-step-method an efficient, fast and non-covalent coupling of the binding protein-target cell-complex to a functionalized carrier is of crucial relevance particularly if the target cells shall be separated from the sample together with the solid carrier.
U.S. Pat. No. 5,252,466 discloses a method to prepare fusion proteins including a tag for in vivo biotinylation and are therefore easy to purify. In this case the biotinylation domains are for example the 1.3S subunit of the Propionibacterium shermanii transcarboxylase, tomato-biotin-protein, the α-subunit of the Klebsiella pneumoniae oxalacetate decarboxylase or the Escherichia coli biotin carboxyl carrier protein, expressed in the same reading frame together with a biotin ligase of the plasmid and therefore are biotinylated in vivo. With the help of a phage display system a proteolytic more stable minimal version of the biotinylation domain of klebsiella oxalacetate decarboxylase was developed, suitable to purify and detect respective fusion proteins (Stolz et al., 1998, FEBS-Lett. 440, 213-217). U.S. Pat. No. 5,874,239 also claims a method for biotinylation of fusion proteins suggesting a number of tags, so called “Avi-tags”, which are, with a length of 13 to 50 amino acids preferably about 20 amino acids, shorter than the biotinylation domains of klebsiella oxalacetate decarboxylase.
Thus, the problem underlying the present invention is to provide more efficient and productive methods and the means to perform said method for fast, simply and efficiently binding, enriching, removing, capturing and detecting gram positive bacteria.
The problem is solved by the subject matter defined in the patent claims.
The following figures illustrate the invention.
A P22 like phage tail protein from Salmonella was cloned at the N-terminus with Strep-tag, Avi-tag and JS-tag and expressed in E. coli HMS 174 (DE3) and E. coli JM83, respectively. The samples were loaded on a 12% SDS-gel and stained with Coomassie. The experiment is described in experiment 1. M: Marker, P: Pellet; S: Supernatant; +: induced; −: not induced; nb: not boiled. The arrows indicate the positions of the phage tail protein monomers and trimers, respectively.
Three different bacteriophage tail proteins (Tsp) from salmonella phages were cloned as fusion proteins with JS-tag and expressed in E. coli HMS174 (DE3). The samples from the cell lysate were loaded on a 12% SDS-gel and stained with Coomassie. Lane 1: Marker (molecular weight from top: 118 kDa, 85 kDa, 47 kDa, 36 kDa, 26 kDa, 20 kDa), lane 2: Pellet (P), not induced, lane 3: supernatant (S), not induced, lane 4-12 (all post induction), lane 4: Felix like tail protein (Felix-Tsp, 48 kDa), P, lane 5: Felix-Tsp, S, boiled, lane 6: Felix-Tsp, not-boiled, lane 7: P22 like tail protein (P22-Tsp, 67 kDa), P, lane 8: P22-Tsp, S, boiled, lane 9: P22-Tsp, S, not-boiled, lane 10: ε15-like tail protein (ε15-Tsp, 93 kDa), P, lane 11: ε15-Tsp, S, boiled, lane 12: ε15-Tsp, S, not-boiled. The arrows indicate the expected positions for the phage tail protein monomers and SDS-resistant trimers (in not boiled samples).
A campylobacter phage tail protein (putative tail fibre protein H for Campylobacter jejuni, Acc. No: ZP01067412) was cloned with N-terminal Avi-tag and JS-tag (version 5b), respectively, in the plasmid pET21d and expressed in E. coli HMS 174 (DE3) and E. coli BL21(DE3), respectively. The experiment is described in experiment 1. A 9% Coomassie-stained SDS-gel is depicted with different samples from the expression and solubility test. P: Pellet; S: Supernatant; (i) induced; −: not boiled; M: Marker. The arrows indicate the positions of the monomers and trimers (in not boiled samples), respectively, of the phage tail proteins after induction.
It is depicted, which part of the introduced listeria is still bound in the binding assay after chemical biotinylation. Protein concentrations of 0.5 μg/ml, 1 μg/ml, 2 μg/ml and 5 μg/ml were used in the test. Incubation with NHS-biotin occurred for 0 min, 20 min, 60 min and 120 min. Avi-CBD served as control.
The time dependence is comparatively analysed according to which very low concentrations of L. monocytogenes can be detected in camembert with the different methods. 5 portions of camembert (25 g each) were contaminated with 0, 2, 4, 15 and 46 CFU (colony forming units) and tested after 4 h, 6 h, and 24 h of concentration according to the 1-step, 2-step or ISO (ISO: 11290-1:1996 FDAM1)-method. For the 1-step and 2-step method the Strep-tag-GFP-CBD511_f2 was used as a specific ligand. The values are determined from 4 experiments each. (0: no colonies on plate, X: <10 colonies, XX: 10-30 colonies, XXX: >30 colonies).
The cell binding capacity of different constructs was tested with the listeria strains ScottA according to the 2-step-method. The following constructs were used: JS-GFP_CBD511_f3 (circle), JS-CBD511_f3 (square) and Avi-GFP CBD511_f3 (triangle). Given is the portion of the listeria attached to the magnetic beads in percent of the introduced cells. All experiments were performed twice and the average values were determined.
The left picture depicts a Coomassie-stained gel of the purification of Avi-GFP-CBD511_f2, the right depicts the end product of the purification of JS-CBD511_f2. The experiment is described in experiment 4b. M: Marker; L: Load on the column; F: flow through; W: Wash fraction; 3-8: fractions containing Avi-GFP-CBD511_f2. 10, 1, 0.1 μg/ml: loaded protein concentrations for JS-CBD511_f2. The positions for the bands for Avi-GFP-CBD511_f2 and JS-CBD511_f2 as well as BirA are indicated.
It was analyzed, at which concentration of specific binding protein a maximum cell binding is achieved. As binding protein JS-CBD511_f3 was introduced in the concentrations 0 μg/ml, 0.02 μg/ml, 0.1 μg/ml, 0.5 μg/ml, 1 μg/ml, 2 μg/ml and 3 μg/ml. The experiment is described in experiment 4b.
Illustrated is the portion of the bound bacteria cells (Listeria ScottA) in dependence of the introduced amount of specific binding protein JS5b-CBD511_f2. Concerning one part of the proteins, BirA was coexpressed in an additional plasmid (triangles), concerning the other part no coexpression occurred with BirA (rhombuses). Furthermore, biotin was additionally added (filled symbols) in 2 experiments, whereas this remained undone in 2 experiments (open symbols).
Bacillus cereus bacteria were bound using a 2-step method with two different CBDs (CBDBa and CBD21) either via hexa-his-tag to nickel-NTA-magnetic beads or via JS-tags to streptavidin-magnetic beads. Depicted is the portion of specific bound bacteria compared to the totally introduced bacteria (concentration 3×103 CFU/ml).
Listeria monocytogenes EGDe cells were concentrated with the method according to the present invention from different media and PBST-buffers, respectively. The experiment is described in experiment 8.
The specific concentration of Bacillus cereus with the constructs strep-tag-CBDBa and JS-tag-CBDBa was analyzed in comparison at biotin concentrations of 0.01 μM, 0.1 μM and 1 μM.
JS-CBD511_f3 as well as magnetic streptavidin beads were incubated in a temperature range from −20° C. to 37° C. and subsequently introduced in cell binding tests with Listeria monocytogenes ScottA.
Listeria were concentrated by using JS5b-CBD511_f2 from Salami, which was contaminated with small amounts of Listeria monocytogenes ScottA (5 CFU/25 g) and after 17 h and 20 h of pre-incubation in LEB-FDA medium, respectively. In
Bacillus cereus is concentrated using the specific JS-tag-CBDBa from a mixture of bacteria (Bacillus cereus (DSMZ345), Salmonella tennessee, Listeria monocytogenes (ScottA), Staphylococcus aureus, E. coli HMS174 (DE3)). Given is the number of bound Bacillus cereus cells in percent compared to the totally recovered cells bound to the magnetic beads (black bars) or in the supernatant and the washing fraction (shaded bars), respectively. On the one hand, the cells were plated on a complete medium (CASO, Merck) to maintain also the other cells from the mixed culture, on the other hand cells were plated on selective plates for Bacillus cereus (PEMBA). As controls experiments were performed such that magnetic beads but no binding protein was added.
The appearance of pathogen germs from the Bacillus cereus group is a problem particularly concerning pre-cooked, highly carbohydrate containing foods like ready-to-eat products or re-warmed rice. For this reason, pre-cooked rice was used as a food sample which only needs to be heated for two minutes in the microwave oven to finish cooking. The food sample was diluted with medium and homogenised and spiked with Bacillus cereus cells in concentrations of 102, 103 or 104 CFU/ml. TSPB-medium was used as a control.
Human blood was spiked with Bacillus cereus (DSMZ345) in a concentration of 103 CFU/ml. Citrate, EDTA or heparin were previously added to the blood samples to inhibit blood coagulation and diluted 1:1 with PBST-buffer. JS-tag-CBD21 (10 μg/ml) (black bars), controls without protein (shaded bars).
The green fluorescence of the spacer GFP was herein used as a marker to visualize the binding of JS-tag-GFP-CBD3626 to the Clostridium bacteria and the magnetic beads, respectively.
JS-tag-CBD-constructs according to the present invention JS-CBDPitti20 (grey bars), JS-CBDOpf (white bars) and JS-CBDUSA (black bars) were used. It is depicted how many percent of the introduced cells were specifically bound in the cell binding test. In experiment 20 the performance of the experiments is described. The given numbers are numbers from the PROFOS strain collection. The Staphylococcus strains are isolates from patients as well as DSMZ and ATCC strains, respectively.
strains: C=Staphylococcus carnosus, D=Staphylococcus epidermidis, E=Staphylococcus equorum, F=Staphylococcus haemolyticus, G=Staphylococcus saprophyticus, H=Staphylococcus sciuri, I=Staphylococcus simulans, J=Staphylococcus warneri, K=Staphylococcus xylosus.
The performance and the principal of the peroxidase test are described in experiment 21.
The performance of the test is described in experiment 22. Two JS-tag-CBD-constructs according to the present invention were introduced: JS-CBDEF0355 (grey bars) and JS-CBDEF1293 (white bars). The buffer control without addition of protein (black bars) is also depicted. The given numbers are strain numbers from the PROFOS strain collection. It is a matter of patient isolates as well as DSMZ and ATCC strains of the species Enterococcus and Staphylococcus.
The invention is described below.
The term “bacteria removal” as used herein means complete or partial removal of bacteria from the sample material.
The term “bacteria enrichment” as used herein means the concentration of the bacteria based on the initial concentration present in the sample. The enrichment has to be carried out generally such that a respective detection step allows an unambiguously positive or unambiguously negative conclusion.
The term “capture” of bacteria means the procedure of the extraction and binding of the bacteria from a sample with the help of the capability of the “CBDs” to specifically bind bacteria from the sample.
The term “sample material” or “sample” as used herein comprises all solutions in which bacteria should be detected or from which bacteria should be removed. Examples for suitable samples are: Aqueous solutions and mixtures of water and organic solvents, foods, media, blood, blood products, plasma, serum, urine, other body fluids, diagnostic samples, protein solutions, water ethanol mixtures, process solutions such as washing solutions for the analysis of the contamination of medical devices or in food processing. Furthermore, comprised are also solutions in which non-aqueous solid substances were solved, which are to be analysed or to be isolated, e.g. proteins, DNA, RNA, sugar, salts, foods, food-media homogenates, medicaments, vaccines, environmental samples, faeces, organic and inorganic chemicals, e.g. NaCl, MgCl2, purine, pyrimidine.
The term “endolysin” as used herein means an enzyme, which serves in its original function for the release of new phages in the end of a respective phage reproduction cycle. Such an endolysin can be for example encoded by the phage genome in nature. These endolysins consist of at least one enzymatic active domain each and an enzymatic non-active domain bound to the cell wall of the respective host cell. Additionally, endolysin has to be understood to comprise also the similarly composed autolysins. These are bacteria encoded in nature and consist also of at least one enzymatic active cell wall hydrolysing domain and an enzymatic non-active domain binding to the cell wall of the target bacterium. Phage encoded endolysins and bacteria encoded autolysins are frequently homologous to each other (Garcia et al., 1988, Proc. Natl. Acad. Sci. USA, 85, 914-918) and the modules are exchangeable and can even be combined with each other in chimera from bacteria and phage encoded sequences (Diaz et al., 1990, Proc. Natl. Acad. Sci. USA, Vol. 87, 8125-8129). Therefore, besides the cell wall binding domains from phage endolysins also respective non-enzymatic active cell binding domains from other cell wall lysing enzymes can be used according to the present invention, e.g. autolysins, bacteriocines or phage tail proteins.
The term “CBD” as used herein means polypeptide-domains and -sequences, respectively, deriving from endolysins or other cell wall lysing enzymes and being responsible for the specific binding of the endolysins or the other cell wall lysing enzymes to the bacteria cell wall. These are enzymatically non-active.
The term “polypeptide according to the present invention” as used herein means a polypeptide comprising an enzymatically non-active cell wall binding domain of an endolysin or of another cell wall lysing enzyme (CBD) and a sequence according to SEQ ID NO:1 or derivatives thereof, wherein the polypeptide comprises besides the cell wall binding domain no further domains of an endolysin or another cell wall lysing enzyme, particularly no complete enzymatic active domain (EAD) of an endolysin.
The term “JS-tag” as used herein means a polypeptide sequence comprising a sequence according to SEQ ID NO:1 or derivatives thereof. The JS-tag derives from the biotin acceptor domain of the α-subunit of Klebsiella pneumoniae oxalacetate decarboxylase and contains the consensus sequence MKM (K is biotinylated) so that the polypeptide can be biotinylated in vivo by the protein biotin ligase. Compared to the complete α-subunit of Klebsiella pneumoniae oxalacetate decarboxylase the JS-tag is truncated. One possible minimal sequence for the JS-tag comprises 66 amino acids corresponding to amino acids 529 to 594 of the Klebsiella pneumoniae oxalacetate decarboxylase (SEQ ID NO:2). The term “JS-tag” also comprises derivatives of the sequence according to SEQ ID NO:1. Derivatives as used herein comprise such sequences still at least 80% homologous to SEQ ID NO:1. Examples for such derivatives are depicted in SEQ ID NO: 2-18.
The term “directed immobilisation” as used herein means that the CBDs are immobilised on suitable surfaces via biotin as specific coupling agent, e.g. via magnetic particles supplied with streptavidin or avidin or other carriers.
The term “surface” or “carrier” as used herein comprises all materials to which a coupling or adhesion of a CBD molecule and polypeptide according to the present invention, respectively, is possible directly or indirectly, e.g. to glass surfaces, to chromatography materials, such as agarose, sepharose, acrylate, to plastic surfaces such as polystyrene, polyethylene, polycarbonate, polypropylene, polysulfone, polymethyl methacrylate, to filter materials or membranes such as cellulose, cellulose acetate, nitrocellulose, PVDF, to magnetic or non-magnetic particles made of glass, latex, plastic, metal, metal oxide.
The term “1-step method” as used herein means a method, in which specific binding proteins, e.g. a polypeptide according to the present invention, were immobilized to a suitable carrier or a surface either directed or undirected already before adding the sample. After incubation of the immobilized binding proteins with the sample, the bacteria-binding protein-carrier-complex is separated from the sample and then optionally washed.
The term “2-step method” as used herein means a method, in which non-immobilized specific binding proteins, e.g. a polypeptide according to the present invention, are contacted and incubated with the sample. The formed bacteria-binding protein-complexes are subsequently contacted with a suitable carrier or a surface such that the bacteria-binding protein-complexes are bound via the binding proteins with the help of the biotinylated affinity tags to the carriers or surfaces. Subsequently, the bacteria-binding protein-carrier-complexes are separated from the sample and optionally washed. The binding proteins are modified with a polypeptide or a chemical group in a way that they bind specifically to a carrier or a surface supplied with the respective binding partner of the polypeptide or the chemical group.
The term “polypeptide” as used herein means a polypeptide chain of at least five amino acids.
The term “bacteria-polypeptide-complex” or “polypeptide-bacteria-complex” as used herein means a complex in which bacteria and the polypeptide according to the present invention (the polypeptides according to the present invention) are present.
The term “carrier-polypeptide-bacteria-complex” as used herein means a complex in which bacteria, the polypeptide according to the present invention (the polypeptides according to the present invention) as well as a carrier (-material) are present.
The present invention relates to a polypeptide comprising
The present invention particularly relates to a polypeptide according to the present invention, which is biotinylated. In a particular embodiment, the cell wall binding domain of an endolysin or another cell wall lysing enzyme (CBD) within the polypeptide according to the present invention exhibits the capability to specifically bind gram-positive bacteria. The polypeptide according to the present invention is suitable to bind, enrich, remove from a sample, capture and/or detect bacteria.
The present invention therefore relates to the use of a polypeptide according to the present invention to bind, enrich, remove from a sample, capture and/or detect bacteria.
Thus, the present invention further relates to a method for the binding, the enrichment, the removal, the capture and/or the detection of bacteria from a sample comprising the steps:
In the method according to the present invention the duration of the incubation of the sample with the respective polypeptides according to the present invention (step a) and the incubation of the bacteria-polypeptide-complex with the carrier material (step b), respectively, has to be adjusted to the respective sample and can vary in an embodiment between several seconds and about 24 h. Step a) of the method according to the present invention, in which the functionalized, i.e. biotinylated polypeptides according to the present invention bind to bacteria, is generally faster than step b), in which the biotinylated polypeptide-bacteria-complex is immobilized on the biotin-binding-carrier. Suitable incubation times for step a) of the method according to the present invention are particularly about 0.1 min to about 10 min, for step b) particularly about 10 min to about 60 min or if necessary also overnight. Whereas in step a) of the method according to the present invention it is generally enough to mix the added polypeptides according to the present invention and the sample thoroughly, it could be necessary during the incubation with the carrier material (step b), e.g. after adding carrier material, to roll the sample container in a lying position to achieve an as efficient as possible binding to the carrier.
At very low bacterial concentrations in the starting material a preincubation phase in a suitable nutrient medium is possibly necessary, to achieve an efficient enrichment and therefore obtaining a suitable sample for the method according to the present invention. Samples containing solid components such as food samples can be homogenised prior to the use in the method according to the present invention and taken up in suitable solutions before they are used in the method according to the present invention.
In the method according to the present invention a carrier is used, which is supplied with a biotin-binding substance, i.e. the carrier was functionalized. The biotin-binding substance should be capable to bind biotin with high affinity. Particularly suitable binding partners, i.e. biotin-binding substances, on the surface of the carriers are for example streptavidin, avidin and biotin-binding variants thereof such as monomeric avidin, avidin with partially acetylated amino groups or partially esterified carboxyl groups thereof. Hydrophilic surfaces are preferred compared to hydrophobic surfaces since they are generally more suitable to bind proteins and to tend to agglutination to a lesser extent.
The bacteria are enriched by separation of the bacteria from the rest of the sample. The bacteria can be enriched in the described method for example on a magnetic basis, via chromatographic methods or in the batch-method. Preferred is a magnetic enrichment since this method, compared to other methods, is very fast, can be miniaturized and also automated. But also the chromatographic method or the batch-method can be used, particularly if the method is primarily not used for the following bacteria detection, but is aiming, e.g. at the isolation of a larger amount of bacteria and to continue to work with these isolated bacteria.
During the enrichment on magnetic basis the polypeptide-bacteria-complexes are incubated with suitable magnetic particles as carrier in step b) of the method according to the present invention. The magnetic particles can exhibit diameters in certain embodiments in the range form about 0.1 μm to about 100 μm. However, preferred are smaller particles with a diameter between about 0.5 μm to about 5 μam, particularly preferred are particles with the diameter of about 0.8 μm to about 2 μm, since smaller particles sediment to a lesser extend and therefore providing a better mixture and exhibiting a relatively larger surface compared to larger particles as well as exhibiting high recovery rates. Examples for suitable magnetic particles are MagPrep-streptavidin particles (Merck), streptavidin-magnetic-particles (Roche), streptavidin-beads (Dynal), streptavidin coupled silica-beads (MicroCoat), streptavidin-coupled polyvinyl-alcohol-beads (PA-streptavidin-beads, Microcoat). After step b) of the method according to the present invention, the carrier-polypeptide-bacteria-complex is separated from the sample magnetically by applying a magnetic field. Suitable magnetic separators are available for example from the companies Ambion, Ademtech, Bilatec, BioLabs, Dynal, Polysciences and Promega.
During the enrichment with the help of the batch-method the bacteria containing sample is primarily incubated with the polypeptide according to the present invention, subsequently carrier material is added, suitable to bind high affinity biotin, mixed and again incubated together. Subsequently the carrier-polypeptide-bacteria-complex can be centrifuged from the sample, sedimented or filtrated. Preferred is the concentration via the batch-method particularly at very low bacteria concentrations.
Alternatively, the polypeptide-bacteria-complexes can be separated from the sample and enriched by applying them on a chromatography column containing biotin-binding column material.
The polypeptide-bacterial-complexes can be separated from the functionalized carrier by displacement, e.g. by adding an appropriate amount of biotin or by the adjustment of conditions, which highly denaturate proteins, such as about 8 M guanidinium chloride or about pH 1.5 and by adding biotinidase, cleaving biotin from the peptides, respectively. It is also possible that only the bacteria are separated from the biotinylated polypeptides according to the present invention by choosing conditions, at which the polypeptides according to the present invention do not bind to their receptor on the bacteria surface anymore. Since different polypeptides according to the present invention are used for the invention, the exact conditions therefore have to be tested in the individual case. This can for example be carried out by introducing a fluorescing marker and monitoring under the fluorescence microscope, wether the previously bound bacteria still fluoresce since their surface is covered with polypeptides according to the present invention. However, generally a change of the ion strength to a very high or a very low ion strength, a change of the pH to very acidic or basic, e.g. 50 mM sodium phosphate and pH 11 for 5 min, the addition of detergents or chemical denaturants such as urea or guanidinium chloride, or a combination of the mentioned possibilities are suitable to prevent the CBD bacteria-binding, since this is the specific protein-protein interaction. However, for a lot of applications, e.g. the subsequent plating and counting of the colonies deriving from the bound bacteria, a separation of the magnetic particles from the bacteria is not necessary at all.
The present invention further relates to methods for the detection of bacteria in a sample. The detection of bacteria comprises further steps subsequently to the above described method steps for enrichment. Depending on the kind of the bacteria, which should be detected, a set of techniques is known by a person skilled in the art, leading to the wished result. A choice of suitable detection methods is mentioned below. Bacteria can for example be detected in the complex together with the carrier and the biotinylated polypeptide according to the present invention or after releasing from the carrier material via selective growth conditions, e.g. plating and incubating on selective media plates. Furthermore, a detection of the bacteria is possible by nucleic acid based methods, i.e. detection of the nucleic acids of the bacteria, e.g. PCR, RT-PCR, PCR-RFLP, rep-PCR-fingerprinting, NASBA, DNA-hybridisation methods for example for certain toxins or other pathogenicity factors, multi-locus sequence typing (MLST), rRNA-comparisons. Further possible is a detection of the bacteria cell wall and their components, respectively, e.g. via cell binding domains of endolysins or antibodies or via FTIR, and the detection of bacteria components, respectively, e.g. proteins via ELISA or enzymes via their activity or multi-locus enzyme electrophoreses (MEE). The bacteria detection is also possible via ATP, which is contained in bacteria, e.g. in a bioluminescence assay via detection using a bacteria specific bacteriophage, e.g. for listeria A511-luxA, (see U.S. Pat. No. 5,824,468). The bacteria can further be detected in the carrier-polypeptide-bacteria-complex or after removal from the carrier material via another specific CBD coupled to a marker. A set of examples are therefore depicted in EP1147419. A conventional detection of a combination of microbiologic, morphologic and/or biochemical detection methods is also possible.
The detection of bacteria components, e.g. of proteins is preferably performed via ELISA or similar techniques (e.g. VIDAS). For the performance of these methods it is necessary to disrupt bacteria prior to the actual detection. This can for example be performed with a lysis protein such as lysozyme or a bacteria specific endolysin. Lyses proteins can for example be selected from the following group (references in brackets are either accession numbers for the NCBI-database (number-letter combination) or publication citations):
The cell lyses can be supported by different further additives such as the addition of proteases e.g. proteinase K and the use of heat, preferably 5 min at about 56° C., subsequently 5 min at about 94° C. or such as the addition of detergents, preferably triton, SDS, tween, Na-desoxycholate or solvents such as DMSO, isopropyl alcohol, ethanol, butanol, chloroform.
The polypeptides according to the present invention comprise polypeptide domains/-sequences of endolysins and autolysins, respectively, the so called CBDs, wherein the polypeptides according to the present invention do not exhibit enzymatic active cell wall hydrolysing regions anymore. The lack of the enzyme activity is necessary to functionally separate the bacteria in the complex with the carrier. A lysis of the bound bacteria and therefore a release of the cell contents preferably occurs purposely, only after the separation if this is necessary e.g. for the subsequent detection reaction. The polypeptides according to the present invention therefore comprise in a particular embodiment besides the cell wall binding domain no further domains of an endolysin, particularly no enzymatic-active domain (EAD) of an endolysin, and in particular embodiments also no further sequences of an endolysin. However, under certain conditions a minor hydrolytic rest activity of the used polypeptide fragments can be tolerable. However, this depends on the respective application, i.e. it has to be checked to what extent the rate of the cell hydrolases can be reconciled with the total duration of the application such that a sufficient amount of intact cells are captured. How a potential rest activity of the CBDs can be detected in a hydrolyse assay is for example described in Loessner et al. (1996, Appl. Environ. Microbiol. 62, 3057-3060).
The polypeptide according to the present invention exhibits in a particular embodiment derivatives of SEQ ID NO:1. Examples for such derivatives are depicted in SEQ ID NOs:2-18, said examples exhibit the following as exemplified variation of SEQ ID NO:1:
The above mentioned derivatives of SEQ ID NO:1 turned out to be advantageous concerning the use in the method according to the present invention.
The modular organisation of endolysins in C-terminal domains (CBDs) responsible for the specific cell binding and N-terminal domains (EADs) including the enzymatic active centre were already described in 1990 by Garcia et al. (1990, Gene, 86, 81-88). The concept of the CBDs was continued in Loessner et al., 2002, (Mol. Microbiol., 44, 335-349) and Loessner, 2005, (Curr. Opin. Microbiol, 8, 480-487). A multitude of CBDs are already described in the state of the art. Frequently, the enzymatic active-domain (EAD) is located at the N-terminus and the CBD at the C-terminus—however, there are exceptions (e.g. Garcia et al., 1988, Proc. Natl. Acad. Sci. USA, 85, 914-918; Loessner, 2005, Curr. Op. Microbiol., 8, 480-487). The EAD is generally well defined and can be found and located relatively easy via sequence and homology comparisons with other hydrolases such as amidases, endopeptidases, glycosidases, transglycosylases, muramidases, with sequence analyses softwares, known in the state of the art, and respective databases with conserved sequence motives (e.g. CDD (Marchler-Bauer et al., 2005; Nucleic Acids Research, 33, D192-D196); Pfam (Finn et al., 2006, Nucleic Acids Research 34, D247-D251) or SMART (Schultz et al., 1998, Proc. Natl. Acad. Sci. USA 95, 5857-5864, Letunic et al., 2006, Nucleic Acids Res 34, D257-D260)). During the identification of the CBD part of endolysins, already known motives for cell binding domains in many cases are found, providing a sound indication for the provision of the CBD sequence for the polypeptides according to the present invention. In the group of endolysins, which bind to streptococci, the CBD is relatively easy to find since mostly about 20 amino acids long choline-binding-motives with conserved aromatic residues (CW_binding—1, pfam01473) appear, which occur frequently in multiple repeats (see Garcia et al., 1990, Gene, 86, 81-88). The about 40 amino acids long LysM domain (pfam01476) can also be partially found in CBDs. This is a widely spread peptidoglycan-binding-module with conserved secondary structure (Bateman & Bycroft, 2000, J. Mol. Biol., 299, 1113-1119). SH3b-domains and SH3—3, SH3—4 or SH3—5 domains (smart00287, pfam08239, pfam06347, pfam08460), respectively, the prokaryotic counterparts to the eukaryotic and viral Scr homology domains, SH3, (Ponting et al, 1999, J. Mol. Biol., 289, 729-745) can also often be found as cell binding motives and CBDs, mainly in staphylococci and enterococci. The peptidoglycan-binding-domain (PG_binding—1, pfam01471; PG binding 2, pfam08823) consists of 3 helices which can be found frequently also N-terminal of the EAD. However, sometimes no direct relationship of the CBD portions to other related bacteriophage endolysins or also to other carbohydrate binding proteins can be found and unique sequence motives or structure modules can barely be defined indicating a CBD. In such cases, the relationship can be determined via the EAD. As a basis for the polypeptide according to the present invention serves in this cases, besides the CBDs known from the state of the art, the portion of an endolysin, which is not occupied by the EAD. The rest of this endolysin (i.e. endolysin minus EAD) can directly be understood and used as CBD, as far as it exhibits cell binding function. However, it can be useful in some embodiments of the present invention to use shorter fragments (e.g. because they exhibit a higher stability) as far as they still exhibit cell binding function. The functional test for a CBD is the detection of the cell binding to the respective: bacteria. Different exemplified assays suitable for that purpose are described in the experiments and figures. Besides an efficient cell binding, the expression rate, solubility, stability and simple purification are further features which should be taken into account concerning the definition of the peptide portion functioning as CBD. Methods to test these features are known from the state of the art by the person skilled in the art. Some examples are therefore also described below in the experiments. Purposely planned CBD portions are for example orientated according to structural standards, which can be assessed by the person skilled in the art on the basis of secondary structure predictions, potential domain linkers and 3D-models. Suitable exemplified methods are for example described in the following publications: Gamier et al., 1996, Methods in Enzymology 266, 540-553; Miyazaki et al., 2002, J. Struct. Funct. Genomics, 15, 37-51; Altschul et al., 1997, Nucleic Acids Res. 17, 3389-3402; Schwede et al., 2003, Nucleic Acids Research 31, 3381-3385. Lund et al, CPHmodels 2.0: X3M a Computer Program to Extract 3D Models. Abstract at the CASP5 conference A102, 2002.
Basically all CBDs known from the state of the art and all CBDs deriving from endolysins according, to the above described method can be used for the polypeptides according to the present invention and in the method according to the present invention.
In a particular embodiment of the invention the cell wall-binding domain of the peptide according to the present invention is selected from the group of cell wall binding domains of the following endolysins and other cell wall lysing enzymes, respectively, consisting of Ply511 (Q38653), Ply 500 (Q37979), Ply 118 (Q37976), PlyPSA (1XOV_A), EGDe (NP—466213), PLyL (1YB0_A, B and C), PlyG (YP—891193), PlyPH (Yoong et al., J. Bac. 2006, 188, 2711-2714), PlyB (2NW0_A and B), PlyBa (CAA72266), Ply21 (CAA72267), Ply12 (CAA72264), of the Enterococcus faecalis V583 prophage endolysins, Ply3626 (WO 03/066845), lysins from Cl. perfringens strain 13 (BAB81921) and strain SM101 (YP—699489), ΦP1 lysin (EP1300082), PlyV12 (NP—049942), PlyC (NP—852017), PlyGBS (AAR99416), Cpl-1 (P15057), Cpl-7 (P19385), Cpl-9 (P19386), Pal Amidase (P19386), Twort Amidase (CAA69021), S. aureus phage PVL amidase (UniProt 080064), P68 lys16 (NP—817332), ΦSA2usa endolysin (YP—494080), Phi11 (NP—803306) and Phi12 Endolysin (NP—803355), cell wall hydrolyses of the Staphylococcus aureus phage Phi 11 (NP—803302), phage B30 endolysin (AAN28166), phage 168 endolysin (M J Loessner et al., J Bacteriol. 1997 May; 179(9): 2845-2851), LysK (O'Flaherty et al., J. Bac., 2005, 187, 7161-7164), S. simulans Lysostaphin (AAB53783), S. capitis ALE-1 endopeptidase (BAA13069), phage PhiNIH1.1 cell wall hydrolase (NP—438163), LytM (AAB62278), Atl (BAA04185), LytA from Streptococcus pneumoniae (CAJ34420), from Staphylococcus aureus strain PS47 deriving from peptidoglycan hydrolase (AAA26662), enterolysin A from Enterococcus faecalis (Q9F8B0), ami autolysin from L. monocytogenes (Milohanic et al.; Infection and Immunity, August 2004, p. 4401-4409, Vol. 72, No. 8), lactobacillus lysin, e.g. lysin of the phage A2 (AJ251788.2 or Q9MCC8), phage. PL-1 amidase (Q9MCC6) (both L. casei) etc.
Particularly suitable for the binding of listeria are fragments of the endolysin Ply511, for the binding of bacillus fragments of the endolysins PlyBa, Ply21, and Ply12, for the binding of clostridium fragments of Ply2636, for the binding of staphylococci fragments of the ΦSA2usa endolysins, of the Staphylococcus bacteriocine lysostaphin, of the lysostaphin like ALE-1, of the staphylococcus self-isolated plyOpf, plyPitti20, plyPitti26 and similar proteins of homologous prophages and for the binding of enterococci fragments from endolysins of the prophage from Enterococcus faecalis V583 as well as CBDEF0355 and CBDEF129.
Examples for CBD sequences which can appear in the polypeptides according to the present invention are:
In a particular embodiment the polypeptide according to the present invention comprises no further sequences of an endolysin besides the cell-wall binding domain. The polypeptide according to the present invention preferably consists of only one CBD and an additional polypeptide sequence according to SEQ ID NO:1 or derivatives thereof, optionally coupled by a linker or spacer.
The person skilled in the art exhibits knowledge about the sequences and structures of domain linker sequences and about their prediction (e.g. George and Heringa, 2003, Protein Eng. 15, 871-879; Bae et al., 2005, Bioinformatics, 21, 2264-2270), respectively. Domain linker sequences are frequently characterized by a relatively high portion of hydrophilic amino acids since they are generally less structured and exposed to solvents such as the short linker sequence AAKNPN (SEQ ID NO: 37) of the listeria endolysin PlyPSA (Korndorfer et al., 2006, J. Mol. Biol., 364, 678-689). Polyglycine linkers are also used traditionally, however, they are often protease sensitive. A special kind of the hydrophilic unstructured linkers is proline and threonine rich sequences also occurring as natural linkers, e.g. TPTPPNPGPKNFTT from Enterolysin A (SEQ ID NO: 36). Proline and threonine rich linker sequences can simply be described by the consensus motive (PT)xP or (PT)xT wherein x is an integer from 1 to 10. Croux et al. (1993, Molec. Microbiol., 9, 1019-1025) describe so-called junction zones between the N— and C-terminal domains, and thus the EADs and CBDs of the endolysins. These mostly relatively short areas are natural linker sequences and are also suitable to link the CBD modules recombinant to the JS-tags with the help of suitable cutting sites to a polypeptide according to the present invention. Particularly suitable examples for specific linkers are depicted, e.g. in SEQ ID NO: 34-38.
The sequence of a polypeptide according to the present invention can be composed as follows:
Several exemplified sequences for polypeptides according to the present invention are depicted in SEQ ID NO: 39-53.
The invention further relates to a non-enzymatic active cell wall binding domain, namely the CBD of Ply21. Surprisingly the hydrolytic non-active CBD21 exhibits a host-spectrum concerning bacteria binding, which comprises, besides almost all bacteria of the bacillus group (except for B. polymyxa and B. sphaericus), also further representatives from important groups of gram-positive bacteria such as staphylococci, enterococci, streptococci and even listeria. Since CBDs usually exhibit a relatively small host spectrum (Loessner 2005, Curr. Opin. Microbiol, 8, 480-487), this feature of broad host specificity for a CBD is very unusual. Therefore, the CBD21 is capable to bind all tested bacteria from the bacillus cereus group and additionally solve the object to generally enrich gram-positive bacteria, especially those deriving from groups, in which a lot of pathogen germs can be found such as staphylococci, streptococci, enterococci, micrococci, bacilli and listeria. In another aspect, the present invention therefore relates to a polypeptide comprising the sequence as depicted in SEQ ID NO: 19, but besides this cell wall binding domain does not comprise further domains of an endolysin. The polypeptide preferably comprises no complete enzymatic active domain of an endolysin. More preferably, the polypeptide comprising the sequence as depicted in SEQ ID NO: 19, does not comprise further sequences of an endolysin.
Due to the broad applicability of the CBD21, the present invention relates also to the use of the CBD21 according to the present invention comprising the sequence according to SEQ ID NO:19, but besides this cell wall binding domain does not comprise further domains of an endolysin, to bind, enrich, remove from a sample, capture and/or detect bacteria selected from the group consisting of staphylococci, streptococci, enterococci, micrococci, bacilli and/or listeria (see table 3).
The polypeptide portion of the polypeptides according to the present invention defining a suitable CBD can additionally be linked to a polypeptide portion, serving as spacer and optionally as marker at the same time. Since CBDs represent regions from bigger proteins, the endolysins, they are generally relatively small with about 100 to 300 amino acids or for certain cell binding motives even less. Therefore, it can be useful in a preferred embodiment to introduce a spacer between the CBD domain and the group responsible for the immobilisation to the carrier. This can prevent that the CBD is denatured by the immobilisation and that it looses its binding capability to the cells. For the binding of the polypeptide-bacteria-complex in the 2-step-method it can be important that the groups responsible for the immobilisation to the surfaces become better accessible if they are not directly coupled to the CBD. The spacer is preferably a well defined, well expressible stable protein module, which interacts with other proteins and surfaces as less as possible (e.g. GFP (green fluorescent protein), MBP (maltose binding protein), GST (glutathione s-transferase)). A particularly suitable example is GFP and variants thereof. Since GFP is highly fluorescing, it is also suitable as a marker. For this reason the polypeptide according to the present invention can be for example monitored during the method. In the functional test, i.e. binding test, a binding of the CBDs and of the polypeptides according to the present invention to the bacteria as well as a binding to the carrier can easily be detected. Further modifications can also serve as a marker, as for example proposed in EP 1147419.
To turn the CBD into the polypeptide according to the present invention the CBD has to be present with a tag, the so-called JS-tag, within a fusion protein. The extremely strong bond between biotin and its binding partner streptavidin and avidin (10−15 M; Gonzales et al., 1997, J. Biol. Chem., 272, 11288-11294), respectively, is advantageous for the functioning of the above mentioned 2-step-method, which turns out to be even better compared to the known 1-step-method, known from the state of the art (e.g. EP1147419). Other tags also suitable to bind proteins to functionalized surfaces are for example the his-tag or the strep-tag (His-Tag 10−6-10−8 M; Nieba et al., 1997, Anal. Biochem., 252, 217-228; Strep-Tag ˜10−6 M, Voss & Skerra, 1997, Protein Eng., 10, 975-982). The binding of the bacteria-polypeptide-complex to the carrier is more efficient in the 2-step-method which affects the binding time as well as results in a lower possible loss of bacteria under difficult conditions, e.g. in food samples, as well as in a higher sensitivity. In a choice of possible biotinylated coupling groups, the JS-tag is proofed to be more preferred. A chemical biotinylation does on the one hand not lead to a defined biotinylation at a certain position, with which a directed immobilisation of the CBDs to the carrier would be possible, which is wished for the functionality of the binding proteins. On the other hand proteins are thereby often inactivated. This can particularly apply to the relatively small protein domains of the CBDs. The Avi-tag, representing something like a minimal sequence, wherein the minimal sequence is still to be biotinylated in fusion proteins in vivo, turned out to be also less suitable compared to the JS-tag since higher protein amounts had to be introduced, to achieve an efficient binding of the bacteria to the magnetic beads. The biotinylation domains proposed in U.S. Pat. No. 5,252,466 for the fusion with proteins are relatively big compared to the Avi-tags. Thus, the biotinylation domain of the Klebsielle pneumoniae oxalacetate decarboxylase for example comprises 595 amino acids, which is disadvantageous concerning both the expression yield and insofar that the large fusion portion is relatively protease sensitive and therefore relatively unstable.
The JS-tag and derivatives thereof turn out to be very good biotinylation tags in combination with CBDs to bind, enrich, remove, capture and detect bacteria in samples. These are segments from the α-subunit of the Klebsiella pneumoniae oxalacetate decarboxylase and derivatives thereof containing the consensus motive (MKM) for the in vivo biotinylation. The polypeptide became more stable, e.g. towards proteolyses, compared to the complete biotinylation domain and is easier to handle as affinity tag, e.g. in cloning, expression and purification. The minimal sequence for the JS-tag is 66 amino acids long, which correspond to the amino acids 529 to 594 of the Klebsiella pneumoniae oxalacetate decarboxylase plus methionine as a start. Particularly suitable are sequences described under SEQ ID NO:1-18. In Cronan (1990, J. Biol. Chem., 265, 10327-10333) it is emphasized that conserved proline and alanine rich regions located N-terminal of the MKM motive should take important structural function for the biotinylation of the lysin by the biotin ligase. This region of the α-subunit of the Klebsiella pneumoniae oxalacetate decarboxylase is even particularly developed with a 22 amino acid long region of P and A. However, it turned out that this region is not necessary for the biotinylation in the described system since the above mentioned minimal sequence does not contain this region and is yet very efficiently biotinylated. It turned out that additionally to the amino acids 529 to 594 of the Klebsiella pneumoniae oxalacetate decarboxylase very short peptides (MVGA) provide a very good N-terminal starting sequence (see SEQ ID NO:8-10 and 16-18).
The fusion between the JS-tag and the CBD, and an additional intermediately introduced spacer module, respectively, can be carried out N-terminally as well as C-terminally of the CBD. Concerning the polypeptides according to the present invention the N-terminal fusion is preferred since the CBD portion of the endolysin is usually located C-terminal and the JS-tag (optionally plus spacer module) can structurally substitute the missing EAD of the endolysin. Since the biotinylation domains in proteins, which are biotinylated in vivo, are almost only located at the C-terminus, it is not obvious that these also function well if used N-terminally. For this reason, in Cronan (1990, J. Biol. Chem., 265, 10327-10333) only C-terminal fusions of proteins with a biotinylation domain of the 1.3 S subunit of the Propionibacterium shermanii transcarboxylase were used. If no spacer molecule is used, the JS-tag can be linked to the CBD via a linker (embodiment see above) or also without linker such that only one to three amino acids are introduced to obtain a restriction cutting site for cloning. The sequence AGAGAGAGS or AGAGAGAGSEL (SEQ ID NO:34 or 35) turned out to be an exemplified suitable linker peptide. However, other linker sequences of proteins with a known structure are linker sequences for relatively unstructured peptides (e.g. proline and threonine rich linkers such as (PT)3T(PT)3T(PT)3) can also be used. An example for a PT-rich linker is TPTPPNPGPKNFTT (SEQ ID NO:36). An example for a (short) hydrophilic linker is AAKNPN (SEQ ID NO:37). Another example for a linker which can be used in the present invention is AGAGAGAEL (SEQ ID NO:38).
The polypeptide according to the present invention for the use in the method according to the present invention can be biotinylated thereby that it is biotinylated in vitro with the help of a biotin ligase under conditions known by the person skilled in the art. Surprisingly, it also turned out that compared to the invention described in U.S. Pat. No. 5,252,466 a coexpression of the biotinylated fusion protein with the biotin ligase (BirA) is not necessary in the preparation of the biotinylated polypeptides according to the present invention in bacteria cells, since the biotinylation also works in the absence of external biotin ligase. Interestingly, in complete medium such as LB not even the addition of biotin to the medium is necessary. Fusion proteins of JS-tag and CBD are efficiently biotinylated in commercially available E. coli expression strains, e.g. BL21 (DE3), HMS174 (DE3), JM83 without the additional coexpression of BirA or even without addition of biotin. In contrast to the proposed use of the biotinylation tags in U.S. Pat. No. 5,525,466 as means for an easier purification of proteins it is not necessary to purify the herein described fusions of JS-tag and CBD via affinity columns for biotin, but can be purified in a conventional way such as via cation or anion exchange chromatography, hydrophobic chromatography, fractionated ammonium sulphate precipitation, etc. On the one hand, this is of advantage since the respective affinity material for biotin is very expensive and regularly exhibits only a low capacity, e.g. chromatography material which carries streptavidin or streptactin coupled, on the other hand the fusion proteins are difficult to release from the affinity material since the binding between biotin and its binding partners is very efficient. This leads to problems in the described purification method. Thus, the target protein is eluted delayed from the column (“smears”) and possibly denatured during elution.
The stability of the constructs according to the present invention depends to a certain extent on the specific features of the used CBDs. The fusion constructs consisting of JS-tag, CBD and optionally linker, spacer and marker, respectively, as well as the functionalized carrier such as magnetic beads can be stored over a longer period of time without loosing their binding capability. Storage is possible in a temperature range from about −20° C. to about 37° C. Preferred is storage at temperatures of about −20° C. to about 10° C. Regularly storage should be carried out nearly neutral pH values (pH 6-7), but storage at pH-values up to pH 10 is also possible if admitted by the CBD portion. Suitable buffer systems for the storage are e.g. 100 mM sodium phosphate buffer, pH 6 to pH 10, 2 mM EDTA or 10 mM imidazol, 100 mM NaCl, pH 7. The addition of generally stabilizing agents such as glycerol or ammonium sulphate with for example 30% of saturation has a positive effect on the storage capability.
With the method according to the present invention under use of the polypeptide constructs according to the present invention basically all gram-positive bacteria such as clostridii, bacilli, listeria, staphylococci, lactobacilli, enterococci, aerococci, pediococci, streptococci, mycoplasma, leuconostoc bind and can therefore be enriched, captured, immobilized and optionally detected. The mode of application depends on the mode of the used samples, as defined above. Particularly suitable is the method to enrich and detect potentially pathogen bacteria from food samples. However, the method is also suitable to enrich and detect pathogen bacteria from medical or diagnostic samples. Furthermore, it is suitable to remove or detect gram-positive bacteria from samples, in which they are undesired for example in pharmaceutical or cosmetic preparations and process solutions. The method according to the present invention safes a lot of time concerning the enrichment in contrast to conventional enrichment methods such as ISO-methods. In combination with a magnetic separation using functionalized magnetic particles for capturing the bacteria-polypeptide-complexes separation methods can be substituted, which are complex and difficult to automatise, e.g. centrifugation steps.
The invention further relates to nucleic acids as well as vectors encoding for the polypeptides according to the present invention as well as cells expressing the nucleic acids and vectors, respectively. The person skilled in the art is able to prepare suitable nucleic acids and vectors encoding for the polypeptides according to the present invention with procedures known in the state of the art. Amino acid sequences of the polypeptides according to the present invention can for example be derived from suitable nucleic acid sequences based on the genetic code. An optimized use of codons can here optionally be considered depending on the chosen expression system. The person skilled in the art is also able to choose suitable vectors, e.g. to ensure the expression of the polypeptides according to the present invention via the above mentioned nucleic acid.
Surprisingly it turned out that the nucleic acid sequence in the N-terminal part of the JS-tag of the translated polypeptide sequence is important for an effective expression. AT-rich sequences showed a significantly more efficient expression compared to GC-rich sequences under maintenance of the amino acid sequence. It is assumed that the development of secondary structure elements at the beginning of the transcribed RNA influences the efficiency of the translation. Three variants (SEQ ID NO: 54 to 56) of these AT-rich sequences turned out to be particularly suitable for fusions between JS-tag and a subsequent CBD (see table 1). Therefore, in a preferred embodiment the part of the sequence of the nucleic acid encoding for the JS-tag in the polypeptides according to the present invention starts with a sequence selected from SEQ ID NO:54 to 56.
The method according to the present invention and the polypeptide fragments according to the present invention, respectively, are characterized by the following advantages:
The present invention further relates to a kit comprising a carrier supplied with a biotin binding substance such as streptavidin or avidin as functional groups, further comprising at least one variant of the polypeptide fragments according to the present invention having a CBD fused with JS-tag as well as the buffer solutions, e.g. washing buffer, elution buffer and/or lyses buffer, necessary for the enrichment and optionally the detection of gram-positive bacteria.
The following examples illustrate the invention and are not to be considered to be limiting. Unless otherwise indicated, molecular biological standard methods were used as described for example in Sambrook et al., 1989, Molecular cloning: A Laboratory Manual 2. Edition, Cold Spring Harbor Laboratory Press, Cold Spring Harbor, New York.
Experiment 1: Expression, Solubility and Functional Assembly of Phage Tail Proteins with Biotinylation Tags
The result of the experiment is depicted in
It has been shown in all three sub-experiments that no efficient amount of functional protein can be obtained with phage tail proteins having specific biotinylation tags. Partially, the expression rate is very poor, partially a major part of the proteins are insoluble and it is not possible to obtain a high portion of native protein, characterized by SDS-resistant, oligomeric forms. Only concerning the P22 similar phage tail protein, bands of monomers and native trimers could be detected on SDS gels. Concerning the other two proteins only very weak monomeric bands could be detected on western-blots such that the position could be localized and the basic functioning of induction and expression could be verified.
Stock solutions of 1 mg/ml each of CBD511 (in PBS; 20 mM sodium phosphate pH 7.4, 120 mM sodium chloride) and NHS-biotin (in DMSO) were prepared. 120 μl NHS-biotin solution was added to 1200 μl protein solution and mixed thoroughly. 300 μl of each sample was taken immediately (0 min value) or after 20 min, 60 min or 120 min and added on ice to 30 μl 1 M tris, pH 8 to stop the reaction. All samples were dialysed against PBS-buffer. The protein concentration was determined of all samples by measuring the absorption and the degree of biotinylation was determined using the HABA-test. The degree of biotinylation was 1.5 to 2.5 biotin molecules per CBD-molecule. The cell binding test was performed with the strain Listeria monocytogenes Scott A, introduced in 500 μl of test sample (buffer PBST; 20 mM sodium phosphate pH 7,4, 120 mM sodium chloride, 0.1% tween 20) in a concentration of 104 CFU/ml. Subsequently, biotinylated CBD511 was added in the concentrations 0.5 μg/ml, 1 μg/ml, 2 μg/ml and 5 μg/ml, respectively and incubated for 1 min. JS-tag CBD511 served as control. MagPrep-streptavidin particles (Merck) were added to 50 μg/ml and the samples were incubated for 20 min at room temperature in an overhead rolator. After 5 min of magnetic separation the supernatant was taken and the magnetic particles were washed once with 500 μl of PBST-buffer (10 min). After a second magnetic separation, the magnetic particles were added in one buffer volume and plated on oxford-plates (undiluted and diluted 1:10). As a control the pooled supernatants after the 1. and 2. magnetic separation were plated and counted after one night. The experiment is depicted in
Experiment 3a: Detection of Listeria in Camembert with the 1-Step-, 2-Step- and the ISO-Method
300 g of camembert from a supermarket were divided sterile in portion units of 25 g and stored in Stomacher bags at −80° C. One portion unit was analysed concerning the presence of listeria according to regulation ISO: 11290-1:1996 FDAM 1. If no listeria contamination could be detected, 5 portion units were thawed at room temperature and infected with different amounts of L. monocytogenes ScottA. Therefore an overnight culture was diluted 1 to 5 and incubated up to an OD600 of about 1 at 37° C. Subsequently serial dilutions were prepared in sterile PBST (20 mM sodium phosphate pH 7.4, 120 mM sodium chloride, 0.05% tween). The portion units were contaminated with 0, 1-10, 11-50, 50-100 and 100-500 CFU/25 g camembert and stored overnight at 4° C. For accurate determination of the cell numbers, duplicates of the dilutions were plated on Oxford Agar (Profos AG), the plates incubated for 24 h at 37° C. and counted. 225 ml Fraser ½ medium (Profos AG) were added sterile to the portion units, homogenised for 1 min in the Stomacher and incubated at 30° C. After incubation time of 4 h, 6 h and 24 h, one ml of each sample was taken.
1-step-method: 300 μg/ml of magnetic particles (Dynabeads Epoxy) coated with Strep-tag-CBD511_f2 were added to 1 ml of homogenate and the sample was incubated for 20 min in an overhead rolator at room temperature.
2-step-method: 5 μg of Strep-tag-GFP-CBD511_f2 fusion protein was added to 1 ml homogenate and mixed shortly. Subsequently MagPrep-Streptavidin particles (Merck) were added to 50 μg/ml and the samples were incubated for 20 min in the overhead rolator at room temperature.
The particle-listeria-complexes were subsequently collected in a magnetic field at the vessel wall and the supernatant was removed. The particle-listeria-complex was washed 3× in 1 ml PBST (20 mM sodium phosphate pH 7.4, 120 mM sodium chloride, 0.05% tween) for. 10 min in the overhead rollator, collected in the magnetic field at the vessel wall and each supernatant discarded. The particle-listeria-complexes were resuspended in 100 μl PBST and plated on Oxford-Agar (Profos AG). After 24 h and 48 h at 37° C. the plates were counted and the portion of the listeria attached to the magnetic particles was calculated in percent of the introduced cells. In parallel the contaminated samples were analysed concerning listeria according to the rule ISO: 11290-1:1996 FDAM 1. Therefore, 100 μl were added to 10 ml of Fraser medium (Profos AG) at the given time points, 24 h at 37° C. incubated in the rolator and subsequently plated on Oxford-Agar (Profos AG). All samples were performed in quadruplets.
It has been shown that with the 1-step-method as well as the 2-step-method the necessary concentration times are significantly shorter compared to the method according to ISO: 11290-1:1996 to detect minor listeria contamination in camembert. Concerning the shorter enrichment time, the results of the two-step-method is better compared to the 1-step-method.
Experiment 3b: Detection of Listeria from Mozzarella
225 ml FDA-medium was added to each of the 25 g of mozzarella and the portions were sterile homogenized in Stomacher bags. The samples were incubated over night at 30° C. Listeria of the strains EGDe (serotype 1/2a) and ScottA (serotype 4b) were added in a concentration of 500 CFU/ml. Prior to the listeria detection the samples were buffered each with 1/10 volumes of PBST.
1-step method: 300 μg/ml of magnetic particles (Dynabeads M270 Epoxy) coated with JS-tag-GFP-CBD511_f3 was added to 1 ml homogenate and the sample was incubated for 20 min in an overhead-rolator at room temperature.
2-step method: 0.5, 2, 5 or 10 μg of the JS-tag-GFP-CBD511_f3 fusion protein was added to 1 ml homogenate and mixed briefly. Subsequently, MagPrep-streptavidin particles (Merck) were added to 50 μg/ml and the samples were incubated for 20 min in an overhead-rolator at room temperature.
The particle-listeria-complexes were subsequently collected in a magnetic field at the vessel wall and the supernatant was removed. The particle-listeria-complex was washed 1×in 1 ml PBST for 10 min in the overhead rollator, collected in the magnetic field at the vessel wall and each supernatant discarded. The particle-listeria-complexes were resuspended in 100 μl PBST and plated on Oxford-Agar (Profos AG). After 24 h at 37° C. the plates were counted and the portion of the listeria attached to the magnetic particles was calculated in percent of the introduced cells.
All approaches were performed twice.
It has been shown that with the help of the JS-tag-GFP-CBD511_f3 fusion protein, listeria can be isolated from foods. Concerning mozzarella, this works significantly better with the strain EGDe than with ScottA. Slightly higher concentrations of protein should be used in foods to achieve a high binding efficiency.
Experiment 4: Cell Binding Capacity of JS-Tag-CBDs
Experiment 4a: Comparison of the Cell Binding Capacity of JS-Tag and Avi-Tag Constructs
The cell binding capacity of the different constructs was tested with the listeria strain ScottA according to the 2-step method. The following constructs were used: JS-GFP_CBD511_f3, JS-CBD511_f3 and Avi-Tag-GFP_CBD511_f3. Since the constructs are different in length, equal molar amounts of binding protein were used. The given amounts of fusion proteins were added to 1 ml of test sample (listeria from fresh pre-culture in a concentration of 104 CFU/ml, PBST buffer) and mixed briefly. Subsequently, MagPrep-streptAvidin particles (Merck) were added to 50 μg/ml and the mixtures were incubated for 20 min in an overhead rolator at room temperature. The particle-listeria-complexes were subsequently collected in a magnetic field at the vessel wall and the supernatant was removed. The particle-listeria-complex was washed 1× in 1 ml PBST for 10 min in the overhead rollator, collected in the magnetic field at the vessel wall and each supernatant discarded. The particle-listeria-complexes were resuspended in 100 μl PBST and plated on Oxford-Agar (Profos AG). After 24 h at 37° C. the plates were counted and the portion of the listeria attached to the magnetic particles were calculated in percent of the introduced cells. All approaches were performed twice and the mean values were calculated. The experiment is depicted in
Experiment 4b: Purification of Avi-GFP-CBD511_f2 and JS-CBD511_f2
Both proteins were purified via cation exchange chromatography after expression in E. coli HMS174, cell harvest, lysis and ammonium sulphate precipitation. Avi-GFP-CBD511_f2 was coexpressed with BirA.
Experiment 4c: Concentration Dependence of the Cell Binding to the Magnetic Particles
It has been analysed from which concentration on of specific binding protein the maximum cell binding is achieved. JS-CBD511_f3 was used as binding protein in the concentrations 0 μg/ml, 0.02 μg/ml, 0.1 μg/ml, 0.5 μg/ml, 1 μg/ml, 2 μg/ml and 3 μg/ml. The experiment was performed analogously to experiment 4a. The result is depicted in
Experiment 5: Optimisation of the Nucleotide Sequence in the N-Terminal Region of the JS-Tag
Although it has been shown that JS-tag CBD constructs containing the nucleotide sequence of the C-terminal part of the Klebsiella pneumoniae oxalacetate decarboxylase are biotinylated in vivo, they exhibited a very poor expression yield for the proteins (see also table 2). As a result the nucleotide sequence encoding the first amino acids of the Klebsiella sequence and the starter peptide (MVGA) additionally introduced by our cells was optimized concerning the expression yield without changing the amino acid sequence. From a whole set of different sequence proposals, 5 variants encoding three different nucleotide sequences turned out to be particularly suitable. All variants have in common that they are AT-rich compared to the original sequence if the choice of codons allows this for the respective amino acid. The variants (variations are highlighted in grey) are summarized in table 1.
The variants JS4a, JS5b, JS5d, JS10a and JS10c turned out to be particularly suitable for the expression of fusion constructs of JS-tag and CBD. There were slight differences concerning the expression between the single variants in dependence of the used CBD portion. However, basically all can be used.
Experiment 6: Coexpression of JS-Tag-CBD-Constructs with BirA
It has been tested if a coexpression of BirA (Biotin ligase) is necessary to efficiently biotinylate JS-tag-constructs in vivo.
The construct JS5b-CBD511_f2 was expressed in the expression strain E. coli BL21 (DE3) in the vector pet21a. If BirA was coexpressed the plasmid pACYC184-BirA was additionally present. Fresh LB-medium (in 2 l flasks, total amount 4 l to 10 l) was inoculated with overnight cultures of the expression strain, the cells were induced at an OD600 of a about 0.4 to 0.6 with 1 mM IPTG and harvested after 4 h. 50 μM biotin was added additionally to a portion of the samples during induction. Each construct was tested twice. After harvesting the cells were centrifuged and lysed. The purification was carried out by fractionated ammonium sulphate precipitation and subsequent cation exchange chromatography. The purity of the proteins was documented in SDS-gels.
The expression yield is significantly better concerning the construct JS-5b (see experiment 5), which is optimized in nucleotide sequence, compared to the original sequence from Klebsiella pneumoniae. The coexpression of BirA does basically not increase the yield of protein based on the introduced amount of bacteria cells.
Cell Binding Tests with the JS-Tag Constructs with and without Coexpression of BirA.
JS5b CBD511_f2 was introduced in cell binding tests with the listeria strain ScottA. 1 ml samples with a listeria concentration of 103 CFU/ml were used. As magnetic particles Streptavidin-magnetic-particles (Roche) were used in a concentration of 0.05 mg/ml. Otherwise the performance of the experiment was carried out according to the 2-step-method as described in experiment 3b. Constructs from the expression with and without BirA as well as with and without the addition of additional biotin (50 μM) were analysed. The result is depicted in
Experiment 7: Comparison of the Bacteria Binding with His-Tag-CBDs and JS-Tag-CBDs
Bacillus cereus (DSM345) was freshly inoculated from a preculture and grown at 30° C. in TS-medium up to an OD600 of about 1. The bacteria were introduced in the test sample (1 ml) in a concentration of 3×103 CFU/ml. For the his-tag-constructs the buffer was nickel buffer A (20 mM Na-phosphate, 500 mM NaCl, 20 mM imidazol, 0.1% tween 20, pH 7.4), for JS-tag-constructs the buffer was PBST. His-tag-CBDBa and JS-tag-CBDBa were added in a concentration of 1 μg/ml, his-tag-CBD21 and JS-tag-CBD21 were added in a concentration of 0.12 μg/ml and incubated at room temperature for 5 min. Subsequently, Ni-NTA-agarose-beads (Qiagen) and MagPrep-streptavidin-beads, respectively, were added in a concentration of about 8×106 particles/ml, rolling incubated for 20 min and separated for 5 min with the magnetic separator. The magnetic particles were washed with one sample volume of buffer. The washing solution and the supernatant with unbound bacteria were plated on CASO-complete medium plates, as well as the taken up magnetic particles with the bound bacteria. After about 18 h at 27° C. the colonies were counted. 2 experiments each were performed. Approaches without added protein served as controls. The experiment is depicted in
Whereas the bacteria were bound very specific with the JS-tag-constructs under the given conditions, the yield concerning bound cells was completely insufficient with the his-tag constructs. There was basically no unspecific binding of bacteria concerning both kinds of magnetic particles.
Experiment 8: Enrichment of Listeria from Different Media and Buffers, Respectively
Listeria monocytogenes EGD was enriched with Avi-CBD511_f2 and JS-CBD511_f2 (5 μg/ml each), respectively, from different media and PBST-buffers, respectively. The listeria were incubated in TB-medium up to an OD600 of about 1 and predilutions were prepared thereof in the given media or buffers. Listeria were used in a concentration of 104 CFU/ml in the test. The cell binding test was carried out according to the 2-step-method with 20 min of rolling incubation with 50 μg/ml MagPrep-streptavidin particles (Merck) and washing once with PBST. The bound and unbound cells were plated on Oxford-Agar and counted after 1 night. Given is the respective mean value of two experiments. The result of the experiment is depicted in
Experiment 9: Comparison of the Bacteria Binding in Biotin Containing Samples with Strep-Tag-CBDs and JS-Tag-CBDs
Biotin is frequently contained in food samples. Since the binding of JS-tags as well as strep-tags to streptavidin competes with biotin, it was analysed, which system is more suitable in biotin containing samples, to enrich bacteria specifically. Bacillus cereus (DSM345) was inoculated freshly from a preculture and grown at 30° C. in TS-medium up to an OD600 of about 1. Bacteria were introduced into the test approach (1 ml) in a concentration of 1×103 CFU/ml. As a test solution PBST was used with the given concentrations of biotin (0.01 μM, 0.1 μM, 1 μM). JS-tag and strep-tag CBDBa, respectively, were added in a concentration of 20 μg/ml and incubated for about 2 min at room temperature with the bacteria. StreptAvidin-PA-beads were added in a concentration of 50 μg/ml rolling incubated for 20 min and for 5 min separated with a magnetic separator. The magnetic particles were washed with one sample volume of buffer. The washing solution and the supernatant with unbound bacteria were plated on CASO-complete medium plates, as well as the taken up magnetic particles with the bound bacteria and incubated overnight at room temperature. After another 4 hours at 30° C. the colonies were counted. Two experiments each were performed. Samples without added protein served as controls. The result is depicted in
Experiment 10: Long Term Stability of the JS-Tag-CBD-Constructs Under Different Conditions
Stock solutions of JS-CBD-511_f3 (about 1 mg/ml) and streptavidin magnetic particles (about 1 mg/ml) were incubated under the given conditions and introduced into the cell binding test with listeria. The concentration of the magnetic particles in the 1 ml test (PBST buffer) was 50 μg/ml, the protein concentration as given and the used bacteria number was 104 CFU/ml.
Experiment 10a: Binding protein and magnetic beads were stored at −20° C., 4° C., RT (about 23° C.) and 37° C. in 100 mM sodium phosphate, pH 6-7, 2 mM EDTA up to 126 days and used in the given concentrations in the listeria binding test. It can be seen that only after 126 days and in this case mainly at an incubation of 37° C. the binding efficiency is significantly reduced.
Experiment 10b: Binding protein and magnetic particles were stored at −20° C., 4° C., RT (about 23° C.) and 37° C. in 10 mM Imidazol, pH 7, 100 mM NaCl plus 30% ammonium sulphate up to 74 days and used in the given concentrations in listeria binding tests. It can be seen that the binding efficiency after 74 days of incubation is slightly reduced at all temperatures, but is still over 50%.
Both buffer systems seem to be suitable for the long-term incubation of JS-CBD-constructs and suitable streptavidin magnetic particles.
The results are depicted in
Experiment 11: Enrichment of Listeria from Different Foods
The experiment is depicted in
Experiment 11a: An overnight culture of Listeria monocytogenes ScottA were diluted 1:5 in FDA-Oxoid medium and grown at 37° C. up to an OD600 of about 1. Milk and homogenized cheese, respectively, were diluted 1:10 in PBST and inoculated with listeria in a concentration of 104 CFU/ml. JS4a-CBD511_f2 was added to Eppendorf-Cups in the given concentrations and mixed with 1 ml sample solution each. As magnetic particles Streptavidin-magnetic-particles (Roche) were used in a concentration of 50 μg/ml and the samples were incubated for 10 min in an overhead rolator. The magnetic particles were separated for 5 minutes in the magnetic separator and subsequently washed ones with 1 ml PBST buffer and taken up in 1 ml of buffer. 100 μl undiluted and 1:10 diluted of the beads fraction and the pooled supernatants (after first magnetic separation and washing step) were plated on Oxford-Agar and incubated at 37° C. overnight. The listeria were counted the next day and calculated how many percent of the found bacteria were bound via the JS4a-CBD511_f2 to the magnetic beads, respectively. It has been shown that from milk as well as from cheese the introduced listeria cells were basically removed completely already at a concentration of specific binding protein of 0.2 μg/ml JS4b-CBD511_f2. 50% of the bacteria can be removed already at very low protein concentrations of 0.02 μg/ml.
Experiment 11b: 25 g each of smoked salmon and salami were homogenised, diluted 1:10 with LEB-FDA-medium, inoculated with Listeria innocua in a concentration of 104 CFU/ml and incubated in a Stomacher-bag for 1 h at room temperature. The used bacterial dilution was plated as a control. 1 ml of each sample of the filtrate of the Stomacher bag was taken and mixed with JS5b-CBD511_f3 (1 μg/ml). No protein was added to the controls. Directly subsequently, 400 μg/ml of PA-streptavidin beads were added and incubated for 20 min in the overhead rolator. The magnetic particles were separated for 5 min in the magnetic separator and subsequently washed once with PBST buffer and taken up in 1 ml of buffer again.
100 μl undiluted and 1:10 diluted of the beads fraction and the pooled supernatants (after first magnetic separation and washing step) were plated on Oxford-Agar and incubated at 37° C. overnight. The listeria were counted the next day and calculated how many percent of the found bacteria were bound each via the JS5b-CBD511_f3 to the magnetic particles. It has been shown that from smoked salmon as well as salami more than 90% of the bacteria can be bound.
Experiment 12: Detection of Listeria from Foods with the NASBA-Technique
An overnight culture of Listeria monocytogenes ScottA was diluted 1:5 in LEB-FDA-medium and grown at 37° C. up to an OD600 of about 1. Serial dilutions up to 102 CFU/ml were prepared thereof. 2×60 g of salami were weighed and contaminated with listeria in a concentration of 5 CFU/25 g food. To each of the 60 g samples, 540 ml of LX-medium (BioMerieux) was added, the samples were homogenised and incubated in Stomacher-bags for 17 h and 20 h, respectively, at 37° C. Subsequently the bacteria were captured with JS5b-CBD511_f2 in the 1 ml sample from the supernatant of the Stomacher-bag. The supernatant was additionally plated and counted as a control. The binding protein was added in a concentration of 1 μg/ml and incubated for 1 min with the sample. Subsequently PA-streptavidin magnetic particles were added in a concentration of 400 μg/ml and incubated for 20 min in an overnight rolator. The magnetic particles were separated for 5 min in an magnetic separator and subsequently washed 3 times with 1 ml each of TT-buffer (50 mM tris, pH 8.0, 0.1% tween 20). A portion of the magnetic particles with the bound bacteria were plated on Oxford-Agar and counted after an incubation of 15 to 20 h, to obtain the binding efficiency. The result is depicted in part A of the figure. A portion of the magnetic particles with the listeria bound via JS-tag-CBDs was introduced for detection in the NASBA. The cells on the beads were lysed with 5 μg/ml of listeria specific endolysin in 100 μl lyses buffer A (21% DMSO, 57 mM Tris, 0.4% Triton X100) for 15 min at room temperature. Subsequently the magnetic beads were separated for 5 min in the magnetic separator and 14 μl of the lysate, each of the NASBA-reaction was used. Test stripes, each for 8 samples were used already carrying precasted listeria-specific primer beads. To each sample 5 μl of an enzyme solution for NASBA was added and the reaction performed according to the manufacturer's protocol. As NASBA-system the Nuclisens EasyQ analyser (BioMerieux) was used together with the respective thermo-block. The data are evaluated by a time dependent fluorescence signal. After successful detection reaction the fluorescence signal increases after about 30 min of detection time to a higher level and remains there. 7 NASBA reactions were performed per experimental approach. The NASBA detection (after 17 h of incubation) is depicted in
It has been shown that in the described system already after 17 h of incubation of the foods with a listeria concentration of 5 CFU/25 g more than 99% of the bacteria can be bound and detected. The detection works conventional via selective plates (see
Comparable experiments were also successfully performed with smoked salmon, shrimps, brie, turkey- and pork-sausage and cream cheese from goat.
Experiment 13: Specific Cell Binding of JS-Tag-CBDs, Deriving from Bacillus Endolysins
For all bacteria strains overnight cultures were grown in complete medium. The overnight cultures were inoculated 1:20 to 1:5 in complete medium (e.g. CASO, LB, TS, TY) and grown further up to an OD)600 of about 1. Growth temperature for all bacillus bacteria was 30° C. for all other strains 37° C. Dilution series were prepared from the precultures. In the test approach (500 μl of volume) the bacteria were used in a concentration of 103 to 104 CFU. For the determination of the accurate cell numbers for each used dilution, the respective controls were plated and counted. JS-tag-CBDBa and JS-tag-CBD21, respectively, each were added in concentrations of 10 μg/ml and 1 μg/ml, respectively, to the test approach (buffer PBST) and incubated about 1 min with the cells. Subsequently the MagPrep-streptavidin particles (Merck) were added in a concentration of 50 μg/ml previously blocked with CASO-tween (0.1%)-solution. The sample with bacteria, binding protein and magnetic particles was incubated for 20 min in an overhead rolator at room temperature. The magnetic particles were separated for 5 min in a magnetic separator, subsequently washed once with 1 ml PBST and subsequently resuspended in buffer. 100 μl each of the resuspended beads fraction with the bound cells and of the pooled fraction of supernatant after magnetic separation and washing solution were plated. After drying, the plates were incubated overnight at the respective growth temperature and counted the next morning and corrected by respective dilution factors. Respectively given is how many percent of the totally bound bacteria are specifically bound via the JS-tag-CBD to the magnetic particles and are separated from the sample. Samples with no added specific binding protein served as control for potentially unspecific binding of the bacteria to the magnetic particles. As further control for the expected cell numbers served the bacteria predilutions, which were also plated and counted. Only experiments were evaluated showing a total number of the recovered cells in the range of 80% to 120% of the totally introduced cells. 2 to 4 experiments were performed per test.
An overview of the respective binding data with different bacillus strains and other gram-positive as well as gram-negative bacteria is depicted in table 3.
Bacillus cereus Group
B. cereus
B. cereus
B. cereus
B. cereus
B. cereus
B. cereus
B. cereus
B. cereus
B. cereus
B. cereus
B. weihenstephanensis
B. weihenstephanensis
B. weihenstephanensis
B. weihenstephanensis
B. thuringiensis
B. thuringiensis
B. thuringiensis tenebrionis
B. mycoides
B. mycoides
B. mycoides
B. pseudomycoides
B. pseudomycoides
B. badius
B. circulans
B. coagulans
B. firmus
B. licheniformis
B. megaterium
B. polymyxa
B. pumilus
B. sphaericus
B. subtilis
B. vallismortis
Listeria monocytogenes
Listeria monocytogenes
Staphyloccocus aureus
Staphyloccocus aureus
Staphyloccocus aureus
Staphyloccocus aureus
Staphyloccocus epidermides
Staphyloccocus epidermides
Staphyloccocus epidermides
Staphyloccocus epidermides
Staphyloccocus hämolyticus
Staphyloccocus hämolyticus
Staphyloccocus hämolyticus
Staphyloccocus hämolyticus
Enteroccocus faecalis
Micrococcus luteus
Streptococcus equi spp equi
Streptococcus mutans
Salmonella tenessee
E. coli HMS
As expected, both CBDs show no binding to gram-negative bacteria. However, the binding specificities of the CBDs both isolated from Bacillus cereus phages are unexpectedly completely different. It has been shown that CBDBA is very specific for bacilli from the Bacillus cereus group. Beyond this group only 2 streptococcus strains are recognized. However, CBD21 exhibits an exceptionally broad binding specificity for gram-positive bacteria. All representatives of the Bacillus cereus group are bound and additionally all further tested bacilli except for B. sphaericus and B. polymyxa. Exceptional is that also gram-positive bacteria from other families were bound. The 6 tested families, characterized in that they exhibit a high pathogen potential were all recognized. CBD21 is therefore particularly suitable to enrich, remove and detect pathogen bacteria in different areas where they pose a problem. In contrast to further tested fragments, which also derived from endolysin plyB21, the herein depicted fragment (SEQ ID NO: 19) was characterized by an increased stability and reduced aggregation susceptibility.
Experiment 14: Specific Binding of Bacillus cereus from a Mixture of Bacteria
Overnight cultures were grown in complete medium of the following bacteria: Bacillus cereus (DSM345), Salmonella tennessee, Listeria monocytogenes (ScottA), Staphylococcus aureus, E. coli HMS174 (DE3). The overnight cultures were inoculated in fresh medium and grown for about 3 h at 37° C. and 30° C. (bacilli), respectively, up to an OD600 of about 1. The strains were used in the test in a dilution of about 103 CFU/ml. The respective dilutions were plated and counted as a control. The test mixture has 1 ml and was performed in PBST (20 mM Na phosphate, 120 mM NaCl, pH 7.4, 0.1% Tween 20). As specific binding protein JS-tag-CBDBa was provided in a concentration of 20 μg/ml. For the control experiments only PBST was added instead of protein. 10 μl each of the bacteria pre-dilution was added such that the concentration of the bacteria in each sample was about 103 CFU/ml. The bacteria were incubated with the cells for 1 min. Subsequently 100 μg/ml magnetic PA-streptavidin beads (Microcoat) were added and the samples were rolling incubated for 20 min at room temperature. The magnetic particles were collected for 5 min in a magnetic separator and the supernatants were taken. The separated magnetic particles were washed once for 5 min with 1 ml PBST. The washing solution was pooled with the supernatants. The magnetic beads were again made up to 1 ml with PBST. 100 μl each from different dilutions were plated on the one hand on CASO-plates (casein, soy bean extract; complete medium, Merck) and on the other hand on PEMBA-plates (selective medium for Bacillus cereus, contains polymyxin B, egg yolk, mannitol) and incubated at 27° C. for about 18 h and subsequently counted. 2 mean values each were determined from 2 experiments. The total number of recovered cells resulted from the sum of cells bound to the magnetic particles and the cell which remained in the supernatant and the washing solution, respectively. The plated cell dilutions were used as a control. The result is depicted in
Experiment 15: Enrichment of Bacillus cereus from Carbohydrate Containing Foods
As food sample, precooked express rice (top-long grain express rice, Uncle Ben's) was used. 5 g of precooked express rice was sterile transferred into a Stomacher-bag, with 50 ml TSPB-medium (TS-complete medium with 0.01 mg/ml polymyxin B) and homogenised for 1 min. An overnight culture of Baciullus cereus (DSMZ345) was inoculated at 30° C. in TS-medium. 2 ml of the overnight culture was transferred in 10 ml fresh TSPB-medium and incubated until an OD600=0.8 was achieved. 990 μl of the food sample from the Stomacher-bag filtrate was spiked with 10 μl each of different dilutions of the bacilli precultures such that the bacillus concentration in the food sample was 102, 103, or 104 CFU/ml. TSPB-medium was spiked as a control. After 5 min of incubation at room temperature 10 μg JS-tag-CBDBa was added per sample and mixed thoroughly. The control samples obtained no protein. After about 1 min 400 μg of streptavidin of PA-magnetic beads (Microcoat) were added and the samples were rolling incubated for 20 min at room temperature. The magnetic beads were collected for 5 min in a magnetic separator and the supernatants were taken. The separated magnetic particles were washed twice for 5 min with 1 ml PBST. The washing solutions were pooled with the supernatants. The magnetic beads were again made up to 1 ml with PBST. 100 μl each of the different dilutions of the samples with the magnetic particles and the supernatants, respectively, were plated onto selective PEMBA-plates, subsequently incubated overnight at 30° C. and then counted. 2 plates each were plated in parallel. The result is depicted in
Experiment 16: Enrichment of Bacillus cereus from Blood
Bacillus cereus (DSMZ345) was grown overnight at 30° C. in TB-medium. Fresh TB-medium was inoculated 1:10 and the bacteria were grown up to an 0D600 of about 1. 0.5 ml each of citrate, EDTA or heparin blood was mixed with 0.5 ml PBST each. All additives in the blood served as anticoagulants. Citrate-blood: 0.106 M citrate was diluted 1:10 with blood. EDTA-blood: 1.2-2 mg EDTA/ml blood. Heparin blood: 10-30 I.U. heparin/ml blood. The 1 ml samples of the buffered blood were spiked with 10 μl each of Bacillus cereus preculture such that the bacteria concentration was about 103 CFU/ml. Therefore 5 μl of a JS-tag-CBD21-protein solution (2 mg/ml) was added and vortexed briefly. The controls obtained no protein. Subsequently 40 μl of magnetic particles (streptavidin-PA-beads, Microcoat, 10 mg/ml) were added and the samples were rolling incubated for 20 min at room temperature. The magnetic particles were separated in a magnetic separator for 5 min. The supernatant, which contains the non-bound bacteria was taken and the beads washed once with 1 ml of PBST (5 min). The washing solution was pooled with the supernatants after magnetic separation. 1 ml PBST was added to the magnetic beads with the bound complex of JS-tag-CBD and bacteria. 100 μl each of a 1:10 dilution of the pooled supernatants and the solution containing the magnetic particles was plated on selective PEMBA-plates, incubated for about 18 h and counted. Mean values of 2 experiments each were formed. The result of the experiment is depicted in
Experiment 17: Binding of Clostridii with JS-tag Constructs of the CBD3626
Based on the endolysin of the Clostridium perfringens phage Φ3626, Ply3626, described in the literature (Zimmer et al., 2002, Appl. Environm. Microbiol., 68, 5311-5317), different variants of a potentially suitable CBD3626 with JS-tag was prepared with molecular biological techniques. Since the expression of all constructs was very poor at the beginning, a synthetic gene was prepared optimized on the different codon uses in clostridium and E. coli. It turned out that conventional variants cloned with his-tag were only very poorly soluble; however, the variants with JS-tag were better soluble. The variants which turned out to be suitable to be expressed stable and soluble such that they could be purified subsequently and used in an functioning binding test with Clostridium perfringens cells, are summarized in table 4. The cloning was carried out in the vector Pet21 d each. The expression was carried out in the strain E. coli HMS174 at 30° C. in LB-medium with ampicillin and chloramphenicol under coexpression of birA on another plasmid.
For the binding test a preculture of Clostridium perfringens was grown at 45° C. overnight in TYG-medium (Trypton, yeast extract, glucose) in an anaerobic box. The binding tests were performed in 1 ml PBST buffer at room temperature. Clostridium cells were provided in a concentration of about 104 CFU/ml and the different JS-tag-CBD3626 constructs were added in a concentration of 25 μg/ml, mixed and incubated for about 2 min. Subsequently, PA-streptavidin beads (Microcoat) were added in a concentration of 100 μg/ml and rolling incubated for 10 min. Samples were taken from each CBD constructs containing GFP as a marker and analysed concerning binding under the fluorescence microscope. The result is depicted in
Experiment 18: Binding of Staphylococci with Specific JS-Tag-CBD-Constructs
Overnight cultures of staphylococcus strains were diluted 1:10 in BHI-medium and incubated at 30° C. up to an OD600 of about 1. The cells were diluted in medium and used in a concentration of 104 CFU/ml in the test (test approach 500 μl, PBST buffer). Specific binding protein JS5_bCBDUSA was added in a concentration of 10 μg/ml and incubated with the cells for 20 min at room temperature. Subsequently PA-strepativin beads (Microcoat) were added in a concentration of 200 μg/ml and incubated for 45 min in an overhead rolator. The magnetic particles with the bound cells were incubated 5 min in the magnetic separator, washed once with 1 ml buffer and resuspended again in 500 μl of buffer. The magnetic particles and the pooled supernatants were plated on CASO-plates and incubated overnight at 37° C. The plates with Staphylococcus aureus and S. haemolyticus could be counted after 16 h, the plates with S. epidermidis not until after 24 h. The portion of bound staphylococci was analyzed in comparison to the totally found bacteria.
Staphylococcus aureus
Staphylococcus haemolyticus
Staphylococcus epidermidis
With the JS5b-CBDUSA-construct different strains of staphylococci can be bound, wherein S. aureus binds best.
Experiment 19: Polypeptide constructs according to the present invention with CBD and JS-tag bind staphylococcus cells significantly better in the bead-sb-binding test than to a CBD with a combination of Strep-tag and His-tag (NS-HIS)
The performance of the experiment was carried out in the cell binding test analogously to experiment 18. JS-tag polypeptide constructs according to the present invention of the cell binding domain of the CBDPitti26 (SEQ ID NO: 28) were used compared to a construct, which contained a strep-tag for the biotin binding as well as a his-tag (combination NS-his). The CBD constructs have the NS-his and JS-tag (variant 5b) at the N-terminus, respectively, followed by a linker sequence (AGAGAGAGSEL) and the CBDPitti26 sequence. Pitti26 is a self-isolate of a staphylococcus phage. The polypeptide constructs were dialysed freshly prior to the test, the protein concentration was determined using UV absorption measurement and the proteins were introduced in the test in a concentration of 10 μg/ml. The binding behaviour to 3 different staphylococcus strains is depicted in table 6.
Staphylococcus
Staphylococcus
Staphylococcus
aureus
epidermidis
haemolyticus
Whereas with the JS-tag-CBD-construct according to the present invention cell binding could be detected at all 3 staphylococcus strains, even though with significant preference for Staphylococcus epidermidis cells, concerning the construct with strep-tag and his-tag just for the Staphylococcus epidermidis cells a very weak binding could be achieved under the same conditions. This shows that the constructs according to the present invention are significantly better suitable for the binding of bacteria cells than CBDs coupled to other tags.
Experiment 20: Detection of the Staphylococci Binding by Staphylococcus Specific JS-Tag-CBD Constructs in the Cell Binding Test
Analogously to the experimental design described in experiment 18, a multitude of different staphylococcus strains were tested under usage of 3 different JS-tag-CBD constructs specific for staphylococcus. The CBD portions of the different constructs derive from the phage ΦSA2usa—JS-CBDUSA, and from endolysins from the phage self-isolates PlyOpf—JS-CBDOpf—and PlyPitti20—JS-CBDPitti20, respectively. The polypeptide sequences can be found under the SEQ ID NOs 21, 31, and 27, respectively. The results of the experiments are depicted in
The polypeptide constructs according to the present invention JS-CBDUSA, JS-CBDOpf and JS-CBDPitti20 bind a multitude of staphylococcus strains specifically in the beads binding test. Thereby Staphylococcus aureus MRSA strains are bound similarly good as non-MRSA strains (
Experiment 21: Detection of the Cell Binding to JS-Tag-CBD-Constructs with the Help of the Peroxidase Test
The peroxidase test is based on a principle that a measuring signal can only be detected if the used StrepTactin-HRP (Horse Radish peroxidase)-conjugate (IBA, Göttingen) can bind via the StrepTacin portion to biotin. This biotin is covalently bound to the biotinylated JS-tag of the polypeptide constructs according to the present invention, which in turn have specifically bound to bacteria cells in the experimental approach and is therefore present in the pellet fraction, which are retained in the centrifugation steps. Non-bound JS-tag-CBD-constructs would be discarded with the respective centrifugation supernatants and would therefore not give a signal. The JS-tag-polypeptide-constructs according to the present invention were dialysed against TE-buffer (20 mM Tris, 50 mM NaCl, 5 mM EDTA; pH 7.0) overnight prior to the use in the peroxidase test and subsequently the protein concentration was determined via UV-absorption by means of the specific extinction coefficient. Microtiter plates (Deepwell) were blocked by incubation with 600 μl PBST each for 1 h at 37° C. Subsequently 200 μl each of bacteria preculture and BHI-medium (as control), respectively, were pipetted into the wells and respective amounts of protein solution (protein concentration 10 μg/ml) and buffer (as control), respectively, were added and incubated for 15 min at room temperature. Tripled determinations were performed each. After incubation the plates were centrifuged for 10 min at 4° C. in a table centrifuge (3,600 rpm), the supernatants were taken and discarded and the pellets washed again with 200 μl PBST. Subsequently the pellets are resuspended in 200 μl StrepTactin-HRP-conjugate solution (1:5000 diluted in PBST) and incubated for 30 min at room temperature. The samples are centrifuged again, washed twice with 200 μl PBST each and the thereby supernatants are discarded each. After the second washing step the pellets are taken up in 200 μl ABTS-reaction solution each and the colour reaction is measured via the absorption at 405 nm (corrected by the background absorption at 600 nm) over maximally 30 min. The ABTS reaction solution is composed as follows: 18 ml Mc-Ilvains-buffer (0.1 M citric acid, 0.2 M Na2HPO4, pH 5.0) plus 2 ml of 1% ABTS (2,2 azino-bis(3-ethyl)benzthiazolin 6-sulfon acid) in water plus 100 μl H2O2 (1% solution).
Results of the peroxidase test by use of different CBD-constructs according to the present invention are depicted in
As depicted in
Experiment 22: Detection of the Cell Binding to Enterococcus Specific JS-Tag-CBD-Constructs with the Help of the Peroxidase Test
The performance of the peroxidase test was carried out as described in experiment 21. The result of the peroxidase test by use of different CBD-constructs according to the present invention is depicted in
| Number | Date | Country | Kind |
|---|---|---|---|
| 10 2006 061 002.4 | Dec 2006 | DE | national |
| Filing Document | Filing Date | Country | Kind | 371c Date |
|---|---|---|---|---|
| PCT/DE07/02320 | 12/21/2007 | WO | 00 | 12/8/2009 |
