TOOLS AND METHODS TO DETECT AND ISOLATE COLIBACTIN PRODUCING BACTERIA

Abstract

The invention relates to the field of medical diagnosis, particularly to methods and antibodies useful for the identification of colibactin producing bacteria (pks+ bacteria). Herein are disclosed peptides and antibodies that allow for the detection and isolation of pks+ bacteria, as well as uses and methods of use of said peptides and antibodies.

Claims

1. An isolated peptide consisting of the amino acid sequence SEQ ID NO: 2.

2. An isolated polynucleotide encoding for the peptide according to claim 1.

3. An antibody that specifically recognizes the peptide according to claim 1.

4. The antibody according to claim 3 comprising a variable region comprising, preferably consisting of, the SEQ ID NO: 7.

5. The antibody according to any of claims 3 to 4 wherein said antibody is conjugated with a detectable label.

6. Pharmaceutical composition comprising the antibody according to any of claims 3 to 5 and preferably a pharmaceutical acceptable adjuvant, a pharmaceutical acceptable vehicle and/or another active principle.

7. A kit comprising the peptide according to claim 1 and/or the antibody according to any of claims 3 to 5.

8. The use of the antibody according to any of claims 3 to 5 or the kit according to claim 7 for the in vitro detection in isolated samples of pks+ bacteria, preferably of E. coli pks+.

9. The use of the antibody according to any of claims 3 to 5 or the kit according to claim 7 for the isolation from isolated samples of pks+ bacteria, preferably of E. coli pks+.

10. Method for the in vitro detection of pks+ bacteria comprising: a. providing an antibody according to any of claims 3 to 5; b. adding the antibody to a solution containing a sample of interest in sufficient conditions and with sufficient time of incubation to allow the antibody to bind to the target, creating a reaction mixture; and c. detecting the presence of the bacterial cells bound to the antibody in the reaction mixture.

11. Method for the isolation from isolated samples of pks+ bacteria comprising steps a. and b. of the method according to claim 10 and a step c. comprising separating the cells bound to antibodies from the reaction mixture.

12. The antibody according to any of claims 3 to 5 or the pharmaceutical composition according to claim 6 for use as a medicament.

13. The antibody according to any of claims 3 to 5 or the pharmaceutical composition according to claim 6 for use in the treatment or prevention of colorectal cancer.

Description

DESCRIPTION OF THE DRAWINGS

[0070] FIG. 1. Dispersion diagrams of representative flow cytometry experiments. Diagrams show the acquisition of marked Escherichia coli Nissle 1917 cell suspension in exponential phase of growth and aeration condition with four different polyclonal antibodies: (A) anti-peptide 1, (B) anti-peptide 2, (C) anti-peptide 3 and (D) anti-peptide 4.

[0071] FIG. 2. Labelling results obtained in nutrient broth at OD.sub.600 of 1.00 with antibody anti-peptide 2. Dispersion diagrams and histograms show the flow cytometry acquisition of E. coli (A) LMG2092 and (B) Nissle 1917. (C) Immunofluorescence microscopy photography shows the binding of FITC-anti-peptide 2 antibody to the Nissle strain.

[0072] FIG. 3. Detection, enrichment and depletion of E. coli pks+ from gut microbiotas. Dispersion diagrams and histograms shows the flow cytometry acquisition in a microbiota from a healthy donor (A) and a patient with Lynch syndrome (B). For each sample, the microbiota without labelling (1-top) and the labelled microbiota (1-bottom) and the negative (2-top) and positive (2-bottom) are shown.

EXAMPLES

1. Materials and Methods

[0073] Distribution and Structure of the Colibactin Genome Island

[0074] Six thousand two hundred twelve E. coli strains isolated from humans and deposited in the PATRIC database (https://www.patricbrc.org/) and the 35 reference strains from NCBI were selected for the study of the distribution and structure of the colibactin genome island (pks+ strains). For colibactin detection, proteomes from the different strains were aligned against the published sequence of the colibactin genome island of the E. coli strain IHE3034 (as set forth with the accession no. AM229678) using BLASTp. For positive detection, 20 out of 23 genes must be detected with less than 5 gaps between consecutive genes. For each gene a cut-off value was set based on the alignment expected value (E-value <1 e.sup.−5) and sequence coverage (identity >80%).

[0075] Selection of Pks+ Unique Peptides

[0076] The proteins of the colibactin genome island of the E. coli strain IHE3034 were manually digested in peptides of 20 amino acids. Later, the proteomes of the 6,212 E. coli strains from the PATRIC database were aligned against these peptides using BLASTp. For positive detection, a cut-off value was set based on the alignment expected value (E-value <1e.sup.−5) and sequence coverage (identity >80%). The 4 peptides, the peptides of the invention, from cytoplasmic membrane proteins with the biggest detection score on the pks+ strains and the lowest detection score on the pks-strains were selected. Subcellular location of the reference colibactin genome island was predicted using the PSORTb v3.0 tool (https://www.psort.org). Transmembrane prediction was performed using the HMMTOP v2.0 tool (www.enzim.hu).

[0077] Bacterial Strains and Growth Conditions

[0078] Escherichia coli LMG2092 and Nissle 1917 were grown in Luria-Bertani broth (LB) containing 10 g/L tryptone (Biokar Diagnostics, France), 5 g/L yeast extract (Biokar Diagnostics, France), 10 g/L of NaCl (Merck, KGaA, Darmstadt, Germany) and 1 litre of deionised H.sub.2O. All cultures were grown on the surface of agar plates at 37° C. in a MG500 anaerobic chamber (Don Whitley Scientific, West Yorkshire 100, UK) with an atmosphere of 10% (v/v) H.sub.2, 10% CO.sub.2, and 80% N.sub.2 for 48 h. After that, one single colony was inoculated in 4 mL of fresh liquid media and all cultures were incubated at 37° C. for 24 h in anaerobic, aerobic and aerobic with shaking conditions. The next day, 100 μL of the bacteria suspension were inoculated into 4 mL of fresh medium and incubated at 37° C. until an optical density (OD600) of about 0.6 after 3 h approximately. After that, cultures were harvested by centrifugation, washed once with bacterial flow cytometry buffer (Miltenyi, Bergisch Gladbach, Germany) and resuspended in the same buffer to an OD600=0.2, which represents around 108 CFUs/mL. The bacteria were also growth in fresh liquid media of Nutrient broth (Oxoid, Ltd., Baingstoke, Hampshire, UK) and nutrient broth supplemented with 2% of glucose (Sigma, St. Louis, Mo.).

[0079] Faecal Sample Collection and Microbiota Separation

[0080] The study sample comprised 5 faecal samples from healthy donors and 5 faecal samples from patients with Lynch syndrome. Fresh faecal material from healthy donors was collected in a sterile container and immediately manipulated and homogenised within a maximum of 2 h from defecation. Nine millilitres of sterile NaCl 0.9% (w/v) was added to 1 g of sample, and the mixture was homogenised in a sterile bag using a laboratory paddle blender (Stomacher Lab Blender 400, Seward Ltd. UK). Microbiota extraction was then performed following the protocol described by Hevia et al. (Sci Rep. 2015; 5:16807. doi:10.1038/srep16807). A solution of Nycodenz® 80% (w/v) (PROGEN Biotechnik GmbH, Heidelberg, Denmark) was prepared in ultrapure water, and sterilised at 121° C. for 15 min. A volume of 3 mL of the diluted, homogenised faecal sample was placed on top of 1 mL of the Nycodenz® solution, and centrifuged for 40 min at 4° C. (9,000 g, TST41.14 rotor, Kontron, Milan, Italy). The upper phase (soluble debris) was discarded after centrifugation, and the layer corresponding to the microbiota was collected, washed once and resuspended in 1 mL of flow cytometry (FC) buffer (1×MACSQuant Running Buffer, Miltenyi Biotec, Germany).

[0081] Polyclonal Antibody Generation

[0082] Polyclonal sera against the purified peptides of the invention from E. coli Nissle 1917 were generated in the Central Facilities of the University of Oviedo (Spain). A rabbit was immunised five times, with an interval of 15 days between immunisations, with 500 μg of peptide dissolved in 1 mL of PBS and mixed with 1 mL of Freund's Incomplete Adjuvant. The rabbit was finally sacrificed by intracardiac puncture and blood was let to coagulate at 37° C. for 4 h and subsequent overnight incubation at 4° C. Serum was separated by centrifugation (30 min, 2,000 g), and used for purifying the IgG. Firstly, ammonium sulphate was added to a final concentration of 45% (w/v), and the mix was incubated overnight at 4° C. After centrifugation (1 h, 10,000 g, 4° C.), the pellet was resuspended in 30 mL of PBS. This was extensively dialysed against PBS, and loaded in a ProteinA Sepharose 4 Fast Flow, previously equilibrated with 10 column volumes of PBS (50 mL). The column was washed with 6 column volumes of PBS, and five fractions of 5 mL were eluted with citric acid 100 mM pH 3.0. pH was corrected in each aliquot by adding 1 mL of 1M Tris-HCl pH 9.0. Fractions were mixed, centrifuged in a Vivaspin 20 device (3,000×g, molecular weight cut-off of 10 kDa) and washed with 20 mL of PBS. Protein concentration was estimated by measuring the A280 and samples were aliquoted and stored at −80° C.

[0083] Antibody Conjugation

[0084] Polyclonal antibodies, anti-peptide according to SEQ ID NO: 1, anti-peptide according to SEQ ID NO: 2, anti-peptide according to SEQ ID NO: 3 and anti-peptide according to SEQ ID NO: 4, serum IgG fractions were conjugated with fluorescein isothiocyanate (FITC) and allophycocyanin (APC) using the commercial kits (Abcam Cambridge, Mass., USA) and following manufacturer's instructions. For FITC and APC conjugation, the anti-peptide polyclonal antibodies were reconstituted in amine-free phosphate buffered saline (PBS). For each antibody, one hundred millilitres of 1.5 mg/mL antibody were added to the reactive dye for each conjugation. Before adding the antibody to the FITC/APC mix, 10 μL of FITC-Modifier or APC-Modifier reagent was added to the antibody. The antibody-dye mixtures were incubated in the dark at room temperature (20-25° C.) for 3 h. After incubation, 10 μL of FITC-Quencher or APC-Quencher reagent was added and mixed gently. The concentration of antibody in the final sample was 20 μg/mL.

[0085] Flow Cytometry Analysis

[0086] Labelled cells with polyclonal antibodies were acquired and analysed in a MACS Quant Flow Cytometer device (Miltenyi Biotec, Germany) using the following acquisition parameters: flow rate set to “low”, uptake volume of 10 μL, FSC set to hyperlogarithmic amplification (370 V), SSC set to hyperlogarithmic amplification (400 V), channel B1 corresponding to the FITC detection set to hyperlogarithmic amplification (370 V) and channel R1 corresponding to the APC detection set to hyperlogarithmic amplification (360 V). At least 10,000 events were acquired in each run.

[0087] Permeabilization of Escherichia coli Nissle 1917 and LMG2092

[0088] One mL of culture in exponential phase of growth was centrifuged at 10,000×g for 5 min. After that, the turbidity of the bacterial suspension was adjusted to an OD600=0.2 (around 1E10.sup.8 CFUs/mL) using FC buffer. Both E. coli were fixed in 4% cold and freshly prepared paraformaldehyde in Phosphate Buffered Saline (PBS). Samples were then incubated at 4° C. for 10 min. Bacteria were washed with FC buffer in order to remove extra-fixative reagents. For permeabilization, Tween 20 at concentration of 2% was added to each tube and incubated for 10 min at room temperature. The samples were washed with FC buffer and then, bacteria were stained with antibodies.

[0089] Detection of Escherichia coli Nissle 1917 and LMG2092 by Flow Cytometry The binding capability of the four anti-peptides antibodies was tested in a pks+ (E. coli Nissle 1917) and a pks− (E. coli LMG2092) strains. Twenty-five μL of the bacterial suspension were mixed with 25 μL of the FITC-conjugated antibody at a final concentration of 20 μg/mL. The samples were incubated for 15 min at RT and then were washed with FC buffer at 13,000 rpm for 5 minutes. Finally, bacteria were resuspended in 1504 of FC buffer and samples were acquired using a MACS Quant Flow Cytometry (Miltenyi Biotec, Germany). Bacteria were labelled and analysed at different optical density (i.e. OD600=0.3, 0.6, 1 and 1.5) and with different medium (i.e. Luria Bertani, Nutrient Broth and Nutrient Broth supplemented with 2% of glucose).

[0090] Detection of Pks+ Strains in a Gut Microbiota

[0091] The detection of pks+ strains was investigated over 5 samples of gut microbiotas from healthy donors and 5 samples from Lynch syndrome patients. Microbiotas were labelled with the antibody of the invention conjugated to APC and incubated for 15 minutes at room temperature. Then, they were washed with FC buffer at 13,000 rpm for 5 minutes and the supernatant was removed. Finally, microbiotas were resuspended in 150 μL of FC buffer and data were acquired and analysed using a MACS Quant Flow cytometer as indicated previously.

[0092] Enrichment and Depletion of Pks+ Strains from a Gut Microbiota

[0093] Microbiotas were labelled with the polyclonal anti-peptide 2 serum IgG fraction conjugated to APC and incubated for 15 minutes at room temperature. Then, they were washed with FC buffer at 13,000 rpm for 5 minutes and the supernatant was removed. For positive selection of E. coli pks+, microbiotas were resuspended in 100 μL of resuspension buffer (PBS supplemented with 2 mM EDTA and 3% (v/v) of de-complemented bovine foetal serum). To this mix, 10 μL of magnetic anti-APC particles were added (BD Biosciences, San José, USA). The mixture was incubated for 15 minutes at 4° C. Then, microbiotas were washed at 13000 rpm for 5 minutes with 1×BD IMag buffer (BD Biosciences, San José, USA) and the column was conditioned with 1×BD IMag buffer (BD Biosciences, San José, USA). Positive fraction was retained in the magnetic column after three washes with 1×BD IMag buffer (BD Biosciences). Finally, the positive fraction was eluted through the 1×BD IMag buffer. Both positive and negative fractions were further analysed by flow cytometry using a MACS Quant Flow cytometer as indicated previously.

[0094] Immunofluorescence Microscopy

[0095] The binding ability of the antibody anti-peptide according to SEQ ID NO: 1, the antibody anti-peptide according to SEQ ID NO: 2 and the antibody anti-peptide according to SEQ ID NO: 3 contained in each of the polyclonal serums was also determined by confocal scanning laser microscope equipped with a Leica DFC365FX digital camera (DMi8, Leica Microsystems). The bacterial species were grown until exponential phase OD600 of 0.7, then were washed with flow cytometry buffer, fixated with paraformaldehyde 4%, permeabilised with flow cytometry buffer supplemented with 2% of Tween 20 and incubated for 15 min with the polyclonal antibody anti-peptides conjugated with FITC (200 μg/mL). Finally, bacteria were washed once with flow cytometry buffer, resuspended in 10 μL of flow cytometer buffer and were analysed with confocal microscopy using a 100× oil objective. The images were acquired with software LasX (Leica Microsystems). The FITC filter cube (excitation 480/40, emission 527/30) was used.

[0096] Validation of the Positive Fraction by PCR Analysis

[0097] Strains retrieved in the positive fraction of the magnetic separation were screened for the presence of pks islands by PCR using primers for the clbB gene, predicted before as present in all the pks+ strains. Two different primers for amplify different regions of the clbB gene were selected, i.e. clbB1 and clbB2. Primers used to amplify these regions were clbB1r consisting of SEQ ID NO: 8 ((5′-CCATTTCCCGTTTGAGCACAC-3′), clbB1f consisting of SEQ ID NO: 9 (5′-GATTTGGATACTGGCGATAACCG-3′), clbB2r consisting of SEQ ID NO: 10 (5′-GCGCTCTATGCTCATCAACC-3′) and clbB2f consisting of SEQ ID NO: 11 (5′-GCGCATCCTCAAGAGTAAATA-3′). For primer ClbB1, the PCR temperature cycling conditions were as follows: initial denaturation for 10 min at 95° C., followed by 30 standard cycles: denaturation at 95° C. for 45 secs, primer annealing for 1 min at 54° C., and primer extension at 72° C. for 1 min, and a final extension step at 72° C. for 5 min. For primer ClbB2, the PCR temperature cycling conditions were as follows: initial denaturation for 7 min at 95° C., followed by 35 standard cycles: denaturation at 95° C. for 30 secs, primer annealing for 30 secs at 55° C., and primer extension at 72° C. for 30 secs, and a final extension step at 72° C. for 7 min.

2. Results

[0098] Distribution of the Pks Island Genes

[0099] Not all of the pks+E. coli strains presented all the genes related to the reference pks island described for strain IHE3034. Table 1 shows the distribution of the pks islands in the strains where it was detected. Most of the genes, i.e. clbP, clbO, clbN, clbM, clbL, clbI, clbH, clbF, clbE, clbD, clbC, clbB, were detected in all the island distributions. However, the intP4, clbS, clbQ, clbK, clbJ, trpA and trpB genes were detected with a percentage of distribution between 97 and 99%, while for the clbR, clbA and trpC genes their percentage were lower, between 70 and 78%, with clbR being the gene detected in only 70% of the strains.

TABLE-US-00001 TABLE 1 Distribution of the colibactin island genes along the pks+ strains. Fourteen strains containing errors were discarded. Gene Gene # name Putative function Detection percentage 1 intP4 P4-like integrase 1,198 of 1,212 (98.84%) 2 clbS Colibactin resistance protein 1,186 of 1,212 (97.85%) 3 clbQ Thioesterase 1,211 of 1,212 (99.92%) 4 cblP FmtA-like protein 1,212 of 1,212 (100.0%) 5 clbO PKS 1,212 of 1,212 (100.0%) 6 clbN NRPS 1,212 of 1,212 (100.0%) 7 clbM MatE-like protein 1,212 of 1,212 (100.0%) 8 clbL Amidase 1,212 of 1,212 (100.0%) 9 clbK PKS/NRPS 1,094 of 1,212 (90.26%) 10 clbJ NRPS 1,202 of 1,212 (99.17%) 11 clbI PKS 1,212 of 1,212 (100.0%) 12 clbH NRPS 1,212 of 1,212 (100.0%) 13 clbG Malonyl-CoA transacylase 1,212 of 1,212 (100.0%) 14 clbF Acyl-CoA-dehydrogenase 1,226 of 1,226 (100.0%) 15 clbE Acyl/D-alanyl carrier protein 1,212 of 1,212 (100.0%) 16 clbD 3-hydroxyacyl-CoA- 1,212 of 1,212 (100.0%) dehydrogenase 17 clbC PKS 1,212 of 1,212 (100.0%) 18 clbB NRPS/PKS 1,212 of 1,212 (100.0%) 19 clbR LuxR-like   856 of 1,212 (70.63%) 20 clbA Phosphopantetheinyl   968 of 1,212 (79.87%) transferase 21 trpA IS 1400 transposase ORFA 1,193 of 1,212 (98.43%) 22 trpB IS 1400 transposase ORFB 1,209 of 1,212 (99.75%) 23 trpC Transposase fragment   741 of 1,212 (61.13%)

[0100] Selection of Pks+ Unique Peptides

[0101] Four peptides, the peptides of the invention, were selected based on their exclusive belonging to the pks+ strains for antibody generation (Table 2). These 20 amino acid length peptides were predicted on 3 cytoplasmic membrane proteins, i.e. clbH (NRPS), clbC (PKS) and clbD (NRPS/PKS).

TABLE-US-00002 TABLE 2 Selected peptides for the antibody generation. All the peptides selected were predicted in the cytoplasmatic membrane. Pks Presence on Presence on Antibody Peptide sequence gene pks− strains pks+ strains 1 LDDHGNPVADGEEGELYLAG clbH 2 of 4986 1,225 of 1,226 (SEQ ID NO: 1) 2 WRDGESQQIHYIGRNDFQIK clbH 2 of 4986 1,225 of 1,226 (SEQ ID NO: 2) 3 AAGDRIYAVIRGSAVNNDGK clbC 2 of 4986 1,224 of 1226 (SEQ ID NO: 3) 4 FNTLVVLENYPVDMTLLSCA clbB 2 of 4986 1,225 of 1226 (SEQ ID NO: 4)

[0102] Labelling Strains with Different Antibodies

[0103] Escherichia coli Nissle 1917 and LMG2092 were growth in LB at 37° C., in aeration, until reached optical density OD.sub.600 of 0.6. Then, bacteria were labelled with four different polyclonal antibodies, targeting each one of the peptides of the invention according to SEQ ID NO: 1-4. The labelling of pks+ (Nissle 1917) and pks− (LMG2092) strains with the antibodies anti-peptide according to SEQ ID NO: 1, anti-peptide according to SEQ ID NO: 2, anti-peptide according to SEQ ID NO: 3 or anti-peptide according to SEQ ID NO: 4, were evaluated by flow cytometry. FIG. 1 represents the specificity of the antibody anti-peptide according to SEQ ID NO: 1, the antibody anti-peptide according to SEQ ID NO: 2, the antibody anti-peptide according to SEQ ID NO: 3 or the antibody anti-peptide according to SEQ ID NO: 4 in the detection of E. coli Nissle, with 50% approximately of bacteria labelled after 15 minutes of incubation with the anti-peptide according to SEQ ID NO: 1, anti-peptide according to SEQ ID NO: 2 or anti-peptide according to SEQ ID NO: 3 antibodies. The labelling of bacteria with anti-peptide according to SEQ ID NO:4 antibody was less performant, so for the following experiments it was discarded.

[0104] Labelling Strains at Different Conditions

[0105] In order to find the best labelling conditions, Escherichia coli Nissle 1917 and LMG2092 were growth in different medium (i.e. LB, nutrient broth and nutrient broth supplemented with 2% of glucose), at different optical density (i.e. 0.3, 0.6, 1 and 1.5) and labelled with antibodies of the invention (Table 3). FIG. 2 shows the best labelling conditions, that was obtained in labelling the cells with the FITC labelled antibody of the invention in nutrient broth at OD.sub.600 of 1.00. The higher emission of E. coli Nissle 1917 (FIG. 1B) and the lower emission of the E. coli LMG2092 (FIG. 1A) can be deduced comparing the dispersion plots.

TABLE-US-00003 TABLE 3 Summary data on the percentage of bacteria labelled with different medium, antibodies and optical density. The data show the mean and the standard deviation. Antibody anti- Antibody anti- Antibody anti- peptide peptide peptide SEQ ID SEQ ID SEQ ID Medium OD.sub.600 NO: 1 NO: 2 NO: 3 Luria 0.3 33.58 ± 2.30 44.73 ± 3.03 46.55 ± 1.45 Bertani 0.6 43.18 ± 2.55 48.42 ± 0.90 51.60 ± 0.94 1 44.13 ± 0.79 50.42 ± 5.84 53.93 ± 2.41 1.5 37.72± 2.18 35.64 ± 0.60 39.64 ± 2.90 Nutrient 0.3 48.56 ± 0.74 45.36 ± 3.61 42.99 ± 6.78 broth 0.6 40.74 ± 4.63 61.37 ± 5.88 43.34 ± 1.78 1 40.26 ± 1.24 63.16 ± 8.82 38.80 ± 4.40 1.5 50.22 ± 7.27 43.04 ± 0.98 46.20 ± 4.91 Nutrient 0.3 45.79 ± 2.26 46.74 ± 3.11 36.70 ± 5.55 broth + 0.6   46 ± 0.23 47.22 ± 1.31 53.45 ± 0.88 glucose 1  49.72 ± 10.16  53.3 ± 7.77 58.74 ± 6.77 1.5 49.52 ± 5.23 62.19 ± 5.97 51.43 ± 5.35

[0106] Confocal Microscopy of the Labelled Strains

[0107] As the antibody of the invention labelled bound to Escherichia coli Nissle 1917 cells in a greater proportion than to LMG2092, some immune-staining pictures were performed. The staining of the E. coli Nissle 1917 with the antibody of the invention conjugated with FITC was analysed using the confocal laser scanning microscope (FIG. 2C).

[0108] Detection, Enrichment and Depletion of Pks+ Strains in Gut Microbiotas

[0109] Ten samples of real gut microbiotas (5 from healthy donors and 5 from Lynch syndrome patients) were analysed to detect, enrich and deplete from pks+ strains.

[0110] Microbiotas were labelled with the antibody of the invention conjugated with APC and further analysed by flow cytometry for pks+ strains detection (FIG. 3). From the labelled microbiotas, negative and positive fractions were obtained by magnetic separation. The results show that it is possible to deplete the microbiotas from pks+ strains in single step using the antibody of the invention. The positive fractions were also cultivated to discover potential false positives in the cells obtained with the magnetic separations. Table 4 shows the identified isolates by 16S sequencing and the pks+ validation by PCR analysis of genes of the colibactin cluster.

TABLE-US-00004 TABLE 4 Isolated strains from the positive fraction of the gut microbiotas. PCR PCR Identity validation validation Isolated 16S percentage (ClbB1) (ClbB2) 6 Escherichia coli 99.37% Positive Negative 10 Klebsiella pneumoniae 98.00% Negative Positive 20 Escherichia coli 98.74% Negative Negative 26 Escherichia coli 98.66% Negative Positive 44 Escherichia coli   100% Positive Positive 45 Enterococcus faecium 99.89% Negative Negative 65 Escherichia coli 93.33% Negative Negative 71 Citrobacter freundii   100% Negative Positive 79 Escherichia coli 99.79% Negative Negative 81 Enterococcus faecium 99.68% Positive Positive 91 Enterobacter cloacae   100% Negative Negative 95 Citrobacter gillenii 99.58% Positive Positive 108 Escherichia coli 99.89% Negative Negative 114 Escherichia coli   100% Positive Positive