Identification And Synthesis Of Drug Candidates Derived From Human Microbiome Metasecretome Proteins

Abstract

The present invention relates to the treatment of diseases relating to proteins of the human microbiome metasecretome and, thus, to microbiome interactions, especially microbiome-host interactions. In particular the present invention relates to a method for identification of secreted peptides and proteins of the human microbiome. The present invention also relates to methods for generating a database of human microbiome metasecretome protein sequences. Furthermore, the present invention relates to a method for preparing a protein of the human microbiome metasecretome as well as to the use of such proteins in medicine.

Claims

1. A method for providing a human microbiota protein drug candidate, the method comprising the following steps: (i) providing a plurality of human microbiota protein sequences and/or a plurality of nucleic acid sequences encoding a plurality of human microbiota proteins; (ii) identifying in the sequences provided in step (i) one or more sequence(s) of (a) human microbiota protein drug candidate(s), wherein the sequence(s) of the human microbiota protein drug candidate(s) is/are selected according to the following criteria: (a) sequence(s) having, or coding for proteins having, a signal peptide; (a) sequence(s) having, or coding for proteins having, a length of 20 - 500 amino acids; and (a) sequence(s) comprising, or coding for (a) protein(s) comprising, at least two cysteine residues and/or a primary and/or a secondary structure element conferring a conformational rigid structure; and (iii) preparing one or more human microbiota protein(s) having, or encoded by, the sequence(s) identified in step (ii).

2-6. (canceled)

7. The method according to claim 1, wherein in step (i) a database of human microbiota protein sequences and/or a database of nucleic acid sequences encoding human microbiota proteins is provided.

8-14. (canceled)

15. The method according to claim 1, wherein the human microbiota protein has a length of 20 - 350 amino acids.

16. The method according to claim 1, wherein the human microbiota protein comprises at least two cysteine residues.

17. The method according to claim 16, wherein the cysteine residues account for more than 4% of the total amino acids of the human microbiota protein and/or wherein the human microbiota protein contains an even number of cysteine residues, e.g. forming at least one or two cysteine pairs.

18. (canceled)

19. (canceled)

20. The method according to claim 1, wherein the primary and/or a secondary structure element conferring a conformational rigid structure is selected from the group consisting of cysteine motif, disulfide bridge, leucine-rich repeat, alpha-helix, beta-sheet and coil.

21. (canceled)

22. The method according to claim 1, wherein the human microbiota protein has a length of 20 - 200 amino acids and comprises at least two cysteine residues, e.g. forming at least one cysteine pair.

23. The method according to claim 1, wherein the human microbiota protein has a length of 50 - 150 amino acids and comprises at least four cysteine residues, e.g. forming at least two cysteine pairs.

24-33. (canceled)

34. The method according to claim 1, wherein the human microbiota protein is a mimic or a secretagogue of a human host protein, e.g. selected from the group consisting of cytokines, interleukins, chemokines, growth factors, neuropeptides and peptide hormones.

35. (canceled)

36. The method according to claim 34, wherein the secretagogue induces secretion of Interleukin-10 (IL-10) by human immune cells.

37. The method according to claim 1, wherein the human microbiota protein is immunomodulatory.

38. The method according to claim 1, wherein the human microbiota protein is prepared in step (iii) by chemical synthesis.

39-42. (canceled)

43. The method according to claim 1, wherein the method comprises an additional step (iv) of determining at least one biological activity of the (obtained) protein(s), in particular relating to an interaction with the human host.

44. (canceled)

45. The method according to claim 43, wherein the structure of the human microbiota protein is determined and compared to the structure of a human host molecule, in particular a human host protein.

46. The method according to claim 43, wherein the biological activity is tested in vitro or in vivo.

47-51. (canceled)

52. The method according to claim 43, wherein cytokine release from human cells exposed to the obtained protein is determined.

53. (canceled)

54. Method for identification of a protein of the human microbiota metasecretome, the method comprising the following steps: (i) providing a plurality of human microbiota protein sequences and/or a plurality of nucleic acid sequences encoding a plurality of human microbiota proteins; (ii) identifying in the sequences provided in step (i) one or more sequence(s) of (a) protein(s) of the human microbiota metasecretome, wherein the sequence(s) of the protein of the human microbiota metasecretome is/are selected according to the following criteria: (a) sequence(s) having, or coding for proteins having, a signal peptide; (a) sequence(s) having, or coding for proteins having, a length of 20 - 500 amino acids; and (a) sequence(s) comprising, or coding for (a) protein(s) comprising, at least two cysteine residues and/or a primary and/or a secondary structure element conferring a conformational rigid structure.

55-62. (canceled)

63. A method for treating an inflammatory disease or an autoimmune disorder, or inducing or enhancing secretion of IL-10 from human cells, the method comprising administering, to a subject in need thereof, a human microbiota metasecretome protein comprising an amino acid sequence according to any one of SEQ ID NOs 1 - 10.

64. (canceled)

65. The method of claim 1, wherein step (iii) further comprises one or more pharmaceutically acceptable carriers, fillers or diluents.

66. The method of claim 1, wherein step (iii) further comprises an encapsulating compound or tablet-forming agent.

Description

BRIEF DESCRIPTION OF THE FIGURES

[0473] In the following a brief description of the appended figures will be given. The figuresare intended to illustrate the present invention in more detail. However, they are not intended to limit the subject matter of the invention in any way.

[0474] FIG. 1 shows for Example 1 an overview of per-species signal peptide prediction performed with Phobius.

[0475] FIG. 2 shows an overview over the workflow of an exemplified embodiment.

[0476] FIG. 3 shows for Example 4 the results of AlphaLISA for IL-10 secretion from human PBMCs by stimulation with 10 human microbiota protein drug candidates identified with a method according to the present invention.

EXAMPLES

[0477] In the following, particular examples illustrating various embodiments and aspects of the invention are presented. However, the present invention shall not to be limited in scope by the specific embodiments described herein. The following preparations and examples are given to enable those skilled in the art to more clearly understand and to practice the present invention. The present invention, however, is not limited in scope by the exemplified embodiments, which are intended as illustrations of single aspects of the invention only, and methods which are functionally equivalent are within the scope of the invention. Indeed, various modifications of the invention in addition to those described herein will become readily apparent to those skilled in the art from the foregoing description, accompanying figures and the examples below. All such modifications fall within the scope of the appended claims.

Example 1: Identification of Secreted Proteins of the Human Microbiome

1. Annotation of Human Microbiome Proteins and Selection of Proteins Having a Signal Peptide

[0478] A first study was performed, wherein the microbiome genes and genomes to process and select from, were selected based on a combination of isolated bacterial genomes that are part of the human gut microbiome and known for their role in immunomodulatory activities, and genes coming from metagenomic catalogues.

[0479] A first list of species was compiled, which included twelve species selected for their role in immuno-modulation and inflammation response control: [0480] Alistipes shahii [0481] Akkermansia muciniphila [0482] Bacteroides fragilis [0483] Bacteroides thetaiotaomicron [0484] Barnesiella intestinihominis [0485] Bifidobacterium breve [0486] Bifidobacterium longum [0487] Burkholderia cepacia [0488] Enterococcus hirae [0489] Fusobacterium varium [0490] Lactobacillus johnsonii [0491] Lactobacillus plantarum

[0492] The protein and gene sequences for the 12 selected bacterial species were downloaded from the Ensembl Bacteria database (http://bacteria.ensembl.org/index.html; P.J. Kersey, J.E. Allen, A. Allot, M. Barba, S. Boddu, B.J. Bolt, D. Carvalho-Silva, M. Christensen, P. Davis, C. Grabmueller, N. Kumar, Z. Liu, T. Maurel, B. Moore, M. D. McDowall, U. Maheswari, G. Naamati, V. Newman, C.K. Ong, D.M. Bolser., N. De Silva, K.L. Howe, N. Langridge, G. Maslen, D.M. Staines, A. Yates. Ensembl Genomes 2018: an integrated omics infrastructure for non-vertebrate species Nucleic Acids Research 2018 46(D1) D802-D808).

[0493] The protein and nucleotide sequence of each gene was then imported into a MySQL database (https://www.mysql.com/; Oracle Corporation, Redwood City, USA) that was created for the study.

[0494] The total number of proteins available for each species is presented in the Table 2 below:

TABLE-US-00002 Total number of protein sequences per species, retrieved from the Ensembl Bacteria database Species Count akkermansia_muciniphila 2138 alistipes_shahii 2563 bacteroides_fragilis 435909 bacteroides_thetaiotaomicron 9624 barnesiella_intestinihominis 2813 bifidobacterium_breve 52264 bifidobacterium_longum 90356 burkholderia_cepacia 54695 enterococcus_hirae 13257 fusobacterium_varium 3008 lactobacillus_johnsonii 14141 lactobacillus_plantarum 110870

[0495] Among others, the different numbers of proteins per species are due to the numbers of genomes and strains of each species in the public database.

[0496] All 791,638 available proteins across the 12 selected species were processed using the Phobius software (A combined transmembrane topology and signal peptide predictor, Stockholm Bioinformatics Centre; URL: http://phobius.sbc.su.se/; Lukas Käll, Anders Krogh and Erik L. L. Sonnhammer. A Combined Transmembrane Topology and Signal Peptide Prediction Method. Journal of Molecular Biology, 338(5):1027-1036, May 2004; Lukas Käll, Anders Krogh and Erik L. L. Sonnhammer. Advantages of combined transmembrane topology and signal peptide prediction--the Phobius web server Nucleic Acids Res., 35:W429-32, July 2007) to predict the presence of a signal peptide. Out of the 791,638 proteins 155,336 proteins were found positive. The Phobius results for each protein were imported into the MySQL database. FIG. 1 shows the number of proteins where Phobius found a signal peptide.

[0497] Next, the complete set of proteins was annotated using HMMSCAN (HmmerWeb version 2.21.0; https://www.ebi.ac.uk/Tools/hmmer/search/hmmscan; Finn, R. D., Clements, J. & Eddy, S. R. HMMER web server: interactive sequence similarity searching. Nucleic Acids Res 39, W29-37, doi:10.1093/nar/gkr367 (2011); Eddy, S. R. Accelerated Profile HMM Searches. PLoS Comput Biol 7, e1002195, doi:10.1371/journal.pcbi.1002195 (2011); R.D. Finn, J. Clements, W. Arndt, B.L. Miller, T.J. Wheeler, F. Schreiber, A. Bateman and S.R. Eddy: HMMER web server: 2015 update. Nucleic Acids Research (2015) Web Server Issue 43:W30-W38) and the PFAM database (Version 31.0; http://pfam.xfam.org/; R.D. Finn, P. Coggill, R.Y. Eberhardt, S.R. Eddy, J. Mistry, A.L. Mitchell, S.C. Potter, M. Punta, M. Qureshi, A. Sangrador-Vegas, G.A. Salazar, J. Tate, A. Bateman: The Pfam protein families database: towards a more sustainable future. Nucleic Acids Research (2016) Database Issue 44: D279-D285; E. L. Sonnhammer, S. R. Eddy, R. Durbin: Pfam: a comprehensive database of protein domain families based on seed alignments. In: Proteins. 28, 1997, S. 405-420) to facilitate a downstream filtering and selection of predicted secreted proteins. All PFAM domain results were imported into the same MySQL database along with protein sequences and Phobius predictions.

2. Removal of Redundant Proteins and Length Filter

[0498] To reduce redundancy within each species, a sequence clustering was performed on the subset of proteins found positive for the signal peptide prediction with Phobius using CD-HIT (http://cd-hit.org; http://weizhongli-lab.org/cd-hit/; Weizhong Li, Lukasz Jaroszewski & Adam Godzik. Clustering of highly homologous sequences to reduce the size of large protein databases. Bioinformatics (2001) 17:282-283; Weizhong Li & Adam Godzik. Cd-hit: a fast program for clustering and comparing large sets of protein or nucleotide sequences. Bioinformatics (2006) 22:1658-1659; Weizhong Li, Limin Fu, Beifang Niu, Sitao Wu and John Wooley. Ultrafast clustering algorithms for metagenomic of CD-HIT-OTU-MiSeqsequence analysis. Briefings in Bioinformatics, (2012) 13 (6): 656-668). An identity cut-off at 75% was applied. The number of non-redundant proteins per species, having a signal peptide, is reported in the Table 3 below:

TABLE-US-00003 Number of signal peptide positive and non-redundant protein sequences per species after clustering with CD-HIT Species Count akkermansia_muciniphila 499 alistipes_shahii 707 bacteroides_fragilis 5105 bacteroides_thetaiotaomicron 2039 barnesiella_intestinihominis 794 bifidobacterium_breve 513 bifidobacterium_longum 785 burkholderia_cepacia 3042 enterococcus_hirae 350 fusobacterium_varium 415 lactobacillus_johnsonii 287 lactobacillus_plantarum 1093

[0499] An additional filter was imposed on the length of the protein sequences, to select only those proteins having a length from 50 to 250 amino acids . The reason for this filtering is to avoid considering at this stage smaller peptides which can result from annotation biases or artifacts, and at the same time limit the maximum length of the selected proteins, to be best suited at this stage for the in vitro synthesis and the downstream laboratory tests. Furthermore, proteins larger than 250 amino acids are usually enzymes, while peptide hormones, growth factors and cytokine-like proteins are generally shorter than 250 amino acids and those types of peptides can be directly tested in vitro on several cellular receptors to assess their possible modulatory effects. The number of per-species proteins after the length filter is presented in the Table 4 below:

TABLE-US-00004 Number of non-redundant proteins having a signal peptide with Phobius analysis and a length between 50 and 250 amino acids Species Count akkermansia_muciniphila 152 alistipes_shahii 175 bacteroides_fragilis 1529 bacteroides_thetaiotaomicron 387 barnesiella_intestinihominis 183 bifidobacterium_breve 210 bifidobacterium_longum 313 burkholderia_cepacia 1185 enterococcus_hirae 157 fusobacterium_varium 160 lactobacillus_johnsonii 104 lactobacillus_plantarum 500

[0500] Accordingly, a list of 5,055 proteins was compiled.

3. Selecting Proteins Having Signal Peptides Based on a Distinct Algorithm and Removal of Incomplete/Truncated Sequences

[0501] An additional filtering step was performed to process the resulting protein sequences with SignalP v4.1 software (Center for biological sequence analysis, Technical University of Denmark DTU; URL: www.cbs.dtu.dk/services/SignalP; Henrik Nielsen, Jacob Engelbrecht, Søren Brunak and Gunnar von Heijne. Identification of prokaryotic and eukaryotic signal peptides and prediction of their cleavage sites. Protein Engineering, 10:1-6, 1997; Thomas Nordahl Petersen, Søren Brunak, Gunnar von Heijne & Henrik Nielsen. SignalP 4.0: discriminating signal peptides from transmembrane regions. Nature Methods, 8:785-786, 2011). Only the proteins showing a positive prediction of the signal peptide (either for the Gram+ or Gram- method used by SignalP) were retained.

[0502] Thereafter, corresponding nucleotide sequences were searched for proper start and stop codons to assure that the original gene was “complete”. Thereby incomplete genes (and truncated protein sequences) were removed from the list.

[0503] The final number of protein sequences at this stage, after the above-described filtering steps and the double signal peptide prediction with Phobius and SignalP methods, included 2,573 sequences of secreted human microbiome proteins.

Example 2: In Silico Cleavage of the Signal Peptide

[0504] For the downstream protein synthesis and analysis only the final “leaderless” proteins are relevant, for example to obtain correctly folded proteins and to correctly calculate the amino acids frequencies in each sequence. Therefore, in the above identified protein sequences the signal peptides were removed.

[0505] To this end, the cleavage site given by the “Y score” from SignalP was used. In cases where both the Gram+ and Gram- SignalP predictions were positive for the same sequence, the smaller cleavage site was considered.

Example 3: Cysteine Motifs Identification

[0506] In order to identify proteins having an increased rigid structure, the final list of 2,573 proteins was searched for cysteine-rich proteins, wherein the cysteines form disulfide bonds.

[0507] To this end, KAPPA (http://kappa-sequence-search.sourceforge.net; Joly V, Matton DP. KAPPA, a simple algorithm for discovery and clustering of proteins defined by a key amino acid pattern: a case study of the cysteine-rich proteins. Bioinformatics. 2015 Jun 1;31(11):1716-23), a recently published algorithm for discovery and clustering of proteins defined by a key amino acid pattern was used to process the 2,573 sequences obtained after filtering, double signal peptide prediction and the in silico cleavage of the signal peptide. With the KAPPA analysis, 25 protein clusters were identified sharing common Cysteine-motifs, the following table summarizes the results of the KAPPA run:

TABLE-US-00005 The proteins of each KAPPA cluster with the median number of cysteines per protein in each cluster Cluster ID Nb of proteins Median Nb of Cys Cluster_01 510 2.0 Cluster_02 196 3.0 Cluster_03 145 4.0 Cluster_04 68 4.0 Cluster_05 54 2.0 Cluster_06 8 3.0 Cluster_07 7 4.0 Cluster_08 7 4.0 Cluster_09 6 2.0 Cluster_10 5 2.0 Cluster_11 4 3.0 Cluster_12 4 4.0 Cluster_13 3 9.0 Cluster_14 3 8.0 Cluster_15 3 3.0 Cluster_16 3 5.0 Cluster 1 7 3 6.0 Cluster_18 3 6.0 Cluster_19 2 8.0 Cluster_20 2 4.0 Cluster_21 2 8.0 Cluster_22 2 8.0 Cluster_23 2 6.0 Cluster_24 2 7.0 Cluster_25 2 4.0 Clustering_singletons 47 7.0

[0508] An overview over the work flow of an exemplified embodiment, in particular wherein the methods described in Examples 1 - 3 were performed, is shown in FIG. 2.

[0509] The human microbiota protein drug candidates identified in Examples 1 - 3 were produced as described herein.

Example 4: Identification of Human Microbiota-Derived Protein Drug Candidates Inducing IL-10 Release from Human Cells

[0510] The aim of this study was to identify proteins expressed by human microbiota, which are able to induce secretion of IL-10 from human immune cells. To this end, a library of proteins expressed by human microbiota was constructed and screened to identify proteins inducing secretion of IL-10 from human immune cells.

Experimental Procedures

Library In Silico Method

[0511] A compound library of secreted proteins from gut commensal bacteria was generated by an in silico-based approach. The library included more than 12,000 proteins predicted from human gut microbiome catalogues and from bacterial species known for their role in immune modulation. To obtain the library, bacterial proteins having a length from 50 to 350 amino acids (length of pre-proteins including the signal peptide) were screened for the presence of secretory signal peptides using the bioinformatic tool Phobius and were annotated using HMMSCAN and the PFAM database. A cut-off at 75% was applied to reduce sequence redundancy.

[0512] In view of the relevance of small cysteine-rich proteins in immune modulation an additional selection criterion was applied to identify cysteine-rich proteins: at least two cysteines were required to be present to form a disulphide bond. To ensure correct synthesis and folding in vitro, the amino acid sequences corresponding to the signal peptide were removed.

Library Cell Free Proteins Synthesis and Quantification

[0513] The protein library was generated using an Escherichia coli Cell Free kit suitable for generation of disulphide bonds (RTS 100 E. coli Disulfide Kit; Biotechrabbit, Hennigsdorf, Germany) according to the supplier’s protocol. The Cell-Free system is based on the continuous exchange between the reaction compartment, containing components for transcription and translation, and the feeding chamber, containing amino acids and other energy components, through a semipermeable membrane.

[0514] Heterologous protein expression using the transcriptional machinery of E. coli was improved through a codon optimization algorithm (Twist Bioscience, San Francisco, USA) applied to all the selected sequences. All synthetized ORFs were subcloned into pIVEX 2.4 vector (Biotechrabbit, Hennigsdorf, Germany) specifically designed for high-yield Cell-Free expression of His-tagged proteins.

[0515] For the detection of His-tagged proteins, the 6His Check kit Gold using the HTRF® technology (Cisbio, Codolet, France) was used according to the supplier’s protocol. Proteins, previously diluted at 1:20 in 1 X PBS, were quantified in 384 well plates against a standard curve of 6xHis GFP at 0.1 .Math.g/mL (ThermoFisher, Waltham, USA) diluted in serial dilution in the lysate used for the Cell-Free synthesis (lysate was also diluted at 1:20 in 1 X PBS).

E. Coli Production of Recombinant Proteins

[0516] DNA from positive hits was subcloned in pET-28a vector carrying an N-terminal 6xHis-Tag (Twist Bioscience, San Francisco, USA) and then transformed in E. coli BL21(DE3) thermo competent cells (New England Biolabs, Ipswich, MA, USA). For the expression of recombinant proteins, pre-cultures of BL21(DE3) clones were performed in LB-medium at 30° C. under shaking (180 rpm) conditions. Cultures were made under the same conditions and the induction was started when an OD600 of 0.4 - 0.8 was reached by using 0.5 mM IPTG. Depending on the properties of each protein, induction was performed for 2 hours to overnight.

[0517] Cultures were centrifuged for 15 min at 4° C., 4500 rpm. Supernatants were removed and the pellets were frozen at -80° C. to break the cells. Pellets were then thawed and resuspended in 1 X BugBuster® (Novagen®, Merck KGaA, Darmstadt, Germany) supplemented with Benzonase® Nuclease (Sigma-Aldrich, St. Louis, USA) and Lysozyme (Sigma-Aldrich, St. Louis, USA). Samples were incubated at room temperature by gentle shaking and centrifuged at 4° C., 15,000 g for 30 minutes. Soluble proteins were purified from supernatants onto Nickel packed columns (Protino®, Macherey-Nagel, Düren, Germany) according to the supplier’s protocol. Proteins produced in inclusion bodies were solubilized from pellets using 8 M urea. Imidazole (250 mM), used for proteins elution, was removed by buffer exchange using 3 kDa cutoff filters (Amicon®, Merck Millipore Ltd., Burlington, USA). Proteins were visualized on 12% Bis-Tris acrylamide gels (ThermoFisher, Waltham, USA) stained with Coomassie Blue (Imperial protein Stain; ThermoFisher, Waltham, USA) and detected by Western Blot using the 6X-His Tag monoclonal antibody HIS.H8 (ThermoFisher, Waltham, USA) diluted at 1 :3000 and revealed using the secondary antibody Goat anti-Mouse IgG H+L WesternDot625 (ThermoFisher, Waltham, USA). Purified proteins were quantified by Bradford protein assay (Bradford M (1976) A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal. Biochem. 72:248-254).

IL10 Screening

[0518] IL-10 screening of the microbiota protein library was performed on CD14 depleted human peripheral blood mononuclear cells (PBMCs). This cellular model was chosen to reduce the background due to cell wall components and other bacterial contaminants possibly present in the lysate of the Cell Free synthesis kit on the synthesised (but not purified) proteins.

CD14-Depleted PBMCs

[0519] PBMCs were isolated from buffy coats as follows: 80 ml PBS were added to 50 ml blood; 4 SepMate™-50 IVD tubes (Stemcell Technologies, Vancouver, Canada) were filled with 15 mL of Ficoll® (Ficoll® Paque Plus; Sigma-Aldrich, St. Louis, USA) per donor, then 30 ml of PBSdiluted blood were gently added. Samples were centrifuged for 20 min at 1200 g at room temperature and washed three times with PBS. To lyse the red blood cells, pellets were resuspended in Red Blood Cells Lysis buffer 1X (Miltenyi Biotec, Bergisch Gladbach, Germany) and incubated for 10 min at room temperature. Cells were then washed with MACS buffer and counted.

[0520] PBMC depletion was performed using the CD14 Microbeads kit (Miltenyi Biotec, Bergisch Gladbach, Germany) according to the supplier’s protocol. Depleted monocytes were resuspended in Iscove’s Modified Dulbecco’s Medium (IMDM; GIBCO™, Life Technologies, Carlsbad, USA) supplemented with 1% L-Glutamine (Sigma-Aldrich, St. Louis, USA), 1% Penicillin-Streptomycin (Sigma-Aldrich, St. Louis, USA) and 10% of heat-inactivated FBS (Sigma-Aldrich, St. Louis, USA).

IL-10 Screening

[0521] Screening was performed in 384 well plates in a final volume of 60 .Math.l. PBMCs were seeded at 72,000 cells/well by multidrop (Hamilton Robotics, Martinsried, Germany) and stimulated for 72 hours with 10% (vol/vol) of the library (proteins) previously diluted at 1:10 with PBS in a humidified 5% CO.sub.2 atmosphere at 37° C. The E. coli lysate included in the cell free kit was used (at the same dilution as the library) as negative control. Phytohaemagglutinin (PHA) at 10 .Math.g/ml (0.087 .Math.M) was used as positive control. To ensure technical robustness, the screening was performed on CD14 depleted PBMCs from at least two different donors.

[0522] IL-10 secretion was measured by AlphaLISA® (IL10 (human) AlphaLISA Detection Kit; PerkinElmer, Waltham MA, USA) on the undiluted supernatants according to the supplier’s protocol. Results were expressed as AlphaLISA signal (Counts). Results were considered as positive hits, either when at least a single raw data signal (of the two signals of the two PBMC donors) was higher than the corresponding plate mean + 3SD (Standard Deviation) or when both raw data signals (of the two signals of the two PBMC donors) were higher than the corresponding plate means + 2SD (Standard Deviation). In order to avoid false positives, the concentration of the potential hits obtained by the primary screening was compared to that of the corresponding plate mean.

[0523] Potential hits were then validated by a new round of Cell-Free synthesis and test on CD14 depleted PBMCs from several donors (as described above). Further characterization was performed on recombinant proteins produced in the E. coli BL21 (DE3) strain transformed with a pET-28a vector containing the target sequence (as described above, see paragraph “ E. coli production of recombinant proteins”). Alternatively, a cell-free production of some proteins was performed by a commercial supplier (Synthelis, La Tronche, France).

Results

Screening

[0524] A total of 11904 proteins of the library was screened on CD14 depleted PBMCs in order to identify proteins able to stimulate IL-10 secretion from human PBMCs. From various potential hits obtained in the primary screening, so far ten proteins were confirmed in the second round of cell-free synthesis. These microbiota proteins, which are able to stimulate IL-10 secretion from human PBMCs, include ID3166 (SEQ ID NO: 1); ID6359 (SEQ ID NO: 2); ID1888 (SEQ ID NO: 3); ID1889 (SEQ ID NO: 4); ID2661 (SEQ ID NO: 5); ID5682 (SEQ ID NO: 6); ID5138 (SEQ ID NO: 7); ID6077 (SEQ ID NO: 8); ID6274 (SEQ ID NO: 9); and ID6298 (SEQ ID NO: 10).

[0525] Results of the AlphaLISA for IL-10 secretion from human PBMCs of the second round of cell-free synthesis are shown in FIG. 1. The data show that ten IL-10 secretagogue proteins were identified by screening of a human microbiome metasecretome protein library on stimulation of IL-10 release from human PBMCs. These microbiota proteins induce IL-10 release from human immune cells.

TABLE-US-00006 TABLE OF SEQUENCES AND SEQ ID NUMBERS (SEQUENCE LISTING): SEQ ID NO Sequence Remarks SEQ ID NO: 1 AFLFTSTGVPKKAAEAAFFLYLNKGTKKGSRSCFFIYLDRGTKKGSRSCFFYLPRQGYQKGSRGCFFIYLDRGTKKGSRGCFFIYLDCEKRAGNVCIRKCRGRYLHKKTPRRYRNAEATCS ID3166 SEQ ID NO: 2 QTRKQREDAKREAWKKERKEKKALEAQQDSVSF MKDTESCCASKAFFSLRSFFHASRLASSLCFLVCAFVTPLKQTSNNAAKNNTFFIIKAVLLINISFR ID6359 SEQ ID NO: 3 ARNYTCDVCGNGTIQIVSSHIIHNVHCGFIPCNKI NGVMDEVVYKTVTENNEACNNCGVSYTYKVYG DMEIICKAKAN ID1888 SEQ ID NO: 4 AEPADTAISERRVELCGNCGGRMVTSTTWGSWYT VAQIKCTHHNYGTDLRQQRDGTATTKCQGCGQ GYTTSKSQTRIVCHGYDS ID1889 SEQ ID NO: 5 AAFVFSNSLKPANASSAESSRLLIHVNSFFSQLGLKPISENLLRKTAHFCEFGMLGILASSACAMFSGAYSAASLPSLRRRGFFISFGVSVACAVCDETIQYFVPGRACRVTDMLIDSAGALCGLAAVLAFCAAIRVRRRRRRN ID2661 SEQ ID NO: 6 LAGPGSGCRFTPSCSTYFIQAVEIHGALKGSLMGI WRILRCNPWGGCGYDPVPPRKPR ID5682 SEQ ID NO: 7 AKLGMAAGAMLVLGLLAAGASGGTLlLAALALCA VTLLCGRKKQ ID5138 SEQ ID NO: 8 VEKKTVITKCAITVNEYREKVVPSMRKIHAIVIFVSYSINHLYKNCEPEQLFSPGRKTKKPPPATCRKRLNLQYF ID6077 SEQ ID NO: 9 EITQPCNHVKSDWIIDKEATCIGSYAFYNCTSLTSIEISTSVTKIKYRAFASCRALNNIYYTGTLTQWNEISKDTNWNWAAPLNCKVICLNGTCYL ID6274 SEQ ID NO: 10 LLVSVCTAAGLLAVAMRQIEPLLAWLRTLEVYFQG QSPAVLLRALGIALVAQFAADTCREAGLCAASTAlE LCGRVLVLLQALPLLRSLLGSFADYLQ ID6298