REPURPOSING COMPOUNDS FOR THE TREATMENT OF INFECTIONS AND FOR MODULATING THE COMPOSITION OF THE GUT MICROBIOME

Abstract

The present invention relates to agents and compositions for the modification of the growth of bacterial cells. Thus, the compounds of the present invention are useful for the prevention and/or treatment of a disease in a subject. In particular, the present invention relates to the field of repurposing pharmaceutical compounds for treatment strategies of infectious diseases, gastrointestinal disorders, inflammatory diseases, proliferative diseases, metabolic disorders, cardiovascular diseases, and immunological diseases. Some of the compounds of the present invention demonstrate high specificity in inhibiting the growth of single bacterial species. Such compounds enable narrow-spectrum antibacterial therapies, constituting a major effort of current and future drug development strategies in order to reduce side effects of antibacterial treatment plans. Particularly interesting compounds of this invention are effective against pathobiological species such as Clostridium difficile, Clostridium perfingens, Fusobacterium nucleatum, and an enterotoxigenic strain of Bacteroides fragilis. Other compounds of the present invention reveal a strong inhibitory effect on a broad spectrum of bacterial species. Such compounds are useful for broad-spectrum antibiotic therapies of infections with unknown causative infecting bacterial species. Both types of compounds, especially the ones with narrow-spectrum antibacterialactivity, can further be used for modulating the microbiome composition and targeting species associated with dysbiosis and disease.

Claims

1. A method for modifying growth of bacterial cells, wherein said method comprises contacting bacterial cells with a compound selected from a Ca-channel inhibitor, Tribenoside, Telmisartan, Azathioprine, Mercaptopurine, Mifepristone, Montelukast, Fentiazac, Meclozine dihydrochloride, Carbenoxolone, Gliquidone, Alfacalcidol, Acarbose, Ethacrynic acid, Chlorpromazine hydrochloride, Cyclosporin A, Flufenamic acid, Aripiprazole, Idebenone, Thioguanosine, Thyroxine (L), Gemcitabine, Folic acid, Etretinate, Paclitaxel, Phenindione, Mometasone furoate, Azacytidine-5, Luteolin, Metixene hydrochloride, Protriptyline hydrochloride, Toltrazuril, Tolnaftate, Meclofenamic acid, Prenylamine lactate, Diacerein, Dicumarol, Clemizole hydrochloride, Loratadine, Troglitazone, Tiratricol, Bepridil hydrochloride, Estradiol Valerate, Anthralin, Aprepitant, Amiodarone hydrochloride, Ethopropazine hydrochloride, Astemizole, Methiothepin maleate, pharmaceutically acceptable salts thereof, and derivatives thereof.

2. The method according to claim 1, wherein said bacterial cells are selected from cells of Gram-positive bacteria, Gram-negative bacteria, Enterobacter, Staphylococcus, Streptococcus, Pseudomonas, Escherichia, Salmonella, Helicobacter, Neisseria, Campylobacter, Chlamydia, Clostridia, Citrobacter, Vibrio, Treponema, Mycobacterium, Klebsiella, Actinomyces, Bacteroides, Bordetella, Borrelia, Brucella, Corynebacteria, Diplococcus, Fusobacterium, Leptospira, Listeria, Pasteurella, Proteus, Rickettsia, Shigella, Yersinia, Parabacteroides, Odoribacter, Faecalibacteria, Collinsella, Eggerthella, Lactonifactor, Pediococcus, Leuconostoc, Lactococcus, Roseburia, Coliform, Bacillus, Franscicella, Acinetobacter, Legionella, Actinobacillus, Coxiella, Kingella kingae, Haemophilus, Bifidobacteria, Mobiluncus, Prevotella, Akkermansia, Bilophila, Blautia, Coprococcus, Dorea, Eubacteria, Lactobacillus, Ruminococcus, Veillonella, Enterococcus, and combinations thereof, optionally wherein said bacteria are antibiotic-resistant bacteria and/or multi-drug resistant bacteria.

3. The method according to claim 1, wherein a therapeutically effective amount of said compound is administered to a subject, thereby modifying the growth of bacterial cells in said subject, wherein said modifying results in the prevention and/or treatment of a disease in said subject and/or in the modification of the composition of the microbiome of said subject.

4. The method according to claim 3, wherein said disease is selected from an infectious disease, a gastrointestinal disorder, an inflammatory disease, a proliferative disease, a metabolic disorder, a cardiovascular disease, and an immunological disease.

5. The method according to claim 1, wherein said compound is in the form of a tablet, coated tablet, effervescent tablet, capsule, powder, granulate, sugar-coated tablet, lozenge, pill, ampoule, drop, suppository, emulsion, ointment, gel, tincture, paste, cream, moist compress, gargling solution, plant juice, nasal agent, inhalation mixture, aerosol, mouthwash, mouth spray, nose spray or room spray.

6. The method according to claim 1, wherein said compound modifies the growth of bacterial cells of a spectrum of bacterial species selected from Gram-positive bacteria, Gram-negative bacteria, Enterobacter, Staphylococcus, Streptococcus, Pseudomonas, Escherichia, Salmonella, Helicobacter, Neisseria, Campylobacter, Chlamydia, Clostridia, Citrobacter, Vibrio, Treponema, Mycobacterium, Klebsiella, Actinomyces, Bacteroides, Bordetella, Borrelia, Brucella, Corynebacteria, Diplococcus, Fusobacterium, Leptospira, Listeria, Pasteurella, Proteus, Rickettsia, Shigella, Yersinia, Parabacteroides, Odoribacter, Faecalibacteria, Collinsella, Eggerthella, Lactonifactor, Pediococcus, Leuconostoc, Lactococcus, Roseburia, Conform, Bacillus, Franscicella, Acinetobacter, Legionella, Actinobacillus, Coxiella, Kingella kingae, Haemophilus, Bifidobacteria, Mobiluncus, Prevotella, Akkermansia, Bilophila, Blautia, Coprococcus, Dorea, Eubacteria, Lactobacillus, Ruminococcus, Veillonella, and Enterococcus, wherein said spectrum consists of less than 20 bacterial species.

7. The method according to claim 6, wherein said bacteria are Clostridia selected from the group consisting of C. difficile, C. butyricum, C. perfringens, C. novyi, C. septicum, C. botulinum, C. tetani, C. haemolyticum, C. carnis, C. histolyticum, C. sordellii, C. septicum, C. tertium, C. sporogenes, C. ramosum, C. inocuum, C. paraputrificum, C. cadaveris, C. bifermentans, C. fallax, and C. clostridioforme, and combinations thereof.

8. The method according to claim 6, wherein said compound is a Ca-channel inhibitor.

9. The method according to claim 6, wherein said bacterial species is C. perfringens, and said compound is selected from Alfacalcidol, Acarbose, Ethacrynic acid, Chlorpromazine hydrochloride, Cyclosporin A, Idebenone, Thioguanosine, Gemcitabine, Etretinate, Paclitaxel, Phenindione, Azacytidine-5, pharmaceutically acceptable salts thereof, and derivatives thereof

10. The method according to claim 6, wherein said bacterial species is Fusobacterium nucleatum, and wherein said compound is selected from Luteolin, pharmaceutically acceptable salts thereof, and derivatives thereof.

11. The method according to claim 6, wherein said bacterial species is an enterotoxigenic strain of Bacteroides fragilis, and wherein said compound is selected from Metixene hydrochloride, Protriptyline hydrochloride, Toltrazuril, Acarbose, Ethacrynic acid, Tolnaftate, Cilnidipine, Meclofenamic acid, Prenylamine lactate, pharmaceutically acceptable salts thereof, and derivatives thereof

12. (canceled)

13. The method according to claim 1, wherein said compound enhances the growth of bacterial cells of bacterial species selected from Lactobacillus, Bifidobacterium, Enterococcus, Streptococcus, Pediococcus, Leuconostoc, Bacillus, Eschericha, Lactococcus, and combinations thereof.

14. A pharmaceutical composition for use in the prevention and/or treatment of a disease in a subject and/or in the modification of the composition of the microbiome of a subject, comprising i) a compound selected from a Ca-channel inhibitor, Tribenoside, Telmisartan, Azathioprine, Mercaptopurine, Mifepristone, Montelukast, Fentiazac, Meclozine dihydrochloride, Carbenoxolone, Gliquidone, Alfacalcidol, Acarbose, Ethacrynic acid. Chlorpromazine hydrochloride. Cyclosporin A, Flufenamic acid, Aripiprazole, Idebenone, Thioguanosine, Thyroxine (L), Gemcitabine, Folic acid, Etretinate, Paclitaxel. Phenindione, Mometasone furoate, Azacvtidine-5, Luteolin. Metixene hydrochloride, Protriptyline hydrochloride, Toltrazuril, Tolnaftate. Meclofenamic acid, Prenylamine lactate, Diacerein, Dicumarol, Clemizole hydrochloride, Loratadine, Troglitazone, Tiratricol, Bepridil hydrochloride, Estradiol Valerate, Anthralin, Aprepitant, Amiodarone hydrochloride. Ethopropazine hydrochloride, Astemizole. Methiothepin maleate. pharmaceutically acceptable salts thereof, and derivatives thereof; and ii) a pharmaceutically acceptable additive, carrier, diluent, solvent, filter, lubricant, excipient, binder, and/or stabilizer.

15. (canceled)

16. The method according to claim 3, wherein the said subject is a mouse, rat, guinea pig, rabbit, cat, dog, monkey, or human.

17. The method according to claim 4, wherein the disease is selected from an infection of the gastrointestinal tract, an infection of the urogenital tract, an infection of the upper lower respiratory tract, an infection of the lower respiratory tract, rhinitis, tonsillitis, pharyngitis, dysbiosis, bronchitis, pneumonia, an infection of the inner organs, nephritis, hepatitis, peritonitis, endocarditis, meningitis, osteomyelitis, an infection of the eyes, an infection of the ears, a cutaneous infection, a subcutaneous infection, an infection after burn, diarrhea, colitis, pseudomembranous colitis, a skin disorder, toxic shock syndrome, bacteremia, sepsis, pelvic inflammatory disease, vaginosis, an infection of the central nervous system, wound infection, intra-abdominal infection, intravascular infection, bone infection, joint infection, acute bacterial otitis media, pyelonephritis, deep-seated abscess, tuberculosis, a gastrointestinal motility disorder, irritable bowel syndrome, constipation, a functional gastrointestinal disorder, gastroesophageal reflux disease, functional heartburn, dysbiosis, dyspepsia, functional dyspepsia, nonulcer dyspepsia, gastroparesis, chronic intestinal pseudo-obstruction, colonic pseudo-obstruction, Crohn's disease, colitis, ulcerative colitis, inflammatory bowel disease, diverticulitis, gluten intolerance and/or lactose intolerance, obesity, stomach rumble, small intestinal bacterial overgrowth (SIBO), small intestinal fungal overgrowth (SIFO), meteorism, flatulence, Crohn's disease, inflammatory bowel disease, ulcerative colitis, collagenous-, lymphocytic-, ischemic-, diversion- and/or indeterminate colitis, periodontal disease, chronic fatigue syndrome, myalgic encephalomyelitis, Behet's disease, atherosclerosis, rheumatoid arthritis, gastric cancer and colorectal cancer.

18. The method according to claim 3, wherein wherein said compound is administered to said subject by oral, intranasal, topical, rectal, bronchial, vaginal, or parenteral administration.

19. The method according to claim 6, wherein said compound is selected from a Ca-channel inhibitor, Tribenoside, Telmisartan, Azathioprine, Mercaptopurine, Mi fepristone, Montelukast, Fentiazac, Meclozine dihydrochloride, Carbenoxolone, Gliquidone, Alfacalcidol, Acarbose, Ethacrynic acid, Chlorpromazine hydrochloride, Cyclosporin A, Flufenamic acid, Aripiprazole, Idebenone, Thioguanosine, Thyroxine (L), Gemcitabine, Folic acid, Etretinate, Paclitaxel, Phenindione, Mometasone furoate, Azacytidine-5, Luteolin, Metixene hydrochloride, Protriptyline hydrochloride, Toltrazuril, Tolnaftate, Meclofenamic acid, Prenylamine lactate, pharmaceutically acceptable salts thereof, and derivatives thereof.

20. The method according to claim 8, wherein said Ca-channel inhibitor is selected from a dihyropyridine, Fendiline hydrochloride, pharmaceutically acceptable salts thereof, and derivatives thereof.

21. The method according to claim 8, wherein said dihyropyridine or said derivative thereof is selected from Lacidipine, Cilnidipine, Amlodipine, and pharmaceutically acceptable salts thereof.

Description

EXAMPLES

[0163] Bacterial Strains and Growth Conditions

[0164] Bacterial isolates used in this study were purchased from DSMZ, BEI Resources, ATCC and Dupont Health & Nutrition, or were gifts from the Denamur Lab (INSERM). All strains were recovered in their recommended rich media (resource and literature). The screen and validation experiments were performed in modified Gifu Anaerobic Medium broth (mGAM) (HyServe GmbH & Co.KG, Germany, produced by Nissui Pharmaceuticals), since almost all species could grow robustly in this medium in a manner that is reflective of their gut abundance. Only one strain was grown in Todd-Hewitt Broth (Sigma-Aldrich), one in a 1:1 mixture of mGAM and Gut Microbiota Medium and for one strain, mGAM was supplemented with 60 mM sodium formate and 10 mM taurine. All media were pre-reduced at least 1 day before use under anoxic conditions in an anaerobic chamber (Coy Laboratory Products Inc) (2% H2, 12% CO2, rest N2) and all experiments were performed under anaerobic conditions at 37 C. unless specified otherwise.

[0165] To select a representative core of species in the human gut microbiome, the inventors analyzed 364 fecal metagenomes of asymptomatic individuals from 3 continents. Species were defined and their abundance quantified as previously described. A core set of 60 microbiome species was defined, and from this core, 31 species were selected for the screen of this invention. 7 additional species were selected which are of great interest.

[0166] Preparation of Screening Plates The Prestwick Chemical Library was purchased from Prestwick Chemical Inc. (Illkirch, France) with compounds coming dissolved in dimethyl sulfoxide (DMSO) at a concentration of 10 mM. Compounds were re-arrayed to redistribute the DMSO control wells in each plate and to minimize the total number of 96- and 384-well plates (4384-well plates or 1496-well plates). At the same time, drugs were diluted to a concentration of 2 mM to facilitate further aliquoting, and these plates were stored at 30 C. For each experimental batch (10 replicates in 96-well plates; 20 replicates in 384-well plates) the inventors prepared drug plates in the respective growth medium (2 for 96-well plates, 1 for 384-well plates), and stored at 30 C. until use (max 2 months). Before inoculation, plates were thawed and pre-reduced in the anaerobic chamber overnight. The Biomek FXP (Beckman Coulter) liquid handling system was used for all rearranging and aliquoting of the library compounds.

[0167] Inoculation

[0168] Strains were grown twice overnight to make sure the inventors had a robustly and uniformly growing culture before inoculating the screening plates. For 96-well plates, the second overnight culture was diluted to fresh medium in order to reach a 2 of the aimed starting optical density (OD) at 578 nm. Next, 50 L of this diluted inoculum was added to wells containing already 50 l of 2 concentrated drug in the respective culture medium using a multichannel pipettor. Final drug concentration was 20 M and each well contained 1% DMSO. For 384-well plates, the inventors inoculated with a 384 floating pin replicator VP384FP6S (V&P Scientific, Inc.), transferring 1 l of appropriately diluted overnight culture to wells containing 50 l of growth media, 1% DMSO and 20 M drug. For bacterial species that reached lower OD in overnight cultures the inventors transferred twice 1 l of appropriately adjusted OD culture. Both for 96- and 384-well plates, the starting OD was 0.01 or 0.05 depending on the growth preference of the species.

[0169] Screening conditions for the screen of the Prestwick Chemical Library

[0170] After inoculation, plates were sealed with breathable membranes (Breathe-Easy) to prevent evaporation and cross-contamination between wells, and incubated at 37 C. without shaking. Growth curves were acquired by tracking OD at 578nm with a microplate spectrophotometer (EON, Biotek). Measurements were taken every 1-3 hrs after 30-60 seconds of linear shaking, initially manually but later automatically using a microplate stacker (Biostack 4, Biotek), fitted inside a custom-made incubator (EMBL Mechanical Workshop). The inventors collected measurements for 16-24 hrs. Each strain was screened in at least three biological replicates.

[0171] Normalization of Growth Curves and Quantification of Growth

[0172] Growth curves were analyzed by plate. All growth curves within a plate were truncated at the time of transition from exponential to stationary. The end of exponential phase was determined automatically by finding the peak OD (using the median across all compounds and control wells, and accounting for a small increase during stationary phase) and verified by inspection. Using this timepoint allowed the inventors to capture effects of drugs on lag phase, growth rate and stationary phase plateau. Timepoints with sudden spikes in OD (e.g. caused by condensation) were removed, and a growth curve was discarded completely if there were too many missing timepoints. Similarly, growth curves were discarded if the OD fell too far outside the normal range (e.g. caused by compounds that are strongly absorbing). Three compounds had to be completely excluded from the analysis, as they mostly caused aberrant growth curves: Chicago sky blue 6B, mitoxantrone, and verteporfin. Growth curves were processed by plate to set the median OD at the start and end timepoints to 0 and 1, respectively. Then, the inventors determined reference compounds across all replicates that did not reduce growth significantly for most drugs: those were compounds for which measurements were available for >95% of replicates, and for which final OD was >0.5 for more than 122 out of 132 replicates. The inventors used these reference compounds as representatives of uninhibited growth. Since wells containing reference compounds outnumbered control wells within a plate, the inventors used control wells only later to verify the p691 value calculation. After determining reference compounds, the inventors rescaled growth curves such that the median growth of reference compounds at the end point is 1.

[0173] While growth curves in control wells and most wells with reference compounds followed the expected logistic growth pattern, a variety of deviations were observed for drugs that influenced growth. To quantify growth without relying on assumptions about the shape of the growth curve, the inventors calculated the area under the curve (AUC) using the trapezoidal rule. While the inventors set the median starting OD to 0, the OD of individual wells deviated from this. The inventors used two different methods to correct for this and determine the baseline for each growth curve. First, a constant shift was assumed, subtracting the same shift to all timepoints of the growth curve such that the minimum is zero. Second, an initial perturbation was assumed that affects initial timepoints more than later timepoints (e.g. condensation). To correct this, the inventors first subtracted a constant shift as above, and then rescaled the curve such that a timepoint with an uncorrected OD of 1 also has an OD of 1 after correction. AUCs were calculated for both scenarios, rescaled such that the AUC of reference compounds is 1, and then for each compound the baseline correction that yielded an AUC closest to 1 (i.e. normal growth) was selected. AUCs are highly correlated to final ODs, with a Pearson correlation of 0.95 across all compounds and replicates. Nonetheless, the inventors preferred to use AUCs to decrease the influence of the final timepoint, which will contain more noise than a measurement based on all timepoints.

[0174] Identification of Drugs with Anticommensal Activity

[0175] The inventors detected hits from normalized AUC measurements using a statistical method that controls for multiple hypothesis testing and varying data quality. The inventors fitted heavy-tailed distributions (scaled Student's t-distribution) to the wells containing reference compounds for each replicate and, separately, to each individual plate. These distributions captured the range of AUCs expected for compounds that did not reduce growth, and represented the null hypothesis that a given drug did not cause a growth defect in the given replicate or plate. The inventors calculated one-sided p-values from the cumulative distribution function of the fitted distribution. Within a replicate, each compound was associated with two p-values: one from the plate on which it was measured, and one for the whole replicate. Of those two, the highest p-value was chosen (conservative estimate) to control for plates with little or high noise, and varying levels of noise within the same replicate. The resulting p-values were well-calibrated (i.e. the distribution of p-values is close to uniform with the exception of a peak at low p-values) and captured the distribution of controls, which were not used for fitting the distribution and kept for validation. The inventors then combined p-values for a given drug and strain across replicates using Fisher's method. Lastly, the inventors calculated the False Discovery Rate (FDR) using the Benjamini-Hochberg method over the complete matrix of p-values (1197 compounds by 40 strains). After inspecting representative AUCs for compoundstrain pairs at different FDR levels, the inventors chose a conservative FDR cut-off of 0.01.

[0176] Drug Indications, Dose, and Administration

[0177] The inventors annotated drugs by their primary target organism on the basis of their WHO Anatomical Therapeutic Chemical (ATC) classification, or, if there were uncertainties, based on manual annotation. Compounds were classified as: antibacterial drugs (antibiotics, antiseptics), anti-infective drugs (acting against protozoa, fungi, parasites or viruses), human-targeted drugs (i.e. drugs whose mechanism of action affects human cells), veterinary drugs (used exclusively in animals), and finally non-drugs (which can be drug metabolites, drugs used only in research, or endogenous substances). If a human-use drug belonged to several classes, the drug class was picked according to this order of priority (from high to low): antibacterial, anti-infective, and human-targeted drug. This ensured that drugs used also as antibacterials were not classified in other two categories.

[0178] Drugs from the Prestwick Chemical Library were matched against STITCH 4 identifiers using CART. Identifiers that could not be mapped were annotated manually. Information about drug indications, dose and administration was extracted from the ATC classification system and Defined Daily Dose (DDD) database. Dose and administration data were also extracted from the Drugs@FDA resource. Doses that were given in grams were converted to mol using the molecular weight stated in the Prestwick library information files. When the dose guidelines mentioned salt forms, the inventors manually substituted the molecular weight. Dose data from Drugs@FDA stated the amount of drug for a single dose (e.g. a single tablet). Analyzing the intersection between Drugs@FDA and DDD, the inventors found that the median ratio between the single and daily doses is two. To combine the two datasets, the inventors therefore estimated the single dose as half of the daily dose.

[0179] In general, it is difficult to estimate intestinal drug concentrations, since those depend on the dose, the speed of dissolution, uptake and metabolization by human cells and by bacteria, binding to proteins, and excretion mechanisms into the gut. To estimate gut concentrations of drugs based on their dose, the inventors relied on a known in situ study. When 40 mg (57 mol) of posaconazole are delivered to the stomach in either an acidic or neutral solution, the maximum concentration in the duodenum reaches 26.310.3 and 13.65.8 M, respectively. The ratio between the dose and the duodenal concentration corresponds to a volume estimate of roughly three liters.

[0180] IC.sub.25 Determination/Screen Validation

[0181] To validate the inventors' screen, the inventors selected 25 drugs including human-targeted drugs (19), anitprotozoals (3), one antiparasitc, one antiviral and one no-drug compound. The human targeted drugs spanned 5 therapeutic classes (ATC codes A, G, L, M, N). The inventors' selection comprised mostly drugs with broad-spectrum activity in the inventors' screen (19 drug hits>10 strains). This bias was for ensuring that the inventors can also evaluate false positives. The inventors chose 15 strains to test IC.sub.25s, spanning different phyla (5) and including both sensitive (E. rectale, R. intestinalis) and resistant species (E. coli ED1a). Compounds of interest were purchased from independent sources and dissolved at 100x starting concentration in DMSO. 2-fold serial dilutions were prepared in 96-well U-bottom plates (same as screen). Each row contained a different drug at eleven 2-fold dilutions and a control DMSO well in the middle of the row (in total 8 drugs per plate). These master plates were diluted to 2 assay concentration and 2% DMSO in mGAM medium (50 l) and stored at 30 C. (<1 month). For the assay, plates were pre-reduced overnight in the anaerobic chamber, and mixed with equal volume (50 l) of appropriately diluted overnight culture (prepared as described for screening section) to reach a starting OD578 of 0.01 and a DMSO concentration of 1% across all wells. OD578 was measured hourly for 24 hrs after 1 min of shaking. Experiments were performed in two biological replicates.

[0182] Growth curves were converted to AUCs as described above, using in-plate control wells (no drug) to define normal growth. For each concentration, the inventors calculated the mean across the two replicates. The inventors further enforced monotonicity to conservatively remove noise effects: if the AUC decreased for lower concentrations, it was set to the highest AUC measured at higher concentrations. The IC25 was defined as the lowest concentration for which a mean AUC of below 0.75 was measured. Additionally, MIC was defined as the lowest concentration for which the AUC dropped below 0.1. In the large-scale screen, the inventors detected significant growth reductions, which do not necessarily correspond to complete growth inhibition. To ensure comparability between the results of the validation procedure and the screen, the inventors used the IC25 metric for benchmarking.

[0183] Analysis of Side Effects

[0184] Side effects (SEs) of drugs were extracted from the SIDER 4.1 database using the mapping between Prestwick compounds and STITCH 4 identifiers described above. In SIDER, SEs are encoded using the MedDRA terminology, which contains lower-level terms and preferred terms. Of these, the inventors used the preferred terms, which are more general. The inventors excluded rare SEs that occurred for less than five drugs from the analysis. Drugs with less than seven associated SEs were discarded. In a first pass, the inventors identified SEs associated with antibiotics in SIDER, by calculating for each SEs its enrichment for systemic antibiotics (ATC code J01) versus all other drugs using Fisher's exact test (p-value cut-off: 0.05, correcting for multiple hypothesis testing using the Benjamini-Hochberg method). Antibiotics are typically administered in relatively high doses, and some of the enriched SEs might therefore be caused by a dose-dependent effect (e.g. kidney toxicity). The inventors therefore used an ANOVA (Type II) to test if the presence of SEs for a drug is more strongly associated with it being an antibiotic or with its (log-transformed) dose. SEs that were more strongly associated with the dose were excluded from the list of antibiotics-related SEs.

[0185] Data on the incidence rates of SEs in patients was also extracted from SIDER 4.1. As different clinical trials can report different incidence rates, the inventors computed the median incidence rate per drug-SE pair. As SIDER also contains data on the incidence of SE upon placebo treatment, the inventors were able to ensure the absence of systematic biases.

[0186] Experimental Validation of Side Effect-Based Predictions

[0187] Selected candidate and control compounds belonged to multiple therapeutic classes (ATC codes A, B, C, G, H, L, M, N, S for candidate compounds and A, C, D, G, H, M N, R, S, V for control compounds).Compounds of interest were purchased from independent sources and if possible, dissolved at 5 mM concentration in mGAM. Lower concentrations were used when solubility limit was reached. Solutions were sterile filtered, and three 4-fold serial dilutions were arranged in 96 well plates, aiming at covering a broad range of drug concentrations. Inoculation and growth curve acquisition was performed as described for the MIC determination experiments.

[0188] Conjugation of the TransBac Overexpression Plasmid Library into E. Coli tolC

[0189] The TransBac library, a new E. coli overexpression library based on a single-copy vector (H. Dose & H. Moriunpublished resource) was conjugated in the BW25113 tolC::Kan strain. The receiver strain (BW25113 tolC::kan) was grown to stationary phase in LB medium, diluted to an OD of 1, and 200 l were spread on a LB plate supplemented with 0.3 mM diaminopimelic acid (DAP). Plates were dried for 1 hour at 37 C. and then a 1536 colony array of the library carried within a donor strain (BW38029 Hfr (CIP8 oriT::cat) dap-75) was pinned on top of the lawn. Conjugation was carried out at 37 C. for 6 hours, and the first selection was done by pinning on LB plates supplemented with tetracycline only (10 g/ml) and growing overnight. Two more rounds of selection followed on LB plates containing tetracycline (10 g/ml) and kanamycin (30 g/ml) to ensure killing of parental strains and select only for to/C mutants carrying the different plasmids.

[0190] Chemical Genomics Screen

[0191] The screen was carried out under aerobic conditions on solid LB Lennox medium (Difco), supplemented with 30 g/ml kanamycin, 10 g/ml tetracycline, the appropriate drug, and 0 or 100 M IPTG. Drugs were used at the following sub-inhibitory concentrations for the to/C mutant: diacerein 20 M, ethopropazine hydrochloride 160 M, tamoxifen citrate 20 M, niclosamide 1.25 M, thioridazine hydrochloride 40 M, methotrexate 320 M, or for the wildtype: metformin 100 mM. The 1536 colony array of BW25113 tolC::kan mutant carrying the TransBac collection was pinned on the drug-containing plates, and plates were incubated for 16-38 hours at 37 C. In the case of metformin the inventors used the version of the TransBac library, in which each plasmid complements its corresponding barcoded single-gene deletion mutant, since the inventors did not need to use the tolC background for sensitizing the cell. Growth of this library was determined at 0 and 100 mM metformin (both in the presence of 0, 50 and 100 M IPTG). All plates were imaged using an 18 megapixel Canon Rebel T3i (Canon inc USA) and images were processed using the Iris software.

[0192] Data Analysis

[0193] The inventors used colony size to measure the fitness of the mutants on the plate. For standardization of colony sizes, the inventors subtracted the median colony size and then divided by a robust estimate of the standard deviation (removing outliers below the 1st and above the 99th percentile). The inventors found edge effects affecting up to five rows and columns around the perimeter of the plate. The inventors therefore first standardized colony sizes across the whole plate using only colony sizes from the inner part of the plate as reference. To remove the edge effects, the inventors subtracted from each column its median colony size, and then from each row its median colony size. Finally, the inventors standardized the adjusted colony sizes using the whole plate as reference. The distribution of adjusted colony sizes was right-skewed (i.e. more outlier colonies with larger size), suggesting a log-normal distribution. At the same time, the presence of outliers suggested that a logarithmic equivalent of the Student's t-distribution with variable degree of freedom would be more suitable. The inventors fitted such a distribution for each plate and calculated p-values for both tails of the distribution. This approach assumes that the overexpression of most genes does not affect growth in response to drug treatment. p-values were combined using Fisher's method across replicates and IPTG concentrations (since the inventors noticed that different IPTG concentrations resulted to largely the same resultsi.e. plasmids are leaky). The inventors corrected for multiple hypothesis testing for each drug individually using the Benjamini-Hochberg method. Analysis of common resistance mechanisms: To determine a relationship between the number of human-targeted drugs (h) and the number of antibacterial drugs (a) that affect each strain, the inventors determined the odds ratio (OR):

[00001]$\mathrm{OR}=\frac{\frac{h}{H-h}}{\frac{a}{A-a}}$

[0194] Where H=204 and A=122 are the numbers of human-targeted and antibacterial drugs that show activity, respectively. The inventors computed the nonlinear least-squares estimate for OR based on the following equation:

[00002]$\frac{h}{H-h}=\mathrm{OR}.Math.\frac{a}{A-a}$

[0195] Results

[0196] A High-Throughput Drug Screen on Human Gut Bacterial Species

[0197] To systematically map interactions between drugs and human gut microbes, the inventors monitored the growth of 40 representative isolates upon treatment with 1197 compounds in modified Gifu Anaerobic Medium broth (mGAM), which partially recapitulates species abundances in the gut, under anaerobic atmosphere, at 37 C. The inventors used the Prestwick Chemical Library, consisting mostly of off-patent FDA-approved compounds and spanning a wide range of chemical and pharmacological diversity. Most compounds are administered to humans (1079), covering all main therapeutic classes. Three quarters (835) are human-targeted drugs (i.e. have molecular targets in human cells), whereas the rest are anti-infectives: 156 with antibacterial activity (144 antibiotics and 12 antiseptics) and 88 mainly effective against fungi, viruses, and protozoan or metazoan parasites (FIG. 1a). All compounds were screened at a concentration of 20 M, which is within the range of what is commonly used in high-throughput drug screens and on average, slightly below doses administered to patients (FIG. 2a).

[0198] To be representative of the gut microbiome of healthy individuals, the inventors profiled a diverse set of ubiquitous gut bacterial species. Prevalence and abundance in the human gut, and phylogenetic diversity were the inventors' main selection criteria. In a few cases the selection was constrained due to strain unavailability or irreproducible growth in mGAM. In total, the inventors included 40 human gut isolates from 38 bacterial species and 21 genera (with E. coli and B. fragilis being represented by two different strains), accounting together for 78% of the assignable median relative abundance of the human gut microbiome at genus level (60% at species level). Most of these strains are commensals, and represent 31 out of the 60 species with available reference genome sequences detected at a relative abundance of 1% and prevalence of 50% in a large collection of fecal samples of asymptomatic humans from three continents. In addition, the set includes a few pathobionts (Clostridium difficile, Clostridium perfringens, Fusobacterium nucleatum and an enterotoxigenic strain of Bacteroides fragilis), a probiotic (Lactobacillus paracasei) and two further commensal Clostridia (C. ramosum and C. saccharolyticum), all human isolates. All bacteria probed were part of a larger resource representing the core of the healthy human gut microbiome.

[0199] The experimental setup included screening all compounds, arrayed in 96- or 384-well plates, in at least three biological replicates for each strain. The inventors measured optical density over time as growth readout, quantifying the area under the growth curve (AUC) up to the transition to stationary phase (estimated on controls with unperturbed growth). Correlation between replicates was very good (median 0.89). The inventors then tested for significant deviations from the normalized AUC distribution of samples with unperturbed growth, combining p-values across replicates and correcting for multiple hypothesis testing on the complete matrix of compounds and strains. If a drug significantly reduced growth of at least one tested strain (FDR<0.01), and thus has potential to modulate the human gut microbiota, the inventors classified it as a hit with anticommensal activity.

[0200] Of the 156 antibacterials present in the inventors' screen, 78% were active against at least one gut commensal species, typically with broad activity spectrum (FIGS. 1a-b). Inactive antibiotics mainly belong to the classes of sulfonamides (active at higher concentrations), aminoglycosides (mostly inactive in anaerobes and/or under anaerobic conditions due to limited drug uptake) and rather specific antimycobacterial drugs. Thus, although antibiotics are used to inhibit pathogens, they also target gut commensals. This is presumably due to their broad spectrum activity. Although the medical importance of this collateral damage of antibiotics to the resident microbiome is becoming increasingly clear, their specific activities against diverse microbiome species had not been mapped at this scale before.

[0201] Interestingly, 27% of the non-antibiotic drugs (i.e. other anti-infectives and human targeted drugs) were also active in the inventors' screen. More than half of the anti-infectives against viruses or eukaryotes exhibited anticommensal activity (47 drugs 53%; FIGS. 1a-b). Antibacterial activity has been previously reported for many of them, including the antifungal imidazoles (10 in the inventors' screen), but for particular others (for example the antivirals efavirenz and trifluridine) such activity is new. More noteworthy and novel is the anticommensal activity observed for human-targeted drugs (24%). In contrast to anti-infectives, most human-targeted drugs were effective against a small subset of strains with a number of notable exceptions: 36 drugs affected >10 strains, with 11 having no previously reported antibacterial activity. From the known ones, auranofin was recently reported to have broad-spectrum bactericidal activity and even target multi-drug resistant isolates. Another compound, the ovulation stimulant clomiphene, targets a widespread and conserved bacterial enzyme in undecaprenyl phosphate synthesis, which is an essential precursor for cell wall carbohydrate polymers. The scaffolds of such non-antibiotic drugs with antimicrobial activity can be used as starting points for repurposing towards broad spectrum antibiotics. On the other hand, the microbial specificity of most human-targeted drugs suggests that their scaffolds could be used in the future for developing narrow-spectrum antibacterials and/or modulators of the microbiome composition.

[0202] When considering the total number of drugs inhibiting each bacterial isolate, it is apparent that some species are influenced more than others, with the abundant Roseburia intestinalis, Eubacterium rectale and Bacteroides vulgatus, being the most susceptible, and -proteobacteria representatives being the most resistant (FIG. 1a). Overall, species with higher relative abundance across healthy individuals were significantly more susceptible to human-targeted drugs in the inventors' screen (FIG. 1c). This suggests that human-targeted drugs have an even larger overall impact to the gut microbiome with key species related to its healthy status, such as major butyrate- (E. rectale, R. intestinalis, Coprococcus comes) and propionate-producers (B. vulgatus, Prevotella copri, Blautia obeum), and enterotype drivers being impacted the most.

[0203] Many More Human-Targeted Drugs are Likely to Inhibit Gut Bacteria

[0204] The notion that the impact of human-targeted drugs on the gut microbiome may be even broader than the inventors' screen revealed is supported by several lines of evidence. First, the drug concentration used in this screen (20 M), is below the median estimated gut concentration of the drugs that the inventors tested (FIG. 2a). Since the effective concentration of a drug in the gut is rarely measured, the inventors relied on available data on recommended administration doses for 653 human-targeted drugs, and used detailed measurements of intestinal concentrations for the case of posaconazole to covert administration doses to estimates of gut concentrations for the other drugs. Interestingly, the human targeted drugs with anticommensal activity in the inventors' screen have lower estimated intestinal concentrations than ones without (FIG. 2a; p=0.001, Wilcoxon rank sum test), suggesting that more human-targeted drugs would inhibit bacterial growth if probed at higher doses, closer to recommended administration levels. A case in point is the antidiabetic drug metformin, which was recently identified as the key contributor to changes in the human gut microbiome composition of type-II diabetes (T2D) patients, but did not show anticommensal activity in the inventors' screen. Metformin reaches up to 10 mM plasma concentration in treated T2D patients, and its small intestine concentration is calculated to be 30-300 fold higher, which are both much higher than the screen concentration (20 M). Similarly, the inventors' estimated gut concentration for metformin is at 5 mM (FIG. 2a). When the inventors probed for higher concentrations of metformin, 5/22 strains had a Minimal Inhibitory Concentration (MIC) <10 mM, and further strains had a MIC an order of magnitude higher, which is still within physiological levels of metformin in the gut.

[0205] Second, benchmarking the inventors' screen with an independent set of targeted validation experiments (MIC testing for 22 selected drugs in a subset of 15 strains; see Methods), revealed excellent precision (96%), but slightly lower recall (85%) due to more false negatives (FNs), i.e. drugs with anticommensal that the inventors missed in the inventors' screen (FIG. 2b). Many FNs were due to screen biases, as they mostly came from a few sensitive chemicals that probably lost activity during the screening process (for example, see loperamide or acarbose) and the inventors' stringent FDR cutoff for calling hits. Indeed, increasing the FDR threshold to 0.1 would almost double the fraction of drugs impacting human gut commensals. Along these lines, the validation experiments revealed that more species were inhibited at higher concentrations, confirming the idea more human-targeted drugs would have had anticommensal activity, if the inventors had screened higher concentrations within the recommended administration doses.

[0206] Third, the human gut microbiome harbors hundreds of species and an even larger diversity of strains, whereas the inventors only screened a small representative subset. Rarefaction analysis indicates that if more gut species were tested, the fraction of human-targeted drugs with impact on commensals would increase (FIG. 2c). In contrast, the number of antibacterial drug hits saturates early within the strains tested, indicating that screening more species would not substantially increase the fraction of antibiotic hits in the screen. Taken together, these results suggest that a considerably higher proportion of human targeted drugs than the 24% the inventors' screen reports is likely to inhibit the growth of human gut microbes.

[0207] Side Effects of Human-Targeted Drugs Validate their Systemic Impact

[0208] Although the inventors demonstrated that human-targeted drugs commonly inhibit gut microbes in vitro, evidence that such effects also manifest in vivo in the human gut currently exists only for a handful of cases. To bridge this gap and address the physiological relevance of the inventors' screen, the inventors looked into the registered effects that these drugs have in humans. The rationale of the inventors was that if human-targeted drugs target the gut microbiome, some of the consequences should be apparent from their side effects, which should exhibit some similarity to those of antibiotics.

[0209] The inventors first identified side effects enriched in antibiotics for systemic use compared to those found in all other drugs in the SIDER database version 4.1 25. The inventors identified 69 side effects that were enriched in antibiotics, excluding side effects that are likely caused by high-dose-related host toxicity. The inventors then tested whether antibiotic related side effects occurred with higher frequency in clinical trials for human-targeted drugs with anticommensal activity compared to inactive compounds in the inventors' screen, which turned out to be true (p =0.002, Wilcoxon rank sum test, FIG. 3a). No significant difference was observed for patients receiving a placebo, suggesting the absence of biases (FIG. 3a).

[0210] The analysis above suggests that the collateral damage of human-targeted drugs on gut microbes can be detected by the higher occurrence of antibiotic-like side effects in patients. The inventors thus wondered whether this side effect signature could be used to predict anticommensal activity of other human-targeted drugs, which the inventors may have missed due to the low drug concentration used in the inventors' screen (FIG. 2a). To test this hypothesis, the inventors screened 26 candidate compounds with high enrichment of antibiotic-related side effects and 16 without (control compounds) in 18 strains (FIG. 3b), in concentrations up to 2500 M. Of the 42 compounds in total, 28 inhibited the growth of at least one strain (FIG. 3b). For both candidate and control compounds, the fraction of growth-inhibiting compounds was close to two thirds, and there was no significant difference in the number of affected strains. However, when the inventors normalized the measured MICs by the recommended drug doses to make amounts comparable between drugs, a significant difference was evident. Drugs predicted to be active had a median MIC across all drug-strain pairs that corresponded to 4.3 drug doses, compared to 12 for control drugs (p=2e-6, one-sided Wilcoxon rank sum test; FIG. 3c). For drugs predicted to have anticommensal activity, 34% of the MICs correspond to less than two drug doses, compared with just 8% for control drugs. Interestingly, all seven NSAIDs among the inventors' predictions showed anticommensal activity at higher concentrations, affecting 6 to 18 strains, and 44% of these MICs corresponded to less than two drug doses. This is consistent with recent metagenomics studies associating NSAID use with microbiome changes.

[0211] In summary, side effect patterns in humans can differentiate human-targeted drugs with anticommensal activity from those without, confirming the physiological relevance of the inventors' in vitro screen. The inventors thus explored next the chemical and biological properties of the interacting drugs and bacteria, respectively.

[0212] Therapeutic Indication Areas and Chemical Properties of Human-Targeted Drugs with Anticommensal Activity

[0213] Drugs from all major indication areas according to the Anatomical Therapeutic Chemical classification (ATC) inhibited growth of at least one gut microbe, with drugs applied topically (classes S and D) exhibiting the lowest hit rate (FIG. 4a). At the other side of the spectrum, antineoplastics, hormones and compounds targeting our nervous system inhibited gut microbes more than other medications (FIG. 4a). Within the ATC classification, three subclasses were significantly enriched in hits: antimetabolites, antipsychotics and calcium channel blockers (FIG. 4a). Antimetabolites are used as chemotherapeutic and immunosuppressant agents with their incorporation into RNA/DNA or their interaction with RNA/DNA synthesis enzymes having cytotoxic effects to human cells. Their molecular targets and cytotoxicity are often conserved in bacteria, explaining the observed effects. In addition, the inventors' results imply that antimetabolites could play a more direct role in mucositis development during chemotherapy.

[0214] The enrichment in antipsychotics is less straightforward, given that these target dopamine and serotonin receptors in the brain, which are absent in bacteria. Although phenothiazines are known to have antibacterial effects, the inventors observed anticommensal activity for nearly all subclasses of the chemically diverse antipsychotics in the inventors' screen, and the pattern of species inhibited was similar even for chemically distinct sub-classes (FIG. 4b). In general, antipsychotics targeted more similar sets of species than expected, based on their chemical similarity (FIG. 4c). This raises the possibility that direct bacterial inhibition may not only manifest as side effects for antipsychotics but also be part of their MoA.

[0215] Many therapeutic sub-classes do not have enough representatives in the inventors' screen to yield a statistically significant enrichment for anticommensal activity. Among them, all three PPIs that are part of the Prestwick Chemical Library exhibited broad anticommensal activity. When comparing the inhibited species in the inventors' study to the microbiome changes in patients using PPIs, the inventors found high concordance. Taxa with reduced abundance in patients included drugstrain pairs in this screen with reduced growth, while enriched taxa were rarely inhibited by PPIs in the inventors' study. This suggests that PPI could also influence directly the gut microbiome composition, in addition to changing the stomach pH and thereby the microbes that can reach our gut.

[0216] As indication areas often contain chemically similar drugs, the inventors explored whether certain chemical properties of drugs can influence their anticommensal activity. To some degree, human-targeted drugs with higher chemical similarity had more similar effects in the screen. The inventors also tested a number of compound properties including complexity, molecular weight, topological polar surface area (TPSA), volume, and XLogP as a measure of hydrophobicity. Complex, heavier and larger compounds preferentially target Gram-positive bacteria, whereas Gram-negative bacteria are protected against such bulkier drugs. This is in accordance with the selective outer membrane barrier of Gram-negative bacteria, which confers protection against bulky and/or hydrophobic drugs. Due to the very large number of chemical moieties present in drugs with anticommensal activity, the inventors did not attempt an exhaustive enrichment analysis. Nevertheless, the inventors did observe reactive nitro-groups to be significantly enriched in drugs with anticommensal activity (p=6.4 e-06), indicating that local chemical properties may confer antibacterial activity. Thus, despite the wide range of indication areas and chemical diversity of anticommensals, some chemical properties of human-targeted drugs associate with their antibacterial spectrum.

[0217] Human-Targeted Drug Consumption May Promote Antibiotic Resistance

[0218] The inventors next investigated whether resistance mechanisms influenced the spectrum of effects observed for gut microbes. Intriguingly, the inventors noticed a strong correlation between resistance to antibacterials and human-targeted drugs (FIG. 5a). This suggested that susceptibility towards xenobiotics in general is determined by intrinsic properties of the individual bacterial strains. These properties go beyond general cell envelope composition, as there is no clear division between Gram-positive and Gram-negative bacteria in the inventors' data (FIG. 5a). The inventors reasoned that more specific, yet common mechanisms could confer resistance against both antibiotics and human-targeted drugs. To test this hypothesis for one of the most common resistance mechanisms against antibiotics, that of efflux pumps, the inventors selected a prominent member, TolC, known to confer resistance to several antibiotics in E. coli and many other bacteria. The inventors profiled an E. coli tolC mutant strain and its parental wildtype (BW25113) against all the compounds of the Prestwick Chemical Library. E. coli lacking TolC did not only become more sensitive to antibacterials (22 hits more than wildtype), but also became equally more sensitive to human-targeted drugs (19 additional hits; FIG. 5a). This confirms the existence of cross-resistance mechanisms between antibiotic and non antibiotic drugs.

[0219] To more systematically elucidate mechanisms conferring resistance against human-targeted drugs, the inventors employed a chemical genomics approach and screened a genome-wide overexpression library in E. coli against seven non-antibiotic drugs (six human-targeted drugs and niclosamide, an antiparasitic) with broad impact on gut microbes in the inventors' screen. Since wildtype E. coli was one of the most resistant species in the inventors' screen (FIG. 5a), the inventors decided to use the to/C mutant that is sensitive to many of these drugs, allowing the inventors to probe further resistance mechanisms. For all tested drugs except metformin, overexpression of to/C rescued E. coli growth, as expected. Furthermore, the inventors identified a number of diverse transporter families contributing to specific resistance against these drugs (FIG. 5b). Many of them have been linked to antibiotic resistance in the past. Resistance was also acquired by overexpression of transcription factors (e.g. rob, which is known to control efflux pump expression), the ribosome maturation factor rrmA, which plays a role in resistance to the antibiotic viomycin, and detoxification mechanisms (nitroreductases are known to modify nitro-containing antibiotics). Interestingly, for the case of methotrexate, the inventors' chemical genomics screen identified the already known primary target in bacteria (E. coli dihydrofolate reductase), illustrating the potential of this approach to identify bacterial MoA of human targeted drugs. All these results support the concept of an overlap between resistance mechanisms against antibiotics and human-targeted drugs. This implies a hitherto unnoticed risk of acquiring antibiotic resistance by consumption of non-antibiotic drugs.

[0220] Table 1 shows isolated strains of the human microbiota used in this study.

TABLE-US-00001 Data base (NT) Phylum Class Order Family Genus 5001 Bacteroidetes Bacteroidia Bacteroidales Bacteroidaceae Bacteroides 5002 Bacteroidetes Bacteroidia Bacteroidales Bacteroidaceae Bacteroides 5003 Bacteroidetes Bacteroidia Bacteroidales Bacteroidaceae Bacteroides 5004 Bacteroidetes Bacteroidia Bacteroidales Bacteroidaceae Bacteroides 5006 Firmicutes Clostridia Clostridiales Clostridiaceae Clostridium 5009 Firmicutes Clostridia Clostridiales Eubacteriaceae Eubacterium 5011 Firmicutes Clostridia Clostridiales Lachnospiraceae Roseburia 5017 Firmicutes Negativicutes Seleno- Veillonellaceae Veillonella mondales 5019 Bacteroidetes Bacteroidia Bacteroidales Prevotellaceae Prevotella 5021 Verruco- Verruco- Verruco- Verruco- Akkermansia microbia microbiae microbiales microbiaea 5022 Actinobacteria Actinobacteria Bifido- Bifido- Bifido- bacteriales bacteriaceae bacterium 5024 Actinobacteria Actinobacteria Corio- Corio- Eggerthella bacteriales bacteriaceae 5025 Fusobacteria Fusobacteria Fuso- Fuso- Fuso- bacteriales bacteriaceae bacterium 5026 Firmicutes Clostridia Clostridiales Clostridiaceae Clostridium 5028 Actinobacteria Actinobacteria Bifido- Bifido- Bifido- bacteriales bacteriaceae bacterium 5032 Firmicutes Clostridia Clostridiales Clostridiaceae Clostridium 5033 Bacteroidetes Bacteroidia Bacteroidales Bacteroidaceae Bacteroides 5036 Proteobacteria Deltaproteobacteria Desulfovibrionales Desulfovibrionaceae Bilophila 5037 Firmicutes Clostridia Clostridiales Clostridiaceae Clostridium 5038 Firmicutes Bacilli Lactobacillales Strepto- Strepto- coccaceae coccus 5042 Firmicutes Bacilli Lactobacillales Lactobacillaceae Lactobacillus 5045 Firmicutes Clostridia Clostridiales Rumino- Rumino- coccaceae coccus 5046 Firmicutes Clostridia Clostridiales Lachnospiraceae Blautia 5047 Firmicutes Clostridia Clostridiales Lachnospiraceae Blautia 5048 Firmicutes Clostridia Clostridiales Lachnospiraceae Copro- coccus 5050 Bacteroidetes Bacteroidia Bacteroidales Bacteroidaceae Bacteroides 5054 Bacteroidetes Bacteroidia Bacteroidales Bacteroidaceae Bacteroides 5064 Bacteroidetes Bacteroidia Bacteroidales Bacteroidaceae Bacteroides 5069 Firmicutes Clostridia Clostridiales Lachnospiraceae Blautia 5071 Bacteroidetes Bacteroidia Bacteroidales Porphyro- Para- monadaceae bacteroides 5072 Firmicutes Bacilli Lactobacillales Strepto- Strepto- coccaceae coccus 5073 Actinobacteria Actinobacteria Corio- Corio- Collinsella bacteriales bacteriaceae 5074 Bacteroidetes Bacteroidia Bacteroidales Porphyro- Para- monadaceae bacteroides 5075 Firmicutes Clostridia Clostridiales Eubacteriaceae Eubacterium 5076 Firmicutes Clostridia Clostridiales Lachnospiraceae Dorea 5077 Proteobacteria Gammaproteo- Enterobacteriales Enterobacteriaceae Escherichia bacteria 5078 Proteobacteria Gammaproteo- Enterobacteriales Enterobacteriaceae Escherichia bacteria 5079 Firmicutes Clostridia Clostridiales Lachnospiraceae Roseburia 5081 Bacteroidetes Bacteroidia Bacteroidales Porphyromonadaceae Odoribacter 5083 Firmicutes Clostridia Clostridiales Clostridiaceae Peptoclostridium Data base Medium (NT) Species Strain Source preference 5001 Bacteroides type strain DSM mGAM vulgatus 1447 5002 Bacteroides VPI 0061 DSM mGAM uniformis 6597 5003 Bacteroides EN-2, VPI DSM mGAM fragilis 2553 2151 nontoxigenic 5004 Bacteroides E50(VPI DSM mGAM thetaiotao- 5482) 2079 micron 5006 Clostridium type strain, DSM mGAM ramosum 113-I, VPI 1402 0427 5009 Eubacterium A1-86 DSM mGAM rectale 17629 5011 Roseburia L1-82 DSM mGAM intestinalis 14610 5017 Veillonella type strain, DSM Todd- parvula Te3 2008 Hewitt + 0.6 % sodium lactate 5019 Prevotella type strain, DSM mGAM copri CB7 18205 5021 Akkermansia type strain, DSM mGAM muciniphila Muc 22959 5022 Bifido- type DSM mGAM bacterium strainE194 20083 adolescentis a (Variant a) 5024 Eggerthella type strain, DSM mGAM lenta 1899 B, 2243 VPI 0255 5025 Fusobacterium type strain, DSM mGAM nucleatum 1612A, VPI 15643 subsp. 4355 Nucleatum 5026 Clostridium type strain, DSM mGAM bolteae WAL 15670 16351 5028 Bifido- type strain, DSM mGAM bacterium E194b 20219 longum subsp. (Variant a) Longum 5032 Clostridium C36 DSM mGAM perfringens 11782 5033 Bacteroides 20656-2-1 ATCC mGAM fragilis 43860 enterotoxigenic (ET) 5036 Bilophila type strain, ATCC mGAM wadsworthia WAL 7959 49260 supplemented [Lab 88- with 60 130H] mM sodium formiate and 10 mM taurine 5037 Clostridium type strain, DSM mGAM saccharo- WM1 2544 lyticum 5038 Strepto- type strain, DSM mGAM coccus 275 20560 salivarius 5042 Lactobacillus LPC-37, Dupont mGAM paracasei ATCC Health SD5275 and Nutrition 5045 Rumino- type strain, ATCC mGAM coccus bromii VPI 6883 27255 5046 Rumino- type strain, ATCC mGAM coccus gnavus VPI C7-9 29149 5047 Ruminococcus type strain, ATCC mGAM torques VPI B2-51 27756 5048 Copro-coccus type strain, ATCC mGAM comes VPI CI-38 27758 5050 Bacteroides Type DSM mGAM caccae strain, 19024/ CCUG ATCC 38735, CIP 43185 104201, JCM 9498, NCTC 13051, VPI 3452A 5054 Bacteroides NCTC ATCC mGAM ovatus 11153, 8483 Type strain 5064 Bacteroides Strain BEI mGAM xylan-isolvens CL03T12C04, Resources HM-722 5069 Blautia obeum type strain DSM mGAM 26238 5071 Para- VPI T4-1 DSM mGAM bacteroides [CIP 19495 merdae 104202T, JCM 9497] 5072 Strepto- type strain DSM mGAM coccus 6778 parasanguinis 5073 Collinsella ATCC DSM mGAM aerofaciens 25986, 3979 type strain VPI 1003 5074 Para- ATCC DSM mGAM bacteroides 8503, 20701 distasonis CCUG 4941, JCM 5825, NCTC 11152 5075 Eubacterium C15-B4, DSM mGAM eligens type strain 3376 5076 Dorea VPI C8-13 DSM mGAM formicigenerans 3992 5077 Escherichia IAI1 Denamur mGAM coli IAI1 Lab (INSERM) 5078 Escherichia ED1a Denamur mGAM coli ED1a Lab (INSERM) 5079 Roseburia A2-183, DSM GMM + hominis type strain 16839 mGAM 5081 Odoribacter 1651/6, DSM mGAM splanchnicus type strain 20712 5083 Clostridium 630 DSM mGAM difficile 27543

[0221] Table 2 shows additional bacteroides used in this study

TABLE-US-00002 5057 Bacteroidetes Bacteroidia Bacteroidales Bacteroidaceae Bacteroides Bacteroides HM-20, BEI mGAM fragilis Strain Resources 3_1_12 5065 Bacteroidetes Bacteroidia Bacteroidales Bacteroidaceae Bacteroides Bacteroides HM-715, BEI mGAM uniformis Strain Resources CL03T00C 23

[0222] Table 3 shows laboratory E. coli strains used in this study

TABLE-US-00003 data base Phylum Class Source Medium preference NT5084 BW25113 BW25113 WT PMID: 10829079 mGAM NT5085 BW25113 BW25113 tolC::kan from Keio collection JW5503-1 mGAM (PMID: 16738554)

[0223] Table 4 shows narrow-spectrum compounds effective against C. difficile. Shown are the adjusted p-values of the compound-induced inhibition on the growth of 40 different bacterial species as indicated. Note: A compound is defined as narrow-spectrum if it inhibits the growth of less than 9 species of 40 different bacterial species tested.

TABLE-US-00004 Meclozine Carbenoxolone Fendiline Telmisartan Azathioprine Mercaptopurine Mifepristone Montelukast Cilnidipine Fentiazac Lacidipine dihydrochloride disodium salt Tribenoside Gliquidone hydrochloride Amlodipine prestwick_ID 1350 94 1469 299 1189 1376 1254 1297 457 837 1019 991 270 1219 n_hit 4 7 4 3 8 5 1 6 8 2 1 1 8 2 Akkermansia 1.000 0.413 0.222 1.000 1.000 1.000 1.000 1.000 0.733 0.852 1.000 1.000 1.000 1.000 muciniphila (NT5021) Bacteroides 0.868 0.175 1.000 0.024 1.000 0.926 1.000 1.000 0.029 0.429 0.058 0.886 0.994 1.000 caccae (NT5050) Bacteroides 0.018 1.000 1.000 0.022 0.340 0.010 0.684 0.154 0.086 0.936 0.068 0.815 0.490 0.780 fragilis (ET) (NT5033) Bacteroides 0.242 0.917 0.907 0.010 0.253 0.340 0.619 1.000 0.059 1.000 0.254 1.000 0.016 0.783 fragilis (NT) (NT5003) Bacteroides 1.000 0.220 1.000 0.114 1.000 1.000 1.000 0.835 0.021 0.344 0.044 1.000 0.952 0.838 ovatus (NT5054) Bacteroides 0.224 0.517 0.450 0.079 0.961 0.859 1.000 0.579 0.015 0.943 0.801 0.773 0.293 1.000 thetaiotaomicron (NT5004) Bacteroides 0.316 0.510 1.000 0.003 0.937 0.507 0.336 1.000 0.016 0.999 0.833 1.000 <0.001 0.637 uniformis (NT5002) Bacteroides 0.118 0.927 1.000 0.021 0.799 0.880 1.000 0.774 0.001 1.000 0.230 1.000 <0.001 0.579 vulgatus (NT5001) Bacteroides 0.783 0.105 1.000 0.131 1.000 1.000 1.000 0.782 1.000 0.356 0.188 1.000 1.000 1.000 xylanisolvens (NT5064) Bifidobacterium 1.000 0.714 0.851 0.724 1.000 NA 1.000 0.002 0.037 0.132 0.733 0.913 1.000 1.000 adolescentis (NT5022) Bifidobacterium 1.000 0.141 0.605 0.853 1.000 0.938 1.000 0.015 0.092 0.150 0.601 1.000 0.689 0.895 longum (NT5028) Bilophila 0.519 1.000 1.000 1.000 0.055 0.730 0.878 0.632 1.000 0.861 0.944 1.000 0.626 0.853 wadsworthia (NT5036) Blautia obeum 0.358 <0.001 0.014 1.000 0.091 0.018 1.000 0.070 <0.001 0.030 0.315 1.000 0.337 1.000 (NT5069) Clostridium 0.907 0.024 1.000 0.886 0.139 NA 1.000 0.839 0.016 0.580 0.415 1.000 0.521 0.828 bolteae (NT5026) Clostridium <0.001 <0.001 <0.001 <0.001 0.001 0.001 0.001 0.001 0.001 0.003 0.005 0.006 0.006 0.009 difficile (NT5083) Clostridium 0.007 0.884 1.000 0.358 <0.001 0.152 1.000 0.004 0.026 <0.001 0.283 0.809 0.350 1.000 perfringens (NT5032) Clostridium 0.262 0.943 0.591 0.581 1.000 0.860 1.000 0.431 0.031 0.035 0.565 0.678 0.741 1.000 ramosum (NT5006) Clostridium 0.693 0.094 0.148 0.561 1.000 0.892 1.000 1.000 0.037 0.712 0.147 0.394 0.435 1.000 saccharolyticum (NT5037) Collinsella 1.000 0.003 0.002 0.820 1.000 1.000 1.000 0.819 <0.001 0.087 0.203 1.000 0.230 1.000 aerofaciens (NT5073) Coprococcus 0.027 0.001 <0.001 0.882 <0.001 NA 1.000 0.009 0.050 0.471 0.411 1.000 0.004 0.145 comes (NT5048) Dorea 1.000 0.629 0.688 0.907 0.350 0.779 1.000 1.000 0.018 0.222 0.098 1.000 0.211 0.996 formicigenerans (NT5076) Eggerthella 0.503 0.008 0.027 1.000 1.000 0.565 0.992 1.000 0.768 0.249 0.815 1.000 1.000 1.000 lenta (NT5024) Escherichia coli 0.452 1.000 1.000 0.410 1.000 0.809 0.895 0.781 0.392 0.862 0.750 1.000 0.228 0.806 ED1a (NT5078) Escherichia coli 0.400 1.000 1.000 0.459 1.000 0.463 1.000 1.000 0.046 0.712 0.767 1.000 0.784 0.636 IAI1 (NT5077) Eubacterium 1.000 0.888 1.000 1.000 0.845 0.137 1.000 1.000 0.001 0.694 1.000 1.000 0.009 0.011 eligens (NT5075) Eubacterium 0.131 0.025 0.092 0.165 0.007 NA 1.000 0.093 0.009 0.923 0.046 1.000 0.005 0.188 rectale (NT5009) Fusobacterium 0.589 0.153 0.189 0.167 1.000 1.000 0.638 1.000 0.569 1.000 0.013 1.000 1.000 1.000 nucleatum (NT5025) Lactobacillus 1.000 0.951 1.000 0.182 0.783 0.945 0.979 0.653 0.158 0.953 1.000 1.000 0.309 1.000 paracasei (NT5042) Odoribacter 1.000 0.369 0.607 0.970 1.000 1.000 1.000 1.000 0.110 0.391 0.564 1.000 0.831 0.985 splanchnicus (NT5081) Parabacteroides 0.004 1.000 1.000 0.019 <0.001 0.003 0.667 0.560 0.161 0.943 0.109 0.018 0.534 0.833 distasonis (NT5074) Parabacteroides 0.116 0.462 0.925 <0.001 0.054 0.010 0.205 1.000 0.068 1.000 0.032 0.990 0.032 1.000 merdae (NT5071) Prevotella copri 0.245 <0.001 <0.001 0.309 0.061 NA 0.720 0.185 0.272 0.729 0.527 0.368 0.009 0.174 (NT5019) Roseburia 1.000 1.000 1.000 0.378 1.000 1.000 1.000 0.609 1.000 0.637 0.822 1.000 1.000 0.001 hominis (NT5079) Roseburia 0.012 0.559 0.952 0.212 0.001 NA 1.000 0.007 0.001 1.000 0.391 1.000 0.003 1.000 intestinalis (NT5011) Ruminococcus 1.000 0.638 1.000 0.923 0.380 0.643 0.964 0.017 0.934 0.199 0.258 0.903 1.000 0.261 bromii (NT5045) Ruminococcus 0.718 0.669 0.770 0.331 0.001 0.083 0.949 0.414 0.021 0.498 0.082 1.000 0.024 1.000 gnavus (NT5046) Ruminococcus 0.322 0.266 0.632 0.114 0.833 0.049 1.000 0.875 0.087 1.000 0.078 1.000 0.519 1.000 torques (NT5047) Streptococcus 0.042 1.000 0.706 0.098 0.083 0.009 0.448 0.035 0.033 1.000 0.024 1.000 0.181 1.000 parasanguinis (NT5072) Streptococcus 0.007 0.986 0.769 1.000 <0.001 0.353 0.192 0.008 0.002 0.041 0.427 0.388 0.416 1.000 salivarius (NT5038) Veillonella 1.000 0.003 1.000 0.867 1.000 1.000 1.000 0.947 0.526 0.106 0.766 1.000 1.000 0.925 parvula (NT5017)

[0224] Table 5 shows narrow-spectrum compounds effective against bacterial species but Clostridium difficile and an enterotoxigenic strain of Bacteroides fragilis. Shown are the adjusted p-values of the compound-induced inhibition on the growth of 40 different bacterial species as indicated. Note: A compound is defined as narrow-spectrum if it inhibits the growth of less than 9 species of 40 different bacterial species tested.

TABLE-US-00005 Ethacrynic Chlorpromazine Cyclosporin Flufenamic Alfacalcidol Acarbose acid hydrochloride A acid Aripiprazole Idebenone Thioguanosine prestwick_ 1211 1174 259 64 435 203 1229 1288 347 ID n_hit 7 6 6 6 5 4 3 3 3 Akkermansia 0.280 0.967 1.000 <0.001 0.835 0.758 0.924 1.000 0.885 muciniphila (NT5021) Bacteroides 0.568 1.000 1.000 0.619 1.000 0.194 1.000 1.000 0.373 caccae (NT5050) Bacteroides 0.784 <0.001 0.001 0.026 0.913 0.027 0.147 0.540 1.000 fragilis (ET) (NT5033) Bacteroides 0.287 0.018 0.001 0.036 1.000 0.094 0.389 0.384 0.310 fragilis (NT) (NT5003) Bacteroides 0.983 0.792 0.349 0.124 0.362 0.001 1.000 0.923 0.712 ovatus (NT5054) Bacteroides 1.000 0.020 0.063 0.013 1.000 0.073 0.934 1.000 0.312 thetaiotaomicron (NT5004) Bacteroides 0.590 0.001 <0.001 0.003 1.000 0.247 0.913 0.596 0.924 uniformis (NT5002) Bacteroides 0.183 0.001 <0.001 <0.001 0.848 0.527 0.612 0.782 1.000 vulgatus (NT5001) Bacteroides 1.000 0.00 0.617 1.000 0.341 0.611 1.000 1.000 0.836 xylanisolvens (NT5064) Bifidobacterium 0.010 1.000 1.000 1.000 0.143 0.298 1.000 1.000 0.234 adolescentis (NT5022) Bifidobacterium 0.008 1.000 1.000 0.713 0.004 0.377 1.000 1.000 0.324 longum (NT5028) Bilophila 0.242 0.713 0.887 0.919 1.000 1.000 1.000 1.000 0.815 wadsworthia (NT5036) Blautia 0.408 1.000 0.792 0.759 0.004 0.056 1.000 1.000 0.519 obeum (NT5069) Clostridium 0.653 1.000 0.006 0.864 1.000 0.501 0.003 0.394 0.549 bolteae (NT5026) Clostridium 0.382 1.000 1.000 0.362 0.611 0.078 0.799 1.000 0.628 difficile (NT5083) Clostridium <0.001 <0.001 0.002 0.001 <0.001 0.678 1.000 0.005 <0.001 perfringens (NT5032) Clostridium 0.027 1.000 0.032 0.957 1.000 0.433 1.000 0.001 0.527 ramosum (NT5006) Closthridiulm 0.680 1.000 0.040 0.202 1.000 0.089 0.997 1.000 0.853 saccharolyticum (NT5037) Collinsella 0.312 1.000 1.000 0.810 0.049 0.705 1.000 1.000 0.249 aerofaciens (NT5073) Coprococcus 0.001 0.431 0.025 0.753 0.430 0.854 0.052 0.145 1.000 comes (NT5048) Dorea 0.011 1.000 0.197 1.000 0.104 1.000 0.139 1.000 0.264 formicigenerans (NT5076) Eggerthella 0.659 1.000 1.000 0.360 1.000 0.389 1.000 1.000 0.887 lenta (NT5024) Escherichia 1.000 1.000 0.826 1.000 1.000 0.734 1.000 1.000 <0.001 coli ED1a (NT5078) Escherichia 1.000 0.652 0.847 1.000 0.613 1.000 1.000 1.000 <0.001 coli IAI1 (NT5077) Eubacterium 0.942 1.000 0.998 0.028 1.000 0.031 1.000 0.666 0.804 eligens (NT5075) Eubacterium 0.003 <0.001 0.589 0.051 0.024 1.000 0.072 1.000 0.931 rectale (NT5009) Fusobacterium 1.000 0.633 0.016 0.721 1.000 0.239 1.000 1.000 0.910 nucleatum (NT5025) Lactobacillus 0.939 1.000 1.000 1.000 0.353 0.692 0.913 0.331 1.000 paracasei (NT5042) Odoribacter 0.556 1.000 0.525 0.062 0.355 0.005 0.182 1.000 0.518 splanchnicus (NT5081) Parabacteroides 0.575 1.000 0.054 0.003 0.496 0.001 <0.001 0.004 0.983 distasonis (NT5074) Parabacteroides 0.441 0.984 0.024 0.004 1.000 0.009 0.137 0.156 1.000 merdae (NT5071) Prevotella 0.250 0.956 0.827 0.032 0.404 0.264 0.008 1.000 1.000 copri (NT5019) Roseburia 0.427 0.375 1.000 0.061 <0.001 0.806 1.000 1.000 0.382 hominis (NT5079) Roseburia 0.001 0.021 0.904 1.000 0.016 1.000 0.021 0.477 1.000 intestinalis (NT5011) Ruminococcus 1.000 0.683 0.248 1.000 1.000 0.838 1.000 0.020 0.730 bromii (NT5045) Ruminococcus 0.029 0.157 0.281 0.973 1.000 0.570 0.966 1.000 0.422 gnavus (NT5046) Ruminococcus 0.092 1.000 0.027 1.000 0.878 1.000 1.000 1.000 0.877 torques (NT5047) Streptococcus 0.006 1.000 0.623 1.000 0.792 0.628 1.000 0.076 0.837 parasanguinis (NT5072) Streptococcus <0.001 0.610 0.553 0.834 0.317 0.765 1.000 0.423 1.000 salivarius (NT5038) Veillonella 0.698 1.000 0.837 0.628 0.003 0.022 1.000 1.000 0.447 parvula (NT5017) Thyroxine Folic Mometasone Azacytidine- (L) Gemcitabine acid Etretinate Paclitaxel Phenindione furoate 5 Luteolin prestwick_ 403 1266 627 1409 155 538 572 866 870 ID n_hit 3 2 2 1 1 1 1 1 1 Akkermansia 1.000 1.000 0.801 1.000 1.000 0.692 1.000 1.000 0.149 muciniphila (NT5021) Bacteroides 1.000 1.000 1.000 0.975 1.000 0.910 1.000 1.000 0.538 caccae (NT5050) Bacteroides 1.000 1.000 0.952 1.000 0.924 0.826 1.000 0.795 0.443 fragilis (ET) (NT5033) Bacteroides 1.000 1.000 0.892 0.978 0.528 0.353 1.000 0.800 0.448 fragilis (NT) (NT5003) Bacteroides 1.000 1.000 1.000 1.000 1.000 0.973 1.000 1.000 1.000 ovatus (NT5054) Bacteroides 1.000 1.000 1.000 1.000 1.000 0.783 1.000 1.000 1.000 thetaiotaomicron (NT5004) Bacteroides 0.806 1.000 1.000 0.657 0.793 0.938 1.000 1.000 0.266 uniformis (NT5002) Bacteroides 1.000 1.000 1.000 0.886 0.699 0.908 1.000 0.998 1.000 vulgatus (NT5001) Bacteroides 1.000 1.000 0.939 1.000 1.000 0.211 1.000 1.000 1.000 xylanisolvens (NT5064) Bifidobacterium 1.000 1.000 0.674 0.875 1.000 0.354 1.000 1.000 0.886 adolescentis (NT5022) Bifidobacterium 1.000 1.000 0.990 1.000 1.000 0.583 1.000 1.000 1.000 longum (NT5028) Bilophila 0.905 1.000 1.000 1.000 1.000 1.000 1.000 0.746 0.383 wadsworthia (NT5036) Blautia 1.000 1.000 1.000 1.000 1.000 0.400 1.000 1.000 1.000 obeum (NT5069) Clostridium NA 1.000 1.000 1.000 0.900 0.943 1.000 1.000 0.176 bolteae (NT5026) Clostridium 0.064 1.000 1.000 0.249 0.668 0.526 1.000 1.000 0.023 difficile (NT5083) Clostridium 1.000 <0.001 0.989 <0.001 <0.001 0.009 0.958 <0.001 0.018 perfringens (NT5032) Clostridium 1.000 1.000 <0.001 1.000 0.897 0.725 0.991 0.525 0.586 ramosum (NT5006) Closthridiulm 0.001 1.000 0.158 1.000 0.894 0.801 1.000 0.991 0.025 saccharolyticum (NT5037) Collinsella 1.000 1.000 1.000 1.000 1.000 0.051 0.108 1.000 1.000 aerofaciens (NT5073) Coprococcus 0.064 0.645 1.000 0.922 1.000 0.967 0.903 1.000 NA comes (NT5048) Dorea 1.000 1.000 0.882 1.000 1.000 0.838 1.000 1.000 0.311 formicigenerans (NT5076) Eggerthella 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 lenta (NT5024) Escherichia 1.000 1.000 1.000 1.000 1.000 0.763 1.000 0.180 1.000 coli ED1a (NT5078) Escherichia 0.997 1.000 0.733 0.810 0.968 0.581 1.000 0.384 0.515 coli IAI1 (NT5077) Eubacterium 0.002 1.000 1.000 0.454 1.000 0.935 1.000 1.000 1.000 eligens (NT5075) Eubacterium 0.043 1.000 0.001 0.225 0.851 1.000 1.000 1.000 0.340 rectale (NT5009) Fusobacterium 1.000 1.000 0.684 1.000 1.000 0.971 1.000 0.778 0.001 nucleatum (NT5025) Lactobacillus 0.009 1.000 1.000 1.000 0.934 0.995 1.000 1.000 1.000 paracasei (NT5042) Odoribacter 1.000 1.000 1.000 1.000 1.000 0.665 1.000 0.657 0.612 splanchnicus (NT5081) Parabacteroides 1.000 0.296 0.824 1.000 0.909 1.000 0.380 1.000 1.000 distasonis (NT5074) Parabacteroides 1.000 1.000 1.000 0.721 0.220 0.826 0.689 1.000 0.330 merdae (NT5071) Prevotella 1.000 <0.001 1.000 0.293 0.950 1.000 1.000 0.042 0.562 copri (NT5019) Roseburia 0.262 1.000 1.000 0.998 1.000 0.490 1.000 1.000 1.000 hominis (NT5079) Roseburia 0.183 1.000 1.000 0.226 0.622 0.902 1.000 0.514 0.430 intestinalis (NT5011) Ruminococcus 1.000 1.000 1.000 1.000 1.000 1.000 0.379 1.000 1.000 bromii (NT5045) Ruminococcus 1.000 0.046 0.424 0.923 0.955 0.918 1.000 0.713 0.033 gnavus (NT5046) Ruminococcus 1.000 1.000 0.823 0.263 1.000 0.995 1.000 1.000 0.061 torques (NT5047) Streptococcus 1.000 0.896 1.000 0.921 0.872 0.691 1.000 0.870 0.077 parasanguinis (NT5072) Streptococcus 1.000 1.000 0.981 1.000 0.754 1.000 0.734 0.988 0.388 salivarius (NT5038) Veillonella 1.000 1.000 0.880 1.000 1.000 0.061 <0.001 1.000 1.000 parvula (NT5017)

[0225] Table 6 shows narrow-spectrum compounds effective against an enterotoxigenic strain of Bacteroides fragilis. Shown are the adjusted p-values of the compound-induced inhibition on the growth of 40 different bacterial species as indicated. Note: A compound is defined as narrow-spectrum if it inhibits the growth of less than 9 species of 40 different bacterial species tested.

TABLE-US-00006 Metixene Protriptyline Prenylamine hydrochloride hydrochloride Toltrazuril Acarbose Ethacrynic acid Tolnaftate Meclofenamic acid lactate prestwick_ID 491 930 1195 1174 259 70 206 560 n_hit 9 8 7 6 6 5 5 5 Akkermansia 0.197 0.039 1.000 <0.001 1.000 0.832 1.000 1.000 muciniphila (NT5021) Bacteroides 0.006 0.005 0.293 1.000 1.000 0.678 0.113 1.000 caccae (NT5050) Bacteroides 0.007 0.009 0.002 <0.001 0.001 0.009 0.001 0.001 fragilis (ET) (NT5033) Bacteroides 0.028 0.003 0.074 0.018 0.001 0.040 0.009 0.253 fragilis (NT) (NT5003) Bacteroides <0.001 <0.001 0.076 0.792 0.349 0.095 0.006 1.000 ovatus (N15054) Bacteroides <0.001 0.001 0.435 0.020 0.063 0.036 0.010 0.090 thetaiotaomicron (NT5004) Bacteroides <0.001 <0.001 0.295 0.001 <0.001 0.024 0.016 0.027 uniformis (NT5002) Bacteroides <0.001 <0.001 0.033 0.001 <0.001 0.034 0.072 <0.001 vulgatus (NT5001) Bacteroides 0.200 0.796 1.000 <0.001 0.617 1.000 0.357 1.000 xylanisolvens (NT5064) Bifidobacterium 1.000 0.928 1.000 1.000 1.000 0.854 0.659 1.000 adolescentis (NT5022) Bifidobacterium 1.000 1.000 1.000 1.000 1.000 0.883 0.995 1.000 longum (NT5028) Bilophila 1.000 0.832 0.812 0.713 0.887 0.638 0.951 0.251 wadsworthia (NT5036) Blautia obeum 0.792 1.000 0.342 1.000 0.792 0.501 0.955 0.860 (NT5069) Clostridium 0.569 1.000 0.910 1.000 0.006 0.897 1.000 0.442 bolteae (NT5026) Clostridium difficile 1.000 0.757 1.000 1.000 1.000 1.000 0.111 0.773 (NT5083) Clostridium 0.337 1.000 0.007 <0.001 0.002 0.007 0.266 0.796 perfringens (NT5032) Clostridium 1.000 1.000 1.000 1.000 0.032 1.000 0.653 1.000 ramosum (NT5006) Clostridium 1.000 1.000 1.000 1.000 0.040 0.072 0.201 0.651 saccharolyticum (NT5037) Collinsella 1.000 0.745 1.000 1.000 1.000 0.451 1.000 1.000 aerofaciens (NT5073) Coprococcus 0.084 0.964 0.040 0.431 0.025 0.464 0.394 0.124 comes (NT5048) Dorea 1.000 0.664 0.792 1.000 0.197 0.161 0.963 0.736 formicigenerans (NT5076) Eggerthella lenta 0.931 0.564 1.000 1.000 1.000 1.000 1.000 1.000 (NT5024) Escherichia coli 1.000 0.302 1.000 1.000 0.826 1.000 0.398 0.438 ED1a (NT5078) Escherichia coli 1.000 0.725 0.502 0.652 0.847 0.558 0.394 0.126 IAI1 (NT5077) Eubacterium 0.023 0.477 1.000 1.000 0.998 1.000 1.000 0.001 eligens (NT5075) Eubacterium 0.047 0.148 0.030 <0.001 0.589 0.173 1.000 0.006 rectale (NT5009) Fusobacterium 1.000 1.000 1.000 0.633 0.016 0.393 0.015 0.550 nucleatum (NT5025) Lactobacillus 1.000 1.000 0.075 1.000 1.000 0.003 0.638 0.034 paracasei (NT5042) Odoribacter 0.001 0.231 0.004 1.000 0.525 0.934 0.020 1.000 splanchnicus (NT5081) Parabacteroides <0.001 0.166 <0.001 1.000 0.054 0.102 0.003 0.014 distasonis (NT5074) Parabacteroides <0.001 0.007 <0.001 0.984 0.024 0.013 0.001 0.011 merdae (NT5071) Prevotella copri 0.018 0.285 0.001 0.956 0.827 0.831 0.316 0.055 (NT5019) Roseburia hominis 1.000 1.000 1.000 0.375 1.000 0.976 1.000 0.768 (NT5079) Roseburia 0.899 1.000 0.007 0.021 0.904 0.008 1.000 0.005 intestinalis (NT5011) Ruminococcus 0.675 1.000 0.126 0.683 0.248 0.531 0.682 1.000 bromii (NT5045) Ruminococcus 0.526 0.971 0.031 0.157 0.281 0.665 0.256 0.029 gnavus (NT5046) Ruminococcus 1.000 1.000 0.182 1.000 0.027 <0.001 0.971 0.699 torques (NT5047) Streptococcus 1.000 0.995 0.438 1.000 0.623 0.236 0.043 0.075 parasanguims (NT5072) Streptococcus 1.000 0.450 0.781 0.610 0.553 0.711 0.337 0.299 salivarius (NT5038) Veillonella parvula 1.000 0.925 1.000 1.000 0.837 0.914 1.000 1.000 (NT5017)

[0226] Table 7 shows compounds effective against a broad spectrum of bacteria. Shown are the adjusted p-values of the compound-induced inhibition on the growth of 40 different bacterial species as indicated. Note: A compound is defined as broad-spectrum compound if it inhibits the growth of 10 species of 40 different bacterial species tested.

TABLE-US-00007 Estradiol Clemizole Bepridil Methiothepin Amiodarone Ethopropazine Diacerein Tiratricol Troglitazone Dicumarol Anthralin Astemizole Loratadine Valerate Aprepitant hydrochloride hydrochloride maleate hydrochloride hydrochloride prestwick_ID 1167 202 1467 785 1224 136 1432 1473 1600 227 368 375 409 840 n_hit 33 14 13 10 11 10 15 11 10 15 14 10 10 11 Akkermansia 0.108 1.000 1.000 0.895 0.944 0.001 1.000 1.000 0.518 0.019 0.538 0.064 1.000 0.049 muciniphila (NT5021) Bacteroides <0.001 0.118 0.020 0.043 1.000 1.000 0.026 0.004 0.004 0.634 0.145 0.097 0.371 <0.001 caccae (NT5050) Bacteroides <0.001 0.015 0.231 0.013 0.051 0.565 0.002 0.007 <0.001 0.437 0.078 0.053 0.348 <0.001 fragilis (ET) (NT5033) Bacteroides <0.001 0.108 0.004 0.002 0.002 0.504 0.110 <0.001 <0.001 0.612 0.294 0.091 0.044 <0.001 fragilis (NT) (NT5003) Bacteroides <0.001 0.003 0.263 0.002 1.000 1.000 0.001 1.000 0.144 0.855 0.128 0.167 1.000 <0.001 ovatus (NT5054) Bacteroides <0.001 0.017 0.107 0.042 0.567 0.649 0.025 0.445 0.464 0.363 0.007 0.024 0.322 <0.001 thetaiotaomicron (NT5004) Bacteroides <0.001 0.507 0.266 0.002 0.238 0.214 0.151 0.032 0.024 0.171 0.081 0.282 0.717 <0.001 uniformis (NT5002) Bacteroides <0.001 0.526 0.025 0.137 0.006 0.323 0.012 <0.001 <0.001 0.012 0.008 0.084 0.003 <0.001 vulgatus (NT5001) Bacteroides <0.001 1.000 0.044 0.025 1.000 1.000 0.030 0.290 0.046 1.000 1.000 1.000 1.000 <0.001 xylanisolvens (NT5064) Bifidobacterium 1.000 <0.001 0.881 1.000 1.000 1.000 0.001 0.308 0.319 0.010 0.098 0.582 0.151 1.000 adolescentis (NT5022) Bifidobacterium 0.898 0.197 0.274 0.969 1.000 0.993 0.001 0.024 0.008 0.059 0.051 0.001 <0.001 1.000 longum (NT5028) Bilophila 0.478 0.921 1.000 1.000 1.000 1.000 1.000 0.369 0.511 0.266 0.981 0.895 0.068 0.559 wadsworthia (NT5036) Blautia obeum <0.001 <0.001 <0.001 0.832 <0.001 <0.001 <0.001 0.692 0.001 <0.001 <0.001 <0.001 <0.001 0.646 (NT5069) Clostridium <0.001 0.049 0.334 1.000 0.007 0.980 0.157 0.661 0.167 0.050 0.001 0.103 0.258 0.768 bolteae (NT5026) Clostridium <0.001 <0.001 <0.001 0.007 0.010 1.000 0.020 0.071 0.109 0.189 0.022 0.073 0.189 0.441 difficile (NT5083) Clostridium <0.001 <0.001 <0.001 0.010 0.001 0.010 <0.001 0.001 0.263 <0.001 0.005 0.101 0.095 1.000 perfringens (NT5032) Clostridium <0.001 0.001 <0.001 0.175 0.121 0.896 0.463 1.000 0.883 <0.001 0.236 0.354 0.930 0.949 ramosum (NT5006) Clostridium <0.001 <0.001 0.061 0.026 0.507 0.148 0.311 1.000 0.974 <0.001 0.334 0.211 1.000 1.000 saccharolyticum (NT5037) Collinsella <0.001 0.020 0.543 0.808 0.152 0.608 <0.001 0.963 0.097 <0.001 0.001 0.003 0.001 0.449 aerofaciens (NT5073) Coprococcus <0.001 0.009 0.004 1.000 <0.001 0.005 <0.001 <0.001 0.018 <0.001 0.089 0.138 <0.001 0.091 comes (NT5048) Dorea <0.001 0.046 0.814 0.813 1.000 <0.001 <0.001 0.403 0.040 0.001 0.009 0.037 0.047 1.000 formicigenerans (NT5076) Eggerthella lenta <0.001 0.006 0.153 0.934 <0.001 0.127 1.000 1.000 1.000 0.796 0.459 0.738 0.871 0.506 (NT5024) Escherichia coli 0.497 1.000 1.000 0.697 1.000 1.000 0.414 0.668 0.799 0.564 0.967 1.000 1.000 0.963 ED1a (NT5078) Escherichia coli 1.000 1.000 1.000 0.871 1.000 1.000 0.740 0.719 0.749 0.215 0.117 1.000 0.996 0.883 IAI1 (NT5077) Eubacterium <0.001 0.292 0.265 1.000 0.060 0.002 <0.001 1.000 1.000 <0.001 <0.001 0.001 <0.001 0.874 eligens (NT5075) Eubacterium <0.001 1.000 0.003 1.000 0.241 0.001 0.001 0.002 0.558 <0.001 0.002 0.002 0.001 0.051 rectale (NT5009) Fusobacterium <0.001 0.342 1.000 0.019 1.000 1.000 1.000 0.849 1.000 0.015 0.527 1.000 1.000 1.000 nucleatum (NT5025) Lactobacillus 0.829 0.595 0.992 1.000 0.483 0.858 0.009 0.013 0.659 0.002 0.019 0.796 0.078 1.000 paracasei (NT5042) Odoribacter <0.001 0.002 0.069 0.001 <0.001 0.878 0.167 0.659 0.018 0.557 0.015 0.004 1.000 <0.001 splanchnicus (NT5081) Parabacteroides <0.001 0.001 0.012 <0.001 0.879 0.107 0.016 0.255 0.002 0.104 0.002 0.004 0.013 <0.001 distasonis (NT5074) Parabacteroides <0.001 0.003 0.812 0.007 <0.001 0.635 0.013 0.001 0.251 0.441 0.007 0.472 0.013 <0.001 merdae (NT5071) Prevotella copri <0.001 0.132 0.006 0.020 0.053 0.008 0.121 0.007 0.016 0.143 0.033 <0.001 0.009 0.024 (NT5019) Roseburia <0.001 1.000 1.000 1.000 1.000 0.010 0.492 1.000 0.001 <0.001 0.118 <0.001 <0.001 1.000 hominis (NT5079) Roseburia <0.001 0.008 0.001 1.000 1.000 0.001 <0.001 0.011 <0.001 <0.001 0.002 0.001 <0.001 0.776 intestinalis (NT5011) Ruminococcus <0.001 0.031 0.002 0.146 0.354 1.000 <0.001 <0.001 0.210 <0.001 0.821 0.163 0.072 0.823 bromii (NT5045) Ruminococcus <0.001 0.036 0.004 0.017 0.050 0.001 0.029 0.028 0.236 <0.001 0.005 0.247 0.029 0.547 gnavus (NT5046) Ruminococcus <0.001 1.000 0.001 1.000 1.000 0.001 0.001 0.001 0.130 <0.001 0.012 1.000 0.117 1.000 torques (NT5047) Streptococcus <0.001 0.012 0.119 <0.001 1.000 0.913 0.055 1.000 0.776 0.757 0.050 1.000 0.326 0.741 parasanguinis (NT5072) Streptococcus <0.001 <0.001 <0.001 <0.001 1.000 1.000 0.363 0.067 1.000 0.123 0.009 0.188 0.880 0.460 salivarius (NT5038) Veillonella <0.001 0.826 1.000 0.968 <0.001 1.000 0.477 1.000 0.009 0.677 0.345 0.223 1.000 0.920 parvula (NT5017)