The project leading to this application has received funding from the European Research Council (ERC) under the European Union's Horizon 2020 research and innovation programme (grant agreement No 866028).

TECHNICAL FIELD OF THE INVENTION

The invention relates to methods for reducing the level of xenobiotic in an environment by contacting the environment with a composition comprising one or more bacterial strains. The invention also relates to bacterial strains for use in method of reducing the level of xenobiotic in a subject. The invention also relates to compositions comprising one or more bacterial strains.

BACKGROUND OF THE INVENTION

Xenobiotics are compounds foreign to the human body, such as environmental contaminants, food contact materials, pesticides, drugs, natural toxins and processing contaminants. It is estimated that an individual human will be exposed to up to 100,000 different xenobiotic compounds throughout their lifetime. This includes man-made chemicals resistant to normal environmental degradation processes, many of which are widely used in industrial, agricultural and domestic applications. Their resistance to degradation results in their accumulation in the environment, in humans, and in animals. Environmental accumulation of xenobiotics is particularly evident in water systems, where xenobiotic levels frequently reach levels which are undesirable and even unsafe; and water treatment facilities are poorly equipped to deal with these types of pollutants.

Once ingested, degradation-resistant xenobiotics can persist for long periods of time (a process known as bioaccumulation). When xenobiotics reach a toxic concentration within the body, they can have a variety of negative physiological effects.

Per- and polyfluoroalkyl substances (PFAS) are amongst the most well-known xenobiotics and are colloquially known as "forever chemicals" because of their persistence in the environment. Following their development in the 1940s and 1950s, these synthetic chemicals were widely used in industrial manufacturing and they quickly became ubiquitous in the environment. Bioaccumulation of PFAS, and toxic effects associated therewith, has been demonstrated in animals and humans.

There is an urgent and unmet need for methods of reducing the level of these types of xenobiotics in the environment, in humans, and in animals. SUMMARY OF THE INVENTION

The inventors have discovered multiple bacterial strains that can bioaccumulate and/or biotransform xenobiotics. Advantageously, these bacterial strains are commonly found in the human microbiome and so do not pose an immediate safety risk to humans and animals. By bioaccumulating and/or biotransforming xenobiotics, the bacterial strains of the invention reduce the bioavailability of xenobiotics, and thereby preventing or reducing their accumulation in the environment, leading to improved environmental quality and reduced bioaccumulation in animals and humans (e.g. via ingestion or absorption). When administered to a subject, bacterial strains of the invention also help reduce the level of xenobiotics which are already present within the subject; and also help prevent future bioaccumulation of xenobiotics.

The invention provides a method for reducing the level of a xenobiotic in an environment, the method comprising contacting the environment with a composition comprising one or more bacterial strains selected from Bacteroides, Collinsella, Coprococcus, Eubacterium, Odoribacter, Parabacteroides, and Roseburia. The invention also provides a method for reducing the level of a xenobiotic in an environment, the method comprising contacting the environment with a composition comprising one or more bacterial strains selected from Bacteroides, Collinsella, Coprococcus, Eubacterium, Odoribacter, Parabacteroides, Roseburia, Escherichia, Phocaeicola, Prevotella, Butyrivibrio, Lacrimispora, Clostridium, Fusobacterium, Agathobacter, Dorea, and Streptococcus.

The invention also provides use of a composition comprising one or more bacterial strains selected from Bacteroides, Collinsella, Coprococcus, Eubacterium, Odoribacter, Parabacteroides, Roseburia, Escherichia, Phocaeicola, Prevotella, Butyrivibrio, Lacrimispora, Clostridium, Fusobacterium, Agathobacter, Dorea, and Streptococcus in a method of reducing the level of a xenobiotic in an environment.

In some embodiments, the environment is an aqueous environment, optionally wherein the aqueous environment is drinking water or wastewater.

In some embodiments, the method comprises detecting the presence and/or measuring the abundance of the one or more bacterial strains in the environment prior to contacting the environment with the composition.

In some embodiments, the method comprises: (i) contacting the environment with one or more bacterial strains; and then (ii) removing the one or more bacterial strains to provide a treated environment. The invention also provides a composition for use in a method of reducing the level of a xenobiotic in a subject, wherein the composition comprises one or more bacterial strains selected from Bacteroides, Collinsella, Coprococcus, Eubacterium, Odoribacter, Parabacteroides, and Roseburia. The invention also provides a composition for use in a method of reducing the level of a xenobiotic in a subject, wherein the composition comprises one or more bacterial strains selected from Bacteroides, Collinsella, Coprococcus, Eubacterium, Odoribacter, Parabacteroides, Roseburia, Escherichia, Phocaeicola, Prevotella, Butyrivibrio, Lacrimispora, Clostridium, Fusobacterium, Agathobacter, Dorea, and Streptococcus.

In some embodiments, the method comprises preventing or treating xenobiotic poisoning in the subject.

In some embodiments, the subject has ingested, is suspected to have ingested, or is at risk of ingesting the xenobiotic.

In some embodiments, the method comprises detecting the presence and/or measuring the abundance of the one or more bacterial strains in the subject prior to administration of the composition to the subject.

In some embodiments, the subject is a human. In some embodiments, the subject is an animal, optionally wherein the animal is a cow, sheep, pig, poultry, cat or dog.

The invention also provides a composition for reducing the level of a xenobiotic, the composition comprising one or more bacterial strains selected from Bacteroides, Collinsella, Coprococcus, Eubacterium, Odoribacter, Parabacteroides, and Roseburia. The invention also provides a composition for reducing the level of a xenobiotic, the composition comprising one or more bacterial strains selected from Bacteroides, Collinsella, Coprococcus, Eubacterium, Odoribacter, Parabacteroides, Roseburia, Escherichia, Phocaeicola, Prevotella, Butyrivibrio, Lacrimispora, Clostridium, Fusobacterium, Agathobacter, Dorea, and Streptococcus.

In some embodiments, the one or more bacterial strains are selected from Bacteroides uniformis, Bacteroides caccae, Bacteroides clarus, Bacteroides dorei, Bacteroides stercoris, Bacteroides thetaiotaomicron, Collinsella aerofaciens, Coprococcus comes, Eubacterium rectale, Odoribacter splanchnicus, Parabacteroides distasonis, Parabacteroides merdae, and Roseburia intestinalis. In some embodiments, the one or more bacterial strains are selected from Bacteroides uniformis, Bacteroides caccae, Bacteroides clarus, Bacteroides dorei, Bacteroides stercoris, Bacteroides thetaiotaomicron, Collinsella aerofaciens, Coprococcus comes, Eubacterium rectale, Odoribacter splanchnicus, Parabacteroides distasonis, Parabacteroides merdae, Roseburia intestinalis, Escherichia coll, Phocaeicola coprocola, Prevotella copri, Bacteroides eggerthii, Prevotella melaninogenica, Bacteroides fragilis, Bacteroides xylanisolvens, Butyrivibrio crossotus, Bacteroides coprocola, Roseburia hominis, Lacrimispora saccharolytica, Clostridium scindens, Fusobacterium nucleatum subsp. Nucleatum, Clostridium difficile, Phocaeicola vulgatus, Agathobacter rectalis, Roseburia inulinivorans, Dorea formicigenerans, Streptococcus salivarius, Fusobacterium nucleatum subsp. Animalis, Fusobacterium nucleatum subsp. Vincenti!, Clostridium hylemonae, and Clostridium sporogenes.

In some embodiments, the composition comprises Bacteroides, optionally wherein the composition comprises Bacteroides uniformis.

In some embodiments, one or more of the bacterial strains is genetically modified. In some embodiments, one or more of the bacterial strains comprises a genetic modification resulting in a decrease or elimination of xenobiotic efflux from the one or more bacterial strains. In some embodiments, the genetic modification comprises deletion or inactivation of at least one gene required for activity of an efflux pump.

In some embodiments, the xenobiotic comprises one or more of a PFA, a bisphenol, and a pesticide.

In some embodiments, the xenobiotic comprises one or more of PFNA, PFOA, PFDeA, bisphenol AF, Boscalid, Propiconazole, Pyrimethanil, tributyl-PC , and triphenyl-PC .

In some embodiments, the xenobiotic comprises a PFA, optionally wherein the PFA comprises PFOA, PFNA, and/or PFDeA.

In some embodiments, the composition comprises Bacteroides and the xenobiotic comprises a PFA.

In some embodiments, the composition comprises Bacteroides uniformis and the PFA comprises PFNA, PFOA and/or PFDeA.

In some embodiments, the composition is formulated for delivery to the intestine, optionally wherein the composition is formulated for oral delivery, nasal delivery and/or rectal delivery.

In some embodiments, the composition further comprises one or more prebiotics for promoting growth of the one or more bacterial strains.

In some embodiments, the one or more prebiotics are selected from arabinoxylan, xylose, fiber dextran, corn fiber, polydextrose, lactose, N-acetyl-lactosamine, glucose, galactose, fructose, rhamnose, mannose, uronic acids, arabinose, fructose, fucose, lactose, galactose, glucose, mannose, D-xylose, xylitol, ribose, xylobiose, sucrose, maltose, lactose, lactulose, trehalose, cellobiose, xylooligosaccharide, fructooligosaccharide, galactooligosaccharide, lactosucrose, and soybean oligosaccharides. In some embodiments, the one or more bacterial strains are lyophilised. In some embodiments, the composition further comprises a carrier, excipient, and/or diluent. In some embodiments, the composition comprises a gastro-resistant coating. In some embodiments, the composition is a timerelease formulation.

The invention also provides a dietary supplement comprising the composition of the invention.

BRIEF DESCRIPTION OF THE FIGURES

Figure 1. Results of the strain bioaccumulation and biotransformation screen.

Figure 2. Bioaccumulation of PFNA and biotransformation of propiconazole by Bacteroides dorei and Bacteroides uniformis.

Figure 3. PFA accumulation by B. uniformis (recovery of PFOA and PFNA in the cell pellet).

Figure 4. PFA accumulation by B. uniformis at a range of different concentrations. Percentage of compound accumulated stays constant for different concentrations.

Figure 5. PFA accumulation by B. uniformis for PFAs with varying chain length. Accumulation increases with increasing chain length.

Figure 6. PFNA accumulation by B. uniformis over the course of 7 days.

Figure 7. Abundant gut bacterial species bioaccumulate and biotransform common chemical pollutants and accumulate and tolerate PFAS over a broad concentration range, a. Specificity of human gut bacteria to sequester (bioaccumulate/biotransform) chemical pollutants as identified using mass-spectrometry. Links between bacterial species and pollutant denotes >20 % depletion and p<0.05 (student t-test). The link thickness is proportional to the median depletion from 6 replicates (3 biological, 2 technical). b,c. Example cases showing bioaccumulation of the PFAS-family compound perfluorononanoic acid (PFNA) (b) and biodegradation of the fungicide propiconazole (c) by B. uniformis. n = 6 (3 biological, 2 technical), d. Bioaccumulation of PFAS compounds with varying chain length by B. uniformis (initial concentration for all compounds = 20 pM). n = 3 technical replicates, e. Kinetics of PFNA depletion during B. uniformis growth starting with low cell density (OD600 = 0.05; initial PFNA concentration = 20 pM). n = 3 biological replicates, f. Kinetics of PFNA depletion by B. uniformis when starting with high cell density (OD600 = 4). Bioaccumulation of ca. 50 % PFNA happens within the time frame of sample collection (ca. 5 min), n = 2 biological replicates, g. PFNA is bioaccumulated by B. uniformis grown at a range of initial concentrations from 0.01 to 100 pM. n = 4 technical replicates, h. Growth sensitivity of gut bacteria to PFAS is independent of bioaccumulation (n = 3 technical replicates); *Bioaccumulating bacteria. Figure 8. Genetic and morphological data support intra-cellular accumulation of PFAS. E. coli efflux pump mutants show increased PFAS bioaccumulation capacity and PFAS bioaccumulating bacteria show distinct morphological features in transmission electron microscopy (TEM). a. PFNA bioaccumulation by live, dead (heat-inactivated) and lysed (heat-inactivated + freeze-thawed + sonicated) B. uniformis, O. splanchnicus and E. coli cultures (OD600 = 3.75) in PBS buffer, n = 3 technical replicates, b. AcrAB-TolC efflux pump schematic. The resting state of the pump is pictured in the left of the image. When the pump encounters a xenobiotic, it changes conformation to export the xenobiotic out of the cell (right), c. Accumulation of PFDeA and PFNA by wild-type E. coli strains and corresponding efflux mutants (E. coli BW25113 tolC, E. coli C43 (DE3) acrAB-tolC). n = 3 technical replicates, d. Efflux mutants show increased PFNA and PFOA sensitivity, n = 3 technical replicates, e- h. TEM of B. uniformis cells grown in mGAM + DMSO (e), 50 pM PFNA (f), 250 pM PFNA (g), or 125 pM PFDeA (h) for 24 h. i,j. TEM of E. coli BW25113 wild-type cells grown in mGAM + DMSO (i) or 250 pM PFNA (j) for 24 h. k,l. TEM of E. coli BW25113 Ato/C cells grown in mGAM + DMSO (k) or 250 pM PFNA (I) for 24 h. m,n. TEM of O. splanchnicus cells grown in mGAM + DMSO (m) or 250 pM PFNA (n) for 24 h. Arrows are pointing at changes in nucleoid appearance in PFAS treated bacteria.

Figure 9. PFAS tolerance and bioaccumulation following adaptive laboratory evolution, a. Five gut bacterial species were evolved through serial passaging in growth medium containing one of four PFAS compounds (500 pM Perfluoroheptanoic acid (PFHpA), 500 pM PFOA, 250 pM PFNA, 125 pM PFDeA) over 20 days. b. Improved growth of adapted B. uniformis population in presence of 125 pM PFDeA and 250 pM PFNA. c., d. Adapted populations retain PFDeA (c) and PFNA (d) bioaccumulation capability, n = 4 independent populations per compound.

Figure 10. Increased PFNA excretion in mouse faeces and Gl tract after PFNA exposure, a. Experimental setup, b., c. Mice colonised with a community of 20 human gut bacterial strains (Com20) show higher PFNA excretion after exposure compared to germfree (GF) controls.

Figure 11. Method setup and results from the community-screen, a. Method workflow for the artificial community experiment, b. Method workflow for the single strain experiment, c. 18 out of 42 tested pollutants were sequestered by at least one synthetic gut bacterial community. Each community consisted of 10 bacteria (community 1: Bacteroides caccae, Bacteroides dorei, Bacteroides thetaiomicron, Bacteroides uniformis, Bacteroides vulgatus, Colinsella aerofaciens, Coprococcus comes, Eubacterium rectale, Parabacteroides merdae, Roseburia intestinalis; community 2: Akkermansia muciniphila, Bacteroides clarus, Bacteroides stercoris, Clostridium difficile, Eggerthella lenta, Eubacterium eligens, Fusobacterium nucleatum subsp. animalis, Odoribacter splanchnicus, Parabacteroides distastonis, Ruminococcus bromii). Coloured squares denote sequestration, i.e., >25 % reduction from the bacteria-free supernatant. PFAS are marked in blue, n = 6 (3 biological and 2 technical replicates).

Figure 12. Within strain variation for PFNA and PFOA bioaccumulation capacity, a. Accumulation capacity of 8 different B. uniformis and 2 E. coli strains grown in mGAM with 20 uM PFOA or PFNA with a starting OD of 0.05 over 24 h. No strain differences were observed, b. Accumulation capacity of 8 different B. uniformis and 2 E. coli strains incubated in PBS (OD600 = 3.75) with 20 uM PFOA or PFNA for 4h. No strain differences were observed, n = 3 technical replicates.

Figure 13. Within strain variation for PFNA and PFOA bioaccumulation capacity. PFNA and PFOA accumulation by B. uniformis in PBS (OD600 = 3.75) containing different concentrations of PFNA ranging from 0.01 to 100 uM over the course of 4 h. n = 3 technical replicates.

Figure 14. Bioaccumulation of PFAS in live and dead bacterial biomass. PFDeA, PFNA and PFOA bioaccumulation by live, dead (heat-inactivated) and lysed (heat-inactivated + freeze-thawed + sonicated) B. uniformis, O. splanchnicus and E. coli cultures (OD600 = 3.75) in PBS buffer (n = 3 technical replicates).

Figure 15. Effect of ALE on the growth in presence of high PFAS concentrations. Growth of bacterial strains in presence of high PFAS concentrations over the course of the adaptive laboratory evolution, n = 4 evolved lineages per compound.

Figure 16. Evolved strains retain PFAS bioaccumulation capacity, a. Strains evolved in presence of PFDeA show unaltered PFDeA bioaccumulation capacity, b. Strains evolved in presence of PFNA show unaltered PFNA bioaccumulation capacity, n = 4 evolved lineages.

Figure 17. Results showing the proteins that are differentially expressed between B. uniformis treated with PFNA (20 uM) in comparison to DMSO. The green and the blue dots mark proteins with a Iog2 abundance ratio >1 or <-l (i.e., two-fold increase or decrease) and a multiple-testing corrected p-value of less than 0.05.

Figure 18. Results from the P. merdae transposon library screen showing reduced fitness for mutants in genes encoding for homologues of the three most upregulated proteins in B. uniformis, all of which are efflux pumps.

Figure 19. A. PFNA accumulation shows a bimodal distribution. B. Gram-positive and gram-negative strains show differences in PFNA accumulation. C. Bacterial phyla show differences in PFNA accumulation. D. Correlation for PFNA accumulation between growth (mGAM) and resting (PBS) assay. DETAILED DESCRIPTION OF THE INVENTION

The invention is based on the surprising discovery of bacterial strains that can reduce the level of a wide variety of xenobiotics. Advantageously, the bacterial strains of the invention are commonly found in the microbiota of humans and can therefore be used in a variety of settings without posing an immediate health risk to humans or animals. The inventors have discovered that the bacterial strains of the invention possess a highly desirable ability to bioaccumulate and/or biotransform xenobiotics. Surprisingly, the inventors found that several bacterial strains that are commonly found in the gut bioaccumulate the "forever chemical" PFA substances, including PFNA (perfluorononanoic acid) and PFOA (perfluorooctanoic acid). The inventors made the particularly surprising discovery that the gut bacterial strains of the invention bioaccumulated PFAS to a much higher degree than bacteria which had been isolated from PFAS contaminated sites (e.g. Pseudomonas sp.). Bacterial strains of the invention advantageously reduce the level of xenobiotics within the environment, thereby improving environmental quality and reducing risk of ingestion and/or absorption by subjects (e.g. humans or animals).

The invention provides a method for reducing the level of a xenobiotic within an environment, the method comprising contacting the environment with a composition comprising one or more bacterial strains selected from Bacteroides, Collinsella, Coprococcus, Eubacterium, Odoribacter, Parabacteroides, and Roseburia. The invention also provides a method for reducing the level of a xenobiotic within an environment, the method comprising contacting the environment with a composition comprising one or more bacterial strains selected from Bacteroides, Collinsella, Coprococcus, Eubacterium, Odoribacter, Parabacteroides, Roseburia, Escherichia, Phocaeicola, Prevotella, Butyrivibrio, Lacrimispora, Clostridium, Fusobacterium, Agathobacter, Dorea, and Streptococcus. As demonstrated herein, bacterial strains from each of these genera can bioaccumulate and/or biotransform xenobiotics, such as PFAS.

As used herein, "biotransformation" of a xenobiotic involves the enzymatic conversion of xenobiotic compounds to metabolites and "bioaccumulation" of a xenobiotic by the bacterial strains of the invention involves uptake and sequestration of the xenobiotic within the bacterial cell.

In some embodiments, reducing the level of the xenobiotic comprises bioaccumulation of the xenobiotic by the one or more bacterial strains. In some embodiments, reducing the level of the xenobiotic comprises biotransformation of the xenobiotic by the one or more bacterial strains.

Contacting the environment with the composition may be by any suitable means. In some embodiments, contacting the environment with the composition comprises applying the composition to the environment using one or more methods selected from spraying, fertigation, and/or injection into the environment.

The environment is typically an ex vivo environment. In some embodiments, the environment is an aqueous environment. In some embodiments, the aqueous environment is in a water treatment plant, e.g. a drinking water treatment plant and/or a wastewater treatment plant. In some embodiments, the aqueous environment is industrial wastewater. In some embodiments, the aqueous environment is drinking water or water intended for drinking.

Advantageously, the bacterial strains of the invention are common in the microbiota of humans, and so their presence e.g. in drinking water does not pose a significant health risk to humans or animals.

In some embodiments, the method for reducing the level of a xenobiotic within an environment is performed in situ.

In some embodiments, the method is a bioremediation method. In some embodiments, the method is a bioremediation method performed in situ.

In some embodiments, the environment is land contaminated with, or suspected of being contaminated with xenobiotics. In some embodiments, the land is agricultural land. In some embodiments, the land is residential or domestic land. In some embodiments, the land is forestry.

In some embodiments, the environment is within a bioreactor (e.g. growth media).

In some instances, it is advantageous to remove the bacterial strains following treatment of an environment, e.g. when bacterial cells have bioaccumulated xenobiotic from the environment. In some embodiments, the method comprises: (a) contacting an environment comprising a xenobiotic, or suspected of comprising a xenobiotic, with the composition to reduce the level of xenobiotic within said environment; and then (b) removing the one or more bacterial strains to provide a treated environment. For example, the environment may be filtered to remove bacterial strains. In some embodiments, the method comprises: (a) contacting an aqueous environment comprising a xenobiotic, or suspected of comprising a xenobiotic, with the composition to reduce the level of xenobiotic within said aqueous environment; and then (b) removing the one or more bacterial strains to provide a treated aqueous environment. For example, the aqueous environment may be filtered to remove bacterial strains. Suitable filter sizes may be easily determined by the user.

In some embodiments, the method comprises contacting an aqueous environment comprising a xenobiotic, or suspected of comprising a xenobiotic, with one or more bacterial strains of the invention, wherein the one or more bacterial strains are immobilised on a support. In this embodiment, the aqueous environment may flow over the support continuously or the support may be added to the aqueous environment for a predetermined period of time. Bacteria can be immobilised on a range of supports, e.g. glass or polymer beads, sand or gravel-type materials, polysaccharide-based matrices, and membranes.

The invention also provides a composition for use in a method of reducing the level of a xenobiotic in a subject, wherein the composition comprises one or more bacterial strains selected from Bacteroides, Collinsella, Coprococcus, Eubacterium, Odoribacter, Parabacteroides, and Roseburia. The invention also provides a method of reducing the level of a xenobiotic in a subject comprising administering to the subject a composition comprising one or more bacterial strains selected from Bacteroides, Collinsella, Coprococcus, Eubacterium, Odoribacter, Parabacteroides, and Roseburia. The invention also provides a composition for use in a method of preventing or treating xenobiotic poisoning in a subject, wherein the composition comprises one or more bacterial strains selected from Bacteroides, Collinsella, Coprococcus, Eubacterium, Odoribacter, Parabacteroides, and Roseburia. The invention also provides a method for preventing or treating xenobiotic poisoning in a subject comprising administering to the subject a composition comprising one or more bacterial strains selected from Bacteroides, Collinsella, Coprococcus, Eubacterium, Odoribacter, Parabacteroides, and Roseburia.

The invention also provides a composition for use in a method of reducing the level of a xenobiotic in a subject, wherein the composition comprises one or more bacterial strains selected from Bacteroides, Collinsella, Coprococcus, Eubacterium, Odoribacter, Parabacteroides, Roseburia, Escherichia, Phocaeicola, Prevotella, Butyrivibrio, Lacrimispora, Clostridium, Fusobacterium, Agathobacter, Dorea, and Streptococcus. The invention also provides a method of reducing the level of a xenobiotic in a subject comprising administering to the subject a composition comprising one or more bacterial strains selected from Bacteroides, Collinsella, Coprococcus, Eubacterium, Odoribacter, Parabacteroides, Roseburia, Escherichia, Phocaeicola, Prevotella, Butyrivibrio, Lacrimispora, Clostridium, Fusobacterium, Agathobacter, Dorea, and Streptococcus. The invention also provides a composition for use in a method of preventing or treating xenobiotic poisoning in a subject, wherein the composition comprises one or more bacterial strains selected from Bacteroides, Collinsella, Coprococcus, Eubacterium, Odoribacter, Parabacteroides, Roseburia, Escherichia, Phocaeicola, Prevotella, Butyrivibrio, Lacrimispora, Clostridium, Fusobacterium, Agathobacter, Dorea, and Streptococcus. The invention also provides a method for preventing or treating xenobiotic poisoning in a subject comprising administering to the subject a composition comprising one or more bacterial strains selected from Bacteroides, Collinsella, Coprococcus, Eubacterium, Odoribacter, Parabacteroides, Roseburia, Escherichia, Phocaeicola, Prevotella, Butyrivibrio, Lacrimispora, Clostridium, Fusobacterium, Agathobacter, Dorea, and Streptococcus. As noted above, the bacterial strains of the invention are highly prevalent in the human microbiome, indicating that they can be safely administered to humans and animals.

As used herein, 'xenobiotic poisoning' embraces any deleterious physiological effects caused by ingestion of xenobiotics. Xenobiotics may be ingested via a variety of routes, e.g. in drinking water and/or foodstuffs, and then absorbed from the intestine by active and/or passive mechanisms. Deleterious physiological effects associated with the absorption and/or bioaccumulation of xenobiotics by a subject include, but are not limited to, inflammation, increased oxidative stress and increased cellular toxicity. In some embodiments, reducing the level of xenobiotics in a subject comprises reducing the level of ingested xenobiotics.

In some embodiments, reducing the level of a xenobiotic in a subject comprises reducing the bioavailability of xenobiotics by bioaccumulation and/or biotransformation. Bioaccumulation results in ingested xenobiotics being sequestered within bacterial cells, thereby reducing the bioavailability of xenobiotic within the subject's body. Biotransformation results in metabolism of the xenobiotics, thereby reducing the bioavailability of xenobiotic within the subject's body. As used herein, the bioavailability of a xenobiotic is the proportion of xenobiotic that is available for absorption, typically from the intestine, into the subject's body. Advantageously, the inventors have demonstrated that mice colonised with the human microbiota bacterial strains of the invention excrete higher levels of xenobiotic than germ-free controls.

In some embodiments, the subject has ingested a xenobiotic, is suspected as having ingested a xenobiotic, or is at risk of ingesting a xenobiotic. In some embodiments, the subject is at risk of ingesting drinking water or foodstuff contaminated with a xenobiotic. In some embodiments, the subject is suspected as having ingested drinking water or foodstuff contaminated with a xenobiotic. In some embodiments (e.g. when the subject is at risk of exposure to a xenobiotic), the composition is for use as a prophylactic or preventative measure. In some embodiments, the subject is at risk of ingesting a xenobiotic during occupational or recreational activities. For example, the subject may work in close contact with xenobiotics or may visit environments contaminated with xenobiotics.

In some embodiments, the method comprises detecting the presence and/or measuring the abundance of the one or more bacterial strains in an environment prior to contacting with the composition. The presence and/or abundance of bacterial strains in an environment can be determined using methods known in the art, e.g. cell-based methods and/or molecular methods (e.g. qPCR). In some embodiments, the method comprises detecting the presence and/or measuring the abundance of the one or more bacterial strains in the subject prior to administration of the composition to the subject. The bacterial strains of the invention are common to the microbiota of humans, although the composition of microbiota varies between individuals. Whilst the microbiota of some individuals will contain each of the bacterial strains of the invention, the microbiota of other individuals may comprise only a subset (or potentially even none) of the bacterial strains of the invention. The presence and/or abundance of bacterial strains in a subject can be determined using methods known in the art, e.g. cell-based methods and/or molecular methods (e.g. qPCR). The amount of bacterial strain administered to the subject may be varied based on the presence and/or abundance of that bacterial strain in the subject. For example, when the subject has a high abundance of a bacterial strain, a lower amount may be administered as compared to the amount administered to a subject in which the bacterial strain is absent or at low abundance; and vice versa.

In some embodiments, the subject is a human. In some embodiments, the subject is an animal, optionally selected from a cow, sheep, pig, poultry (e.g. chicken, turkey), cat or dog.

The invention also provides a composition for reducing the level of xenobiotic in an environment, the composition comprising one or more bacterial strains selected from Bacteroides, Collinsella, Coprococcus, Eubacterium, Odoribacter, Parabacteroides, and Roseburia. The invention also provides a composition for reducing the level of xenobiotic in an environment, the composition comprising one or more bacterial strains selected from Bacteroides, Collinsella, Coprococcus, Eubacterium, Odoribacter, Parabacteroides, Roseburia, Escherichia, Phocaeicola, Prevotella, Butyrivibrio, Lacrimispora, Clostridium, Fusobacterium, Agathobacter, Dorea, and Streptococcus. As demonstrated herein, bacterial strains from each of these genera can bioaccumulate and/or biotransform xenobiotics, such as PFA.

References herein to "the composition" will be understood to embrace the composition of the invention, compositions for use in the methods of the invention, and compositions for use in the therapeutic methods of the invention.

In some embodiments, the composition comprises two or more bacterial strains selected from Bacteroides, Collinsella, Coprococcus, Eubacterium, Odoribacter, Parabacteroides, and Roseburia. In some embodiments, the composition comprises 3 or more (e.g. 4 or more, 5 or more, 6 or more, or 7) bacterial strains selected from Bacteroides, Collinsella, Coprococcus, Eubacterium, Odoribacter, Parabacteroides, and Roseburia. In some embodiments, the composition comprises two or more bacterial strains selected from Bacteroides, Collinsella, Coprococcus, Eubacterium, Odoribacter, Parabacteroides, Roseburia, Escherichia, Phocaeicola, Prevotella, Butyrivibrio, Lacrimispora, Clostridium, Fusobacterium, Agathobacter, Dorea, and Streptococcus. In some embodiments, the composition comprises 3 or more (e.g. 4 or more, 5 or more, 6 or more, 7 or more, 8 or more, 9 or more, 10 or more, 11 or more, 12 or more, 13 or more, 14 or more, 15 or more, 16 or more, or 17) bacterial strains selected from Bacteroides, Collinsella, Coprococcus, Eubacterium, Odoribacter, Parabacteroides, Roseburia, Escherichia, Phocaeicola, Prevotella, Butyrivibrio, Lacrimispora, Clostridium, Fusobacterium, Agathobacter, Dorea, and Streptococcus. As demonstrated herein, bacterial strains from each of these genera can bioaccumulate and/or biotransform xenobiotics.

In some embodiments, the composition comprises one or more of Bacteroides uniformis, Bacteroides caccae, Bacteroides clarus, Bacteroides dorei, Bacteroides stercoris, Bacteroides thetaiotaomicron, Collinsella aerofaciens, Coprococcus comes, Eubacterium rectale, Odoribacter splanchnicus, Parabacteroides distasonis, Parabacteroides merdae, and Roseburia intestinalis.

In some embodiments, the composition comprises one or more of Bacteroides uniformis, Bacteroides caccae, Bacteroides clarus, Bacteroides dorei, Bacteroides stercoris, Bacteroides thetaiotaomicron, Collinsella aerofaciens, Coprococcus comes, Eubacterium rectale, Odoribacter splanchnicus, Parabacteroides distasonis, Parabacteroides merdae, Roseburia intestinalis, Escherichia coll, Phocaeicola coprocola, Prevotella copri, Bacteroides eggerthii, Prevotella melaninogenica, Bacteroides fragilis, Bacteroides xylanisolvens, Butyrivibrio crossotus, Bacteroides coprocola, Roseburia hominis, Lacrimispora saccharolytica, Clostridium scindens, Fusobacterium nucleatum subsp. Nucleatum, Clostridium difficile, Phocaeicola vulgatus, Agathobacter rectalis, Roseburia inulinivorans, Dorea formicigenerans, Streptococcus salivarius, Fusobacterium nucleatum subsp. Animalis, Fusobacterium nucleatum subsp. Vincenti!, Clostridium hylemonae, and Clostridium sporogenes. As demonstrated herein, each of these bacterial strains can bioaccumulate and/or biotransform xenobiotics.

In some embodiments, the composition comprises two or more of Bacteroides uniformis, Bacteroides caccae, Bacteroides clarus, Bacteroides dorei, Bacteroides stercoris, Bacteroides thetaiotaomicron, Collinsella aerofaciens, Coprococcus comes, Eubacterium rectale, Odoribacter splanchnicus, Parabacteroides distasonis, Parabacteroides merdae, and Roseburia intestinalis. In some embodiments, the composition comprises 3 or more (e.g. 4 or more, 5 or more, 6 or more, 7 or more, 8 or more, 9 or more, 10 or more, 11 or more, 12 or more, or 13) of Bacteroides uniformis, Bacteroides caccae, Bacteroides clarus, Bacteroides dorei, Bacteroides stercoris, Bacteroides thetaiotaomicron, Collinsella aerofaciens, Coprococcus comes, Eubacterium rectale, Odoribacter splanchnicus, Parabacteroides distasonis, Parabacteroides merdae, and Roseburia intestinalis.

In some embodiments, the composition comprises two or more of Bacteroides uniformis, Bacteroides caccae, Bacteroides clarus, Bacteroides dorei, Bacteroides stercoris, Bacteroides thetaiotaomicron, Collinsella aerofaciens, Coprococcus comes, Eubacterium rectale, Odoribacter splanchnicus, Parabacteroides distasonis, Parabacteroides merdae, Roseburia intestinalis, and Escherichia coli. In some embodiments, the composition comprises 3 or more (e.g. 4 or more, 5 or more, 6 or more, 7 or more, 8 or more, 9 or more, 10 or more, 11 or more, 12 or more, 13 or more, 14 or more, 15 or more, 16 or more, 17 or more, 18 or more, 19 or more, 20 or more, 21 or more, 22 or more, 23 or more, 24 or more, 25 or more, 26 or more, 27 or more, 28 or more, 29 or more, 30 or more, 31 or more, 32 or more, 33 or more, 34 or more, 35 or more, or 36) of Bacteroides uniformis, Bacteroides caccae, Bacteroides clarus, Bacteroides dorei, Bacteroides stercoris, Bacteroides thetaiotaomicron, Collinsella aerofaciens, Coprococcus comes, Eubacterium rectale, Odoribacter splanchnicus, Parabacteroides distasonis, Parabacteroides merdae, Roseburia intestinalis, Escherichia coli, Phocaeicola coprocola, Prevotella copri, Bacteroides eggerthii, Prevotella melaninogenica, Bacteroides fragilis, Bacteroides xylanisolvens, Butyrivibrio crossotus, Bacteroides coprocola, Roseburia hominis, Lacrimispora saccharolytica, Clostridium scindens, Fusobacterium nucleatum subsp. Nucleatum, Clostridium difficile, Phocaeicola vulgatus, Agathobacter rectalis, Roseburia inulinivorans, Dorea formicigenerans, Streptococcus salivarius, Fusobacterium nucleatum subsp. Animalis, Fusobacterium nucleatum subsp. Vincentii, Clostridium hylemonae, and Clostridium sporogenes.

In some embodiments, one or more bacterial strains is genetically modified. In some embodiments, one or more of the bacterial strains comprises a genetic modification resulting in a decrease or elimination of xenobiotic efflux from the one or more bacterial strains. In some embodiments, the genetic modification comprises deletion or inactivation of at least one gene required for activity of an efflux pump. In some embodiments, the genetic modification comprises deletion or inactivation of at least one gene encoding an efflux pump or component thereof. Efflux pumps are a common mechanism used by several bacterial species to reduce intra-cellular concentration of toxic compounds. By disrupting or destroying the activity of efflux pumps, bioaccumulation by the bacterial strains is increased. Deletion or inactivation of genes may be achieved using any suitable method known in the art, such as site directed mutagenesis, homologous recombination, CRISPR-Cas9 based methods, or transcription activator-like effector nuclease (TALEN) based methods. In some embodiments, the genetic modification comprises deletion or inactivation of at least one gene required for an RND (resistance-nodulation-division) family transporter or a homolog thereof. RND family transporters are widespread, particularly amongst Gram-negative bacteria.

In some embodiments, the RND family transporter is the AcrAB-TolC complex or a homolog thereof. In some embodiments, the genetic modification comprises deletion or inactivation of at least one gene encoding a TolC family protein or a homolog thereof, e.g. tolC or a homolog thereof. In some embodiments, the genetic modification comprises deletion or inactivation of one or more (e.g. one, two, or all three) of acrA or a homolog thereof, acrB or a homolog thereof, and tolC or a homolog thereof.

In some embodiments, the genetic modification comprises deletion or inactivation of a gene encoding a membrane fusion protein (MFP) subunit of a RND family efflux transporter (e.g. UniProt accession number: R9I2M9) or a homolog thereof.

In some embodiments, the genetic modification comprises deletion or inactivation of a gene encoding a hydrophobe/amphiphile efflux-1 (HAE1) family RND transporter (e.g. UniProt accession number R9I2L8) or a homolog thereof.

In some embodiments, the genetic modification comprises deletion or inactivation of a gene encoding an outer membrane factor (OMF) lipoprotein of a NodT family efflux transporter (e.g. UniProt accession number R9I2R1) or a homolog thereof.

As used herein, homologs include functional and structural homologs which may be identified using methods known in the art, e.g. by sequence and/or structural homology.

In some embodiments, the composition comprises Bacteroides. In some embodiments, the composition comprises one or more (e.g. 2 or more, 3 or more, 4 or more, 5 or more, or 6) of Bacteroides uniformis, Bacteroides caccae, Bacteroides clarus, Bacteroides dorei, Bacteroides stercoris, and Bacteroides thetaiotaomicron. In some embodiments, the composition comprises Bacteroides. In some embodiments, the composition comprises one or more (e.g. 2 or more, 3 or more, 4 or more, 5 or more, 6 or more, 7 or more, 8 or more, 9 or more, or 10) of Bacteroides uniformis, Bacteroides caccae, Bacteroides clarus, Bacteroides dorei, Bacteroides stercoris, and Bacteroides thetaiotaomicron, Bacteroides eggerthii, Bacteroides fragilis, Bacteroides xylanisolvens, and Bacteroides coprocola.

In some embodiments, the composition comprises Bacteroides uniformis. In some embodiments, the composition comprises one or more (e.g. 2 or more, 3 or more, 4 or more, 5 or more, 6 or more, 7 or more, or 8) of Bacteroides uniformis DSM110080, Bacteroides uniformis DSM108148, Bacteroides uniformis DSM108147, Bacteroides uniformis DSM108145, Bacteroides uniformis DSM108146, Bacteroides uniformis HM-716, Bacteroides uniformis DSM6597, and Bacteroides uniformis HM-715. The inventors found that these eight different strains of B. uniformis exhibit comparable levels of PFNA and PFOA bioaccumulation, indicating that bioaccumulation of these compounds is a species-level property of B. uniformis.

In some embodiments, the composition comprises a Bacteroides that is genetically modified. In some embodiments, the Bacteroides comprises a genetic modification resulting in a decrease or elimination of xenobiotic efflux. In some embodiments, the genetic modification comprises deletion or inactivation of at least one gene required for activity of an efflux pump. In some embodiments, the genetic modification comprises deletion or inactivation of at least one gene required for activity of the R9I2M9 efflux transporter RND family, the R9I2L8 hydrophobe/amphiphile efflux-1 (HAE1) family RND transporter, and/or the R9I2R1 NodT family efflux transporter.

In some embodiments, the composition comprises Parabacteroides. In some embodiments, the composition comprises one or more (e.g. both) of Parabacteroides distasonis and Parabacteroides merdae.

In some embodiments, the composition comprises a Parabacteroides that is genetically modified. In some embodiments, the Parabacteroides comprises a genetic modification resulting in a decrease or elimination of xenobiotic efflux. In some embodiments, the genetic modification comprises deletion or inactivation of at least one gene required for activity of an efflux pump. In some embodiments, the genetic modification comprises deletion or inactivation of at least one gene required for activity of a homolog of the Bacteroides R9I2M9 efflux transporter RND family, the R9I2L8 hydrophobe/amphiphile efflux-1 (HAE1) family RND transporter, and/or the R9I2R1 NodT family efflux transporter.

In some embodiments, the composition comprises Bacteroides and/or Parabacteroides. In some embodiments, the composition comprises one or more (e.g. 2 or more, 3 or more, 4 or more, 5 or more, 6 or more, 7 or more, or 8) of Bacteroides uniformis, Bacteroides caccae, Bacteroides clarus, Bacteroides dorei, Bacteroides stercoris, Bacteroides thetaiotaomicron, Parabacteroides distasonis and Parabacteroides merdae. In some embodiments, the composition comprises Bacteroides and/or Parabacteroides. In some embodiments, the composition comprises one or more (e.g. 2 or more, 3 or more, 4 or more, 5 or more, 6 or more, 7 or more, 8 or more, 9 or more, 10 or more, 11 or more, or 12) of Bacteroides uniformis, Bacteroides caccae, Bacteroides clarus, Bacteroides dorei, Bacteroides stercoris, Bacteroides thetaiotaomicron, Parabacteroides distasonis, Parabacteroides merdae, Bacteroides eggerthii, Bacteroides fragilis, Bacteroides xylanisolvens, and Bacteroides coprocola. In some embodiments, the composition comprises Collinsella. In some embodiments, the composition comprises Collinsella aerofaciens.

In some embodiments, the composition comprises Coprococcus. In some embodiments, the composition comprises Coprococcus comes.

In some embodiments, the composition comprises Eubacterium. In some embodiments, the composition comprises Eubacterium rectale.

In some embodiments, the composition comprises Odoribacter. In some embodiments, the composition comprises Odoribacter splanchnicus.

In some embodiments, the composition comprises Roseburia. In some embodiments, the composition comprises Roseburia intestinalis. In some embodiments, the composition comprises one or more (e.g. 2 or more, or 3) of Roseburia intestinalis, Roseburia hominis, and Roseburia inulinivorans.

In some embodiments, the composition comprises Escherichia. In some embodiments, the composition comprises E. coli.

In some embodiments, the composition comprises an E. coli strain comprising a genetic modification resulting in a decrease or elimination of xenobiotic efflux. In some embodiments, the E. coli is genetically modified to delete or inactivate one or more genes encoding the AcrAB-TolC efflux pump. In some embodiments, the E. coli is genetically modified to delete or inactivate tolC. In some embodiments, the E. coli is E. coli BW25113 delta-tolC. In some embodiments, the E. coli is genetically modified to delete or inactivate toiC, acrA and/or acrB. In some embodiments, the E. coli is E. coli C43 (DE3) delta-AcrA-AcrB-TolC.

In some embodiments, the composition comprises Phocaeicola. In some embodiments, the composition comprises Phocaeicola coprocola and/or Phocaeicola vulgatus.

In some embodiments, the composition comprises Prevotella. In some embodiments, the composition comprises Prevotella copri and/or Prevotella melaninogenica.

In some embodiments, the composition comprises Butyrivibrio. In some embodiments, the composition comprises Butyrivibrio crossotus.

In some embodiments, the composition comprises Lacrimispora. In some embodiments, the composition comprises Lacrimispora saccharolytica. In some embodiments, the composition comprises Clostridium. In some embodiments, the composition comprises one or more (e.g. 2 or more, 3 or more, or 4) of Clostridium scindens, Clostridium difficile, Clostridium hylemonae, and Clostridium sporogenes.

In some embodiments, the composition comprises Fusobacterium. In some embodiments, the composition comprises Fusobacterium nucleatum. In some embodiments, the composition comprises one or more (e.g. 2 or more, or 3) of Fusobacterium nucleatum subsp. Nucleatum, Fusobacterium nucleatum subsp. Animalis, and Fusobacterium nucleatum subsp. Vincenti!.

In some embodiments, the composition comprises Agathobacter. In some embodiments, the composition comprises Agathobacter rectalis.

In some embodiments, the composition comprises Dorea. In some embodiments, the composition com prises Dorea formicigenerans.

In some embodiments, the composition comprises Streptococcus. In some embodiments, the composition comprises Streptococcus salivarius.

In some embodiments, the composition comprises one or more (e.g. 2 or more , 3 or more, 4 or more, 5 or more, 6 or more, 7 or more, 8 or more, 9 or more, 10 or more, 11 or more, 12 or more, 13 or more, 14 or more, 15 or more, 16 or more, 17 or more, 18 or more, 19 or more, or 20) of Phocaeicola vulgatus, Bacteroides uniformis, Bacteroides fragilis, Bacteroides thetaiotaomicron, Erysipelatoclostridium ramosum, Agathobacter rectalis, Roseburia intestinalis, Veillonella parvula, Eggerthella lenta, Fusobacterium nucleatum, Enterocloster bolteae, Clostridium perfringens, Lacrimispora saccharolytica, Streptococcus salivarius, Ruminococcus gnavus, Bariatricus comes, Parabacteroides merdae, Streptococcus parasanguinis, Collinsella aerofaciens, and Dorea formicigenerans.

In some embodiments, the composition comprises Erysipelatoclostridium ramosum. In some embodiments, the composition comprises Veillonella parvula. In some embodiments, the composition comprises Eggerthella lenta. In some embodiments, the composition comprises Enterocloster bolteae. In some embodiments, the composition comprises Clostridium perfringens. In some embodiments, the composition comprises Lacrimispora saccharolytica. In some embodiments, the composition comprises Ruminococcus gnavus. In some embodiments, the composition comprises Bariatricus comes. In some embodiments, the composition comprises Streptococcus parasanguinis.

In some embodiments, the composition comprises one or more (e.g. 2 or more, 3 or more, 4 or more, 5 or more, 6 or more, 7 or more, 8 or more, 9 or more, or 10) of Bacteroides caccae, Bacteroides dorei, Bacteroides thetaiomicron, Bacteroides uniformis, Bacteroides vulgatus, Colinsella aerofaciens, Coprococcus comes, Eubacterium rectale, Parabacteroides merdae, and Roseburia intestinalis. In some embodiments, the composition comprises one or more (e.g. 2 or more, 3 or more, 4 or more, 5 or more, 6 or more, 7 or more, 8 or more, 9 or more, or 10) of Akkermansia muciniphila, Bacteroides clarus, Bacteroides stercoris, Clostridium difficile, Eggerthella lenta, Eubacterium eligens, Fusobacterium nucleatum subsp. animalis, Odoribacter splanchnicus, Parabacteroides distastonis, and Ruminococcus bromii.

In some embodiments, the composition comprises Colinsella aerofaciens. In some embodiments, the composition comprises Akkermansia muciniphila. In some embodiments, the composition comprises Eggerthella lenta. In some embodiments, the composition comprises Eubacterium eligens. In some embodiments, the composition comprises Ruminococcus bromii.

In some embodiments, the xenobiotic comprises one or more of a PFA (per- and polyfluoroalkyl substances), a bisphenol, and a pesticide.

The terms "PFA" and "PFAS" are used interchangeably herein. Both terms refer to per- and polyfluoroalkyl substances.

In some embodiments, the xenobiotic comprises one or more of PFNA, PFOA, PFDeA, bisphenol AF, Boscalid, Propiconazole, Pyrimethanil, tributyl-PC , and triphenyl-PC .

In some embodiments, the xenobiotic is a PFA. In some embodiments, the xenobiotic comprises one or more (e.g. 2 or more, or 3) of PFNA (Perfluorononanoic acid), PFOA (Perfluorooctanoic acid), and PFDeA (Perfluorodecanoic acid).

In some embodiments, the xenobiotic is a bisphenol. In some embodiments, the xenobiotic comprises bisphenol AF.

In some embodiments, the xenobiotic is a pesticide. In some embodiments, the xenobiotic comprises one or more (e.g. 2 or more, or 3) of Boscalid, Propiconazole, and Pyrimethanil.

In some embodiments, the xenobiotic is a PFA and the composition comprises one or more (e.g. two or more, or 3) of Bacteroides, Odoribacter, and Parabacteroides. In some embodiments, the xenobiotic is a PFA and the composition comprises one or more (e.g. 2 or more, 3 or more, 4 or more, 5 or more, 6 or more, 7 or more, 8 or more, or 9) of Bacteroides uniformis, Bacteroides caccae, Bacteroides clarus, Bacteroides dorei, Bacteroides stercoris, Bacteroides thetaiotaomicron, Odoribacter splanchnicus, Parabacteroides distasonis, and Parabacteroides merdae. In some embodiments, the PFA comprises PFOA and/or PFNA. In some embodiments, the xenobiotic is a PFA and the composition comprises one or more (e.g. two or more, 3 or more, 4 or more, 5 or more, 6 or more, 7 or more, 8 or more, 9 or more, 10 or more, 11 or more, 12 or more, 13 or more, or 14) of Bacteroides, Odoribacter, Parabacteroides, Roseburia, Escherichia, Phocaeicola, Prevotella, Butyrivibrio, Lacrimispora, Clostridium, Fusobacterium, Agathobacter, Dorea, and Streptococcus. In some embodiments, the xenobiotic is a PFA and the composition comprises one or more (e.g. 2 or more, 3 or more, 4 or more, 5 or more, 6 or more, 7 or more, 8 or more, 9 or more, 10 or more, 11 or more, 12 or more, 13 or more, 14 or more, 15 or more, 16 or more, 17 or more, 18 or more, 19 or more, 20 or more, 21 or more, 22 or more, 23 or more, 24 or more, 25 or more, 26 or more, 27 or more, 28 or more, 29 or more, 30 or more, 31 or more or 32) of Bacteroides uniformis, Bacteroides caccae, Bacteroides clarus, Bacteroides dorei, Bacteroides stercoris, Bacteroides thetaiotaomicron, Odoribacter splanchnicus, Parabacteroides distasonis, Parabacteroides merdae, Phocaeicola coprocola, Prevotella copri, Bacteroides eggerthii, Prevotella melaninogenica, Bacteroides fragilis, Bacteroides xylanisolvens, Butyrivibrio crossotus, Bacteroides coprocola, Roseburia hominis, Parabacteroides distasonis, Lacrimispora saccharolytica, Clostridium scindens, Fusobacterium nucleatum subsp. Nucleatum, Clostridium difficile, Phocaeicola vulgatus, Agathobacter rectalis, Roseburia inulinivorans, Dorea formicigenerans, Streptococcus salivarius, Fusobacterium nucleatum subsp. Animalis, Fusobacterium nucleatum subsp. Vincenti!, Clostridium hylemonae, and Clostridium sporogenes. In some embodiments, the PFA comprises PFOA and/or PFNA. As demonstrated herein, each of these bacterial strains can efficiently bioaccumulate PFAs, such as PFNA.

In some embodiments, the xenobiotic is a PFA and the composition comprises Bacteroides. In some embodiments, the xenobiotic is a PFA and the composition comprises one or more (e.g. 2 or more, 3 or more, 4 or more, 5 or more, or 6) of Bacteroides uniformis, Bacteroides caccae, Bacteroides clarus, Bacteroides dorei, Bacteroides stercoris, and Bacteroides thetaiotaomicron. In some embodiments, the xenobiotic is PFA and the composition comprises one or more (e.g. 2 or more, 3 or more, 4 or more, 5 or more, 6 or more, 7 or more, 8 or more, 9 or more, or 10) of Bacteroides uniformis, Bacteroides caccae, Bacteroides clarus, Bacteroides dorei, Bacteroides stercoris, and Bacteroides thetaiotaomicron, Bacteroides eggerthii, Bacteroides fragilis, Bacteroides xylanisolvens, and Bacteroides coprocola. In some embodiments, the xenobiotic is a PFA and the composition comprises Bacteroides uniformis. In some embodiments, the PFA comprises PFNA, PFOA, and/or PFDeA.

In some embodiments, the xenobiotic is a PFAS, optionally PFNA, PFOA, and/or PFDeA, and the composition comprises an E. coli strain comprising a genetic modification resulting in a decrease or elimination of xenobiotic efflux. In some embodiments, the E. coli is genetically modified to delete or inactivate one or more genes encoding the AcrAB-TolC efflux pump. In some embodiments, the E. coli is genetically modified to delete or inactivate tolC. In some embodiments, the E. coli is E. coli BW25113 delta-tolC. In some embodiments, the E. coli is genetically modified to delete or inactivate tolC, acrA and/or acrB. In some embodiments, the E. coli is E. coli C43 (E)E3) delta-AcrA-AcrB-TolC. As demonstrated herein, E. coli strains comprising a genetic modification resulting in a decrease or elimination of xenobiotic efflux demonstrate high PFNA accumulation.

In some embodiments, the xenobiotic is a PFAS, optionally PFNA, PFOA, and/or PFDeA, and the composition comprises Phocaeicola. In some embodiments, the xenobiotic is a PFAS, optionally PFNA, PFOA, and/or PFDeA, and the composition comprises Phocaeicola coprocola and/or Phocaeicola vulgatus.

In some embodiments, the xenobiotic is a PFAS, optionally PFNA, PFOA, and/or PFDeA, and the composition comprises Prevotella. In some embodiments, the xenobiotic is a PFAS, optionally PFNA, PFOA, and/or PFDeA, and the composition comprises Prevotella copri and/or Prevotella melaninogenica.

In some embodiments, the xenobiotic is a PFAS, optionally PFNA, PFOA, and/or PFDeA, and the composition comprises Butyrivibrio. In some embodiments, the xenobiotic is a PFAS, optionally PFNA, PFOA, and/or PFDeA, and the composition comprises Butyrivibrio crossotus.

In some embodiments, the xenobiotic is a PFAS, optionally PFNA, PFOA, and/or PFDeA, and the composition comprises Lacrimispora. In some embodiments, the xenobiotic is a PFAS, optionally PFNA, and the composition comprises Lacrimispora saccharolytica.

In some embodiments, the xenobiotic is a PFAS, optionally PFNA, PFOA, and/or PFDeA, and the composition comprises Clostridium. In some embodiments, the xenobiotic is a PFAS, optionally PFNA, PFOA, and/or PFDeA, and the composition comprises one or more (e.g. 2 or more, 3 or more, or 4) of Clostridium scindens, Clostridium difficile, Clostridium hylemonae, and Clostridium sporogenes.

In some embodiments, the xenobiotic is a PFAS, optionally PFNA, PFOA, and/or PFDeA, and the composition comprises Fusobacterium. In some embodiments, the composition comprises Fusobacterium nucleatum. In some embodiments, the xenobiotic is a PFAS, optionally PFNA, PFOA, and/or PFDeA, and the composition comprises one or more (e.g. 2 or more, or 3) of Fusobacterium nucleatum subsp. nucleatum, Fusobacterium nucleatum subsp. animalis, and Fusobacterium nucleatum subsp. vincentii.

In some embodiments, the xenobiotic is a PFAS, optionally PFNA, PFOA, and/or PFDeA, and the composition comprises Agathobacter. In some embodiments, the xenobiotic is a PFAS, optionally PFNA, PFOA, and/or PFDeA, and the composition comprises Agathobacter rectalis. In some embodiments, the xenobiotic is a PFAS, optionally PFNA, PFOA, and/or PFDeA, and the composition comprises Dorea. In some embodiments, the xenobiotic is a PFAS, optionally PFNA, PFOA, and/or PFDeA, and the composition comprises Dorea formicigenerans.

In some embodiments, the xenobiotic is a PFAS, optionally PFNA, PFOA, and/or PFDeA, and the composition comprises Streptococcus. In some embodiments, the xenobiotic is a PFAS, optionally PFNA, PFOA, and/or PFDeA, and the composition comprises Streptococcus salivarius.

In some embodiments, the xenobiotic is a bisphenol and the composition comprises one or more (e.g. 2 or more, 3 or more, 4 or more, 5 or more, or 6) of Bacteroides, Collinsella, Eubacterium, Odoribacter, Parabacteroides, and Roseburia. In some embodiments, the xenobiotic is a bisphenol and the composition comprises one or more (e.g. 2 or more, 3 or more, 4 or more, 5 or more, 6 or more, 7 or more, 8 or more, 9 or more, or 10) of Bacteroides uniformis, Bacteroides caccae, Bacteroides clarus, Bacteroides dorei, Bacteroides thetaiotaomicron, Collinsella aerofaciens, Eubacterium rectale, Odoribacter splanchnicus, Parabacteroides merdae, and Roseburia intestinalis. In some embodiments, the bisphenol comprises bisphenol AF.

In some embodiments, the xenobiotic is a pesticide and the composition comprises one or more (e.g. 2 or more, 3 or more, 4 or more, 5 or more, or 6) of Bacteroides, Coprococcus, Eubacterium, Odoribacter, Parabacteroides, and Roseburia. In some embodiments, the xenobiotic is a pesticide and the composition comprises one or more (e.g. 2 or more, 3 or more, 4 or more, 5 or more, 6 or more, 7 or more, 8 or more, 9 or more, 10 or more, 11 or more, or 12) of Bacteroides uniformis, Bacteroides caccae, Bacteroides clarus, Bacteroides dorei, Bacteroides stercoris, Bacteroides thetaiotaomicron, Coprococcus comes, Eubacterium rectale, Odoribacter splanchnicus, Parabacteroides distasonis, Parabacteroides merdae, and Roseburia intestinalis. In some embodiments, the pesticide comprises boscalid, propiconazole, and/or pyrimethanil.

In some embodiments, the xenobiotic comprises tributyl-PC and the composition comprises one or more (e.g. 2 or more, 3 or more, 4 or more, or 5) of Bacteroides, Coprococcus, Odoribacter, Parabacteroides, and Roseburia. In some embodiments, the xenobiotic comprises tributyl-PC and the composition comprises one or more (e.g. 2 or more, 3 or more, 4 or more, 5 or more, 6 or more, 7 or more, 8 or more, 9 or more, or 10) of Bacteroides uniformis, Bacteroides caccae, Bacteroides clarus, Bacteroides dorei, Bacteroides stercoris, Coprococcus comes, Odoribacter splanchnicus, Parabacteroides distasonis, Parabacteroides merdae, and Roseburia intestinalis.

In some embodiments, the xenobiotic comprises triphenyl-PC and the composition comprises one or more (e.g. 2 or more, or 3) of Bacteroides, Coprococcus, and Odoribacter. In some embodiments, the xenobiotic comprises triphenyl-PO ₄ and the composition comprises one or more (e.g. 2 or more, or 3) of Bacteroides dorei, Coprococcus comes, and Odoribacter splanchnicus.

It will be appreciated that xenobiotics may adopt a different form depending on their surrounding environment. For example, xenobiotics in solution may become deprotonated. The invention is not limited to any particular xenobiotic form, and embraces e.g. salts, hydrates, solvates, crystalline forms, amorphous forms, and mixtures thereof, of xenobiotics described herein.

The bacterial strains may be wild-type or genetically modified. The composition typically comprises viable cells. In some embodiments, the composition comprises at least 10 ⁴ bacterial cells, optionally at least 10 ⁵ , at least 10 ⁶ , at least 10 ⁷ , at least 10 ⁸ , at least 10 ⁹ , at least IO ¹⁰ , at least 10 ¹¹ , at least 10 ¹² , at least 10 ¹³ , at least 10 ¹⁴ , or at least 10 ¹⁵ bacterial cells.

In some embodiments, the composition comprises dividing cells e.g. cells in log phase. In some embodiments, the composition comprises non-dividing cells e.g. cells which are resting or inactive.

In some embodiments, the composition comprises one or more bacterial strains which have been adapted for bioaccumulation and/or biotransformation of xenobiotics. In some embodiments, the one or more bacterial strains which have been adapted for bioaccumulation and/or biotransformation of xenobiotics by adaptive evolution. Adaptive evolution may comprise exposing one or more bacterial strains to media comprising a xenobiotic, allowing the bacterial strain to divide for a predetermined period of time or until a predetermined cell density is reached, and then transferring a subset of the bacterial strain population to new media comprising the xenobiotic. Multiple serial transfers (e.g. at least 5, at least 10, at least 15, at least 20, at least 25, at least 30, at least 35, at least 40, at least 45, or at least 50 serial transfer) are typically performed to provide a population of adapted bacterial cells. In some embodiments, the concentration of the xenobiotic may be increased after a number of serial transfers, e.g. after 5 serial transfers. The bioaccumulation and/or biotransformation activity of the adapted bacterial cell population may be compared to the bioaccumulation and/or biotransformation activity of the starting bacterial cell population to identify adapted bacterial cell populations having improved activity.

In some embodiments, the level of a xenobiotic is reduced by at least 5%, e.g. at least 5%, at least 10%, at least 15%, at least 20%, at least 25%, at least 30%, at least 40%, at least 50%, at least 60%, at least 70%, at least 80%, at least 90%, or at least 95%, as compared to the level of the xenobiotic in the contacted area of the environment, prior to contact with the composition. In some embodiments, the level of reduction is assessed less than 6 hours after contact with the composition, e.g. less than 6 hours, less than 5 hours, less than 4 hours, less than 3 hours, less than 2 hours, less than 1 hour, or less than 30 minutes after contact with the composition. In some embodiments, the level of reduction is assessed more than 6 hours after contact with the composition, e.g. more than 6 hours, more than 12 hours, more than 24 hours, more than 48 hours, more than 72 hours or more than 96 hours, more than one week, or more than one month after contact with the composition.

In some embodiments, the level of a xenobiotic is reduced by at least 5%, e.g. at least 5%, at least 10%, at least 15%, at least 20%, at least 25%, at least 30%, at least 40%, at least 50%, at least 60%, at least 70%, at least 80%, at least 90%, or at least 95%, as compared to the level of the xenobiotic in the subject, prior to administration of composition to the subject. In some embodiments, the level of xenobiotic in the subject refers to the level of xenobiotic in the intestine of the subject. In some embodiments, the level of reduction is assessed less than 6 hours after administration of the composition, e.g. less than 6 hours, less than 5 hours, less than 4 hours, less than 3 hours, less than 2 hours, less than 1 hour, or less than 30 minutes after administration. In some embodiments, the level of reduction is assessed more than 6 hours after administration of the composition, e.g. more than 6 hours, more than 12 hours, more than 24 hours, more than 48 hours, more than 72 hours or more than 96 hours, more than one week, or more than one month after administration.

In some embodiments, the composition is formulated for oral delivery, rectal delivery, and/or nasal delivery (e.g. via a nasoduodenal tube). In some embodiments, the composition is formulated for delivery to the intestine of a subject.

In some embodiments, the composition is in the form of a capsule, a tablet, a powder, or a fluid. In some embodiments, the one or more bacterial strains are lyophilised.

In some embodiments, the composition comprises one or more prebiotics for promoting growth of the one or more bacterial strains. In some embodiments, the method comprises administering one or more prebiotics to the subject prior to, consecutively with, or subsequent to administration of the composition.

The invention provides a dietary supplement comprising the composition of the invention. In some embodiments, the dietary supplement comprising one or more prebiotics. In some embodiments, the dietary supplement is a food or a drink. In some embodiments, the dietary supplement is a tablet or capsule.

In some embodiments, the one or more prebiotics are selected from arabinoxylan, xylose, fiber dextran, corn fiber, polydextrose, lactose, N-acetyl-lactosamine, glucose, galactose, fructose, rhamnose, mannose, uronic acids, arabinose, fructose, fucose, lactose, galactose, glucose, mannose, D-xylose, xylitol, ribose, xylobiose, sucrose, maltose, lactose, lactulose, trehalose, cellobiose, xylooligosaccharide, fructooligosaccharide, galactooligosaccharide, lactosucrose, and soybean oligosaccharides.

In some embodiments, the composition comprises a sweetener, flavouring agent, and/or colouring agent. Sweeteners include, but are not limited to, glucose, dextrose, fructose, and saccharin. Flavouring agents include, but are not limited to, synthetic and natural oils, and plant extracts. Suitable colouring agents include food colouring.

In some embodiments, the composition further comprises a carrier, excipient, and/or diluent. Generally, the carrier is a pharmaceutically-acceptable carrier. Non-limiting examples of pharmaceutically acceptable carriers include water, saline, and phosphate-buffered saline. Non-limiting examples of excipients include mannitol, lactose, starch, magnesium stearate, sodium saccharine, cellulose, and magnesium carbonate. Non-limiting examples of diluents include sugars (e.g. monosaccharides, disaccharides and oligosaccharides), calcium sulphate dihydrate, dextrins, starch, and glycine.

In some embodiments, the composition comprises a preservative. Non-limiting examples of preservatives include sucrose, sodium ascorbate, glutathione, cryoprotectant (e.g. a disaccharide, a polyol, or a polysaccharide).

In some embodiments, the composition comprises a gastro-resistant coating. In some embodiments, the composition is in the form of a capsule or tablet comprising a gastro-resistant coating. Advantageously, gastro-resistant coatings can be used to help target the composition of the invention e.g. to the intestine. Advantageously, this helps increase the abundance of viable bacterial strains within the intestine.

Gastro-resistant coatings are typically selected from fatty acids, waxes, shellac, plastics and plant fibers, and include e.g. hydroxypropyl methyl cellulose phthalate, methyl acrylate-methacrylic acid copolymers, cellulose acetate succinate, cellulose acetate trimellitate, sodium alginate, hydroxypropyl methyl cellulose acetate succinate, polyvinyl acetate phthalate (PVAP), cellulose acetate phthalate (CAP), zein, methyl methacrylate-methacrylic acid copolymers and enteric coating aqueous solution (e.g. ethylcellulose, medium chain triglycerides, oleic acid, sodium alginate, stearic acid).

In some embodiments, the composition is a time-release formulation. In some embodiments, the composition is in the form of a time-release capsule or tablet. Time-release formulations include sustained release formulations (where prolonged release is intended), pulse release formulations and delayed release formulations (e.g. to target different regions of the digestive tract, e.g. different regions of the intestine). In some embodiments, time-release formulations of the invention are formulated to allow the composition to be released gradually into the digestive tract (e.g. to distribute the bacterial strains throughout the intestine) or to be released in a delayed manner (e.g. to delay release of the bacterial strains until the intestine has been reached). Time-release formulations are known in the art, and typically include e.g. polymeric based components or coating membranes. Typically, time-release formulations may be formulated to release within 1-10 hours, optionally 2-8 hours, 3-6 hours following ingestion.

EXAMPLES

The invention will be further clarified by the following examples, which are intended to be purely exemplary of the invention and are in no way limiting.

Example 1

Multiple bacterial strains that are prevalent in the microbiome of humans were investigated for their ability to biotransform and/or bioaccumulate xenobiotics.

Experimental setup:

For each xenobiotic, a set concentration of 20 pM was screened. Each xenobiotic-strain interaction was screened in three biological replicates with two technical replicates each. The screen was carried out under anaerobic conditions in 96-well plates in 500 pl volume. Plates containing 250 pl Modified Gifu Anaerobic Media (mGAM) with 40 pM xenobiotic concentration were prepared the evening prior and placed into the anaerobic chamber over night to ensure anaerobe conditions for inoculation. On the day of the screen each well was inoculated with 250 pl of a second passage culture to reach a starting OD ₆ oo of 0.05. For compound control wells bacteria-free mGAM was added to the respective wells.

Plates were incubated at 37°C for 24 h, after which they were removed from the anaerobic chamber for further processing. 100 pl whole culture was transferred to fresh 96-well plates and stored at -80°C until extraction. The rest of the culture was centrifuged at 25°C and 4000 rpm for 15 min. 100 pl supernatant was transferred to fresh 96-well plates and stored at -80°C until extraction.

50 pl sample was extracted with 200 pl ice-cold MeOH:ACN (1:1) containing 60 pM internal standard and incubated at 4°C for 30 min. Plates were then centrifuged at 4°C and 4000 rpm for 10 min. Supernatants were transferred to 96-well plates for LC-MS analysis. Samples for concentration calibration and bacteria-free compound controls were processed in the same way. Data analysis:

Calibration curves were used to estimate the concentrations of target compounds in each sample. Statistical analysis was performed in RStudio Version 1.3.1093 using t-Test. The p-values were corrected for number of compounds tested using the false discovery rate (FDR) method. Additionally, the median of each strain was compared to the median of the compound control. The cut off value for significant reduction was set to p.adj < 0.05 and median reduction of >20%. Bioaccumulation was defined as significant reduction in concentration in the supernatant, but not the whole lysate sample and biotransformation was defined as significant reduction in concentration in both supernatant and lysate sample.

Results:

The level of the exemplary xenobiotics PFNA, PFOA, bisphenol AF, Boscalid, Propiconazole, Pyrimethanil, tributyl-PC , and triphenyl-PCU were found to be significantly reduced by at least one of the bacterial strains tested. Bisphenol AF (BPAF), PFOA, PFNA and Triphenyl-PO4 were mainly bioaccumulated, while Boscalid, Propiconazole, Pyrimethanil and Tributyl-PO4 were mainly biotransformed (Figure 1). Representative data demonstrating bioaccumulation of PFNA and biotransformation of propiconazole by Bacteroides dorei and Bacteroides uniformis is provided in Figure 2.

Discussion:

The inventors discovered that a variety of gut bacterial strains are able to reduce the concentration of xenobiotics via bioaccumulation and/or biotransformation. Bisphenol (exemplified by bisphenol AF) was found to be accumulated by several bacterial strains to up to 60%. PFAs (exemplified by PFOA and PFNA) were also shown to be accumulated up to 60% and 75% respectively. Without wishing to be bound by theory, the inventors believe that results indicating biotransformation of PFOA and PFNA may be a result of incomplete extraction from cells, and that strains demonstrating biotransformation (/.e. reduced levels of PFOA and PFNA in the supernatant) are more likely to be bioaccumulating these compounds. Pesticides (e.g. Boscalid, Propiconazole, and Pyrimethanil) were typically biotransformed by the bacterial strains. The results presented herein indicate that gut bacterial strains can be used to reduce the level of a range of xenobiotics via bioaccumulation and biotransformation. Example 2

To validate the experimental setup and extraction method, the inventors used PFAs and B. uniformis as a model system. The PFOA and PFNA bioaccumulation assay was repeated in B. uniformis using glass vials. Extraction was expanded by sonication to improve extraction of xenobiotic compounds from the cells.

Experimental setup:

PFOA and PFNA were tested at a concentration of 20 pM in B. uniformis in three technical replicates. The screen was carried out under anaerobic conditions in glass vials in a 3 ml volume. Glass vials containing 1.5 ml mGAM with 40 pM xenobiotic concentration were prepared the evening prior and placed into the anaerobic chamber over night to ensure anaerobic conditions for inoculation. On the day of the screen each well was inoculated with 1.5 ml of a second passage culture to reach a starting ODsoo of 0.05. For the compound control bacteria-free mGAM was added to the respective glass vials.

The vials were incubated at 37°C for 24 h, after which they were removed from the anaerobic chamber for further processing. 1 ml whole culture was transferred to a fresh glass vial. The rest of the culture was centrifuged at 25°C and 4000 rpm for 15 min and 1 ml supernatant was transferred to a fresh glass vial. The rest of the supernatant was removed from the cell pellet and the pellet was resuspended in 2 ml water, of which 1 ml was transferred to a fresh glass vial. All samples were stored at -80°C until extraction.

For extraction samples were thawed and sonicated for 3 min. Each 1 ml sample was then extracted with 4 ml ice-cold MeOH:ACN (1:1) containing 60 pM internal standard, sonicated again for 3 min and incubated at 4°C for 10 min. Vials were then centrifuged at 4°C and 4000 rpm for 15 min. Supernatants were transferred to LC-MS vials for analysis. Samples for concentration calibration and bacteria-free compound controls were processed in the same way.

Data analysis:

The median of each sample was compared to the median of the compound control.

Results:

PFOA was reduced in concentration by B. uniformis to 32% and the amount recovered from the cell pellet was 20%, while PFNA was reduced in concentration to 42%, and 48% were recovered from the cell pellet, supporting the accumulation of PFOA and PFNA by B. uniformis (Figure 3). Example 3

B. uniformis response in PFA accumulation at varying doses, accumulation of PFAs with varying chain length and PFNA accumulation over time

To determine capacity of B. uniformis to accumulate PFNA, concentrations between 0.78 and 500 pM were tested in a resting cell assay. In addition, the accumulation capability of B. uniformis was tested for two PFAs with different chain lengths. Finally the accumulation of PFNA was measured over the course of 7 days to determine how quickly PFNA is accumulated and whether it is released again over time.

Experimental setup:

PFNA was tested at a concentration between 0.78 and 500 pM in B. uniformis in three technical replicates. In addition, two different PFAs with varying chain length (PFNA and PFDeA) were tested in three technical replicates at a set concentration of 20 pM. The screen was carried out under anaerobic conditions in 96-well plates in 400 pl volume. Plates containing 200 pl PBS with 2x xenobiotic concentration were prepared the evening prior and placed into the anaerobic chamber over night to ensure anaerobic conditions for inoculation. On the day of the assay, the bacterial culture was centrifuged, supernatant removed and the bacterial pellet resuspended in PBS to reach an ODsoo of 7.5. Each well was inoculated with 200 pl of culture in PBS to reach a starting ODsoo of 3.75. For compound control wells bacteria-free PBS was added to the respective wells.

Plates were incubated at 37°C for 4 h, after which they were removed from the anaerobic chamber for further processing. 50 pl whole culture was transferred to fresh 96-well plates and stored at -80°C until extraction. The rest of the culture was centrifuged at 4°C and 4000 rpm for 10 min. 50 pl supernatant was transferred to fresh 96-well plates and also stored at -80°C until extraction.

50 pl sample was extracted with 200 pl ice-cold MeOH:ACN (1:1) containing 60 pM internal standard and incubated at 4°C for 10 min. Plates were then centrifuged at 4°C and 4000 rpm for 15 min. Supernatants were transferred to 96-well plates for LC-MS analysis. Samples for concentration calibration and bacteria-free compound controls were processed in the same way.

Data analysis:

Calibration curves were used to estimate the concentrations of target compounds in each sample. Statistical analysis was performed in Rstudio Version 1.3.1093 using t-Test. The p-values were corrected for number of concentrations tested using the FDR method. Additionally, the median concentration of each sample was compared to the median of the compound control. Cut off values for significant reduction were set to p.adj < 0.05 and median reduction of > 20 %.

Results:

PFNA was significantly accumulated at all tested concentrations at a constant rate of between 25 and 40 % (Figure 4).

The inventors found that PFNA was accumulated to 25% and PFDeA was accumulated to 60% by B. uniformis (Figure 5). These results may indicate that the accumulation capacity of B. uniformis increases with increasing PFA lipophilicity.

Over the course of 7 days the percentage of accumulated PFNA stayed constant. Even at 0 time point 30 % of PFNA was accumulated. This suggests that within the time it took to collect the samples after adding PFNA, B. uniformis had already accumulated a significant amount of the compound and this didn't change over the course of further 7 days (Figure 6) indicating that once bioaccumulated, the xenobiotic is not released back into the environment.

LCMS methods used in the Examples

In total two Quadrupole Time-of-Flight (QTOF) methods were set up to detect the xenobiotics.

QTOF method in positive scanning mode

A method from Agilent (Aimei Zou, S. P., et al., Agilent Technologies, Inc. Comprehensive LC/MS/MS Workflow of Pesticide Residues in Food Using the Agilent 6470 Triple Quadrupole LC/MS System, (2020)) was adapted to enable analysis of a wide range of xenobiotics. In short, LC-MS analysis was performed on an Agilent 1290 Infinity II LC system coupled with an Agilent 6546 LC/Q-TOF (Agilent). The separation was performed using an ZORBAX RRHD Eclipse Plus column (C18, 2.1 x 100 mm, 1.8 pm; Agilent) with the ZOBRAX Eclipse Plus (C18, 2.1 x 5 mm, 1.8 pm; Agilent) guard column at 40°C. The multisampler was kept at a temperature of 5°C. The injection volume was lpL and the flow rate was 0.4mL/min. The mobile phases consisted of A: water + 0.1% formic acid + 5mM ammonium formate; B: methanol + 0.1% formic acid + 5mM ammonium formate. The lOmin gradient started with 5% solvent B, which was increased to 30% by lmin and then further increased to 100% by 7min and held for another 3min, followed by a 5min equilibration to the starting conditions (5% mobile phase B). The QTOF MS scan operated in positive scanning mode (30-1500m/z). The source parameters were as follows: gas temperature: 200°C, drying gas: 9L/min, nebulizer: 20psi, sheath gas temperature: 400°C, sheath gas flow: 12L/min, VCap: 3000V, nozzle voltage: 0V, fragmentor: 110V, skimmer 45V, Oct RF Vpp: 750V. The online mass calibration was performed using a reference solution (121.05 and 922.01 m/z). The compounds were identified based on their retention time, accurate mass and fragmentation patterns.

QTOF method in negative scanning mode

A QTOF method in negative scanning mode was designed to detect compounds not detectable in positive scanning mode (Jurek, A. & Leitner, E. Food Additives & Contaminants: Part A 35, 2256-2269, (2018)). In short, LC-MS analysis was performed on an Agilent 1290 Infinity II LC system coupled with an Agilent 6546 LC/Q-TOF (Agilent). The separation was performed using an ZORBAX RRHD Eclipse Plus column (C18, 2.1 x 100 mm, 1.8 pm; Agilent) with the ZOBRAX Eclipse Plus (C18, 2.1 x 5 mm, 1.8 pm; Agilent) guard column at 40°C. The multisampler was kept at a temperature of 5°C. The injection volume was lpL and the flow rate was 0.4mL/min. The mobile phases consisted of A: water + 5 mM ammonium acetate + 0.03% acetic acid; B: methanol + 5 mM ammonium acetate + 0.03% acetic acid. The lOmin gradient started with 35% solvent B, which was increased to 100% by 9min and held for 1 min, followed by a 5min equilibration to the starting conditions (35% mobile phase B). The QTOF MS scan operated in negative scanning mode (30-1500m/z). The source parameters were as follows: gas temperature: 200°C, drying gas: 9L/min, nebulizer: 20psi, sheath gas temperature: 400°C, sheath gas flow: 12L/min, VCap: 3000V, nozzle voltage: 0V, fragmentor: 110V, skimmer 45V, Oct RF Vpp: 750V. The online mass calibration was performed using a reference solution (112.99 and 1033.99 m/z). The compounds were identified based on their retention time, accurate mass and fragmentation patterns. Some compounds not detectable with these mobile phases were run on the same method but with the mobile phases A: water; B: methanol.

Data analysis:

The Agilent MassHunter Qualitative Analysis 10.0 software was used to qualify the selected xenobiotic standards. TIC, EIC and ElC-fragment graphs were extracted for each compound. The Agilent MassHunter TOF Quantitative Analysis (Quant-My-Way) software was used to quantify the xenobiotic compounds in each sample.

Example 4

The inventors performed a community-based screening approach, wherein the ability of a mix of gut bacterial strains to sequester the pollutant compounds during 4 h exposure was assessed (Fig. 11a). 18 pollutants were found to be sequestered to more than 25 % by one or both synthetic communities (Fig. 11c). Ten of these compounds were then tested for sequestration by individual strains during 24 h incubation period (Fig. lib); seven pollutants were found to be sequestered by at least one of the bacterial strains (Fig. 7a). By comparing the whole culture and supernatant concentrations to compound controls the inventors were able to distinguish the observed sequestration between bioaccumulation and biodegradation. In the context of this example, bioaccumulation was defined as compound sequestration to at least 20 % from the supernatant but complete recovery the whole culture sample, whereas biodegradation was defined as both supernatant and whole culture sample showing more than 20 % sequestration (Fig. 7b, c). These interactions provide a starting point to mechanistically understand the role of the microbiota in pollutant xenobiotics.

The degree of bioaccumulation over 24 h for 20 pM PFNA exposure varied from 25 % (P. merdae) to 74 % (O. splanchnicus) and for PFOA from 23 % (P. merdae) to 58 % (O. splanchnicus). Following the growth of B. uniformis and concentration of PFNA over the course of 11 h showed that growth correlated with sequestration from the media (Fig. 7e). When spiking 20 pM PFNA into an already grown culture of B. uniformis, 50 % of PFNA was bioaccumulated within minutes (Fig. 7f). This was surprising as previous reports from environmental bacteria isolated from PFAS contaminated sites, Pseudomonas sp., showed much lower efficiency with ~40 % bioaccumulation of perfluorohexane sulfonate (PFHxS) over 5 days, and that following solvent pre-conditioning to facilitate sequestration. B. uniformis accumulated 60-70 % PFNA at sub-micromolar concentrations and circa 55 % at concentrations ranging from 1-100 pM PFNA (Fig. 7g, Fig. 13). In addition, B. uniformis and other abundant gut bacterial strains showed consistent growth up to high micromolar concentration of PFAS (Fig. 7h). In an experiment with 250 pM PFNA exposure, B. uniformis accumulated circa 40 % of the chemical corresponding to apparent intracellular concentration of circa 18 mM, well above that of most native metabolites.

A previous study indicated that accumulation in lipid bilayers may be the main mechanism of PFAS bioaccumulation and so the inventors considered the specificity of bioaccumulation. While all PFAS bioaccumulators were gram-negative species, not all tested gram-negative species accumulated PFAS to the same extent, including Escherichia coli (Fig. 12a, b). This specificity is inconsistent with membrane interaction being the main mechanism underlying PFAS bioaccumulation. To test the underlying differences between species, the inventors tested whether inactive cell mass (i.e., dead, lysed cells) could bioaccumulate PFAS. In a resting cell assay in PBS buffer (OD = 3.75) B. uniformis and O. splanchnicus bioaccumulated PFOA, PFNA and PFDeA to ~20 %, ~55 %, and ~85 % respectively, with dead and lysed cultures accumulating similar amounts. Live E. coli on the other hand bioaccumulated much lower levels of PFOA, PFNA and PFDeA ( ~5 %, ~25 % and ~40 %), while in dead and lysed state amounts similar to B. uniformis and O. splanchnicus were bioaccumulated (Fig. 8a, Fig. 14). Thus, transport across the membrane seems to distinguish E. coli from other bioaccumulating gram negative species. Further, the observed degree of bioaccumulation is unlikely to be membrane only as that would imply up to 1 molecule of PFNA per 2 molecules of lipids in the cell membrane, which would be physiologically improbable since the bioaccumulating bacteria showed robust growth and accumulation capacity over a considerable range of PFNA concentrations. This data suggests that PFAS bioaccumulation is intracellular and not a passive phenomenon driven by attachment to membrane lipid bilayer.

The inventors next tested how cells would cope with high mM intra-cellular levels of highly effective surfactants like PFAS and maintain their growth. To gain insights into morphology of bioaccumulating cells, the inventors used Transmission Electron Microscopy (TEM) to compare control, DMSO treated, cells with those bioaccumulating PFNA or PFDeA. The PFAS treated cells featured remarkable change in nucleoid appearance in TEM (Fig. 9e-n). This suggests interaction of PFNA and PFDeA with intracellular proteins and other macromolecules. The containment of the xenobiotic in/around these granular structures thus appears to be an effective mechanism for the cells to maintain their viability and growth.

The intra-cellular storage of PFAS supports existence of transport mechanisms capable of importing and/or exporting these organic fluorosurfactants without affecting the integrity of the cell membrane. To verify this, the inventors measured bioaccumulation of E. coli gene knock-out mutants that lacked one or more genes coding for efflux pump proteins (Fig. 8b). Efflux pumps are a common mechanism used by several bacterial species to reduce intra-cellular concentration of toxic compounds. The inventors reasoned that E. coli did not bioaccumulate PFNA to the same extent as other tested gramnegative bacteria because it could, at least to some extent, pump out PFAS. Confirming this hypothesis, two efflux pump mutants showed appreciable bioaccumulation, up to 30-40 % PFNA and 60-70 % PFDeA from the medium, while the wildtype strains only accumulated 10-20 % PFNA and 30- 40 % PFDeA (Fig. 8c). Further, these mutants showed increased sensitivity to high concentrations of PFNA, but not PFDeA (Fig. 8d). These results further attest that PFAS bioaccumulation is not a passive, membrane interaction driven, process, and bacterial cells can modulate the bioaccumulation through the transport machinery.

The inventors next used an alternative strategy to ascertain the biological nature of PFAS bioaccumulation, viz., adaptive laboratory evolution (Fig. 9a). The inventors reasoned that if the cells had mechanisms to modulate bioaccumulation and/or to cope with high intra-cellular levels, resistance to high PFAS exposure would rapidly evolve under natural selection. In addition, the inventors wanted to determine whether evolution with high PFAS concentrations would lead to strains with altered bioaccumulating capacity. The inventors therefore evolved, through serial transfer, B. uniformis, B. thetaiomicron, P. merdae, C. difficile and E. coli BW ATolC in a medium with 500 pM PFHpA, 500 pM PFOA, 250 pM PFNA or 125 pM PFDeA. Within 20 transfers (20 days) - corresponding to circa 100 generations - the growth of B. uniformis improved from 2 % to 76 % for PFDeA and from 7 % to 48 % for PFNA compared to the untreated control populations (Fig. 9b). This rapid adaptation affirms the evolvability of bacteria-PFAS interactions. For other tested strains and compounds no significant adaptation of growth was observed (Fig. 15). When comparing bioaccumulation capability of parental vs. evolved populations for PFNA and PFDeA, no differences were observed (Fig. 9c, d, Fig. 16a, b). This is promising regarding uses of bacterial strains for PFAS removal under high and/or chronic exposure environments, as while growth of bacteria may change, the results described herein suggest that bioaccumulation capabilities remain.

To determine the in vivo relevance of gut bacterial accumulation of PFAS, germfree C57BL/6 mice or C57BL/6 mice colonized with a community of 20 human gut bacterial strains (Com20) were administered a one-time oral dose of PFNA (10 mg/kg body weight) administered by gavage. Faecal samples were collected over the following 2 days and on day 3 colon and small intestine content samples were taken post-euthanization (Fig. 10a). Colonized mice showed a substantially higher excretion of PFNA at all follow-up time points (3h, p=0.009, fold change=9; Id, p=0.001, fold change=2.9; 2d, p<0.001, fold change=2.9; 3d colon, p<0.001, fold change=3.4; 3d small intestine, p=0.007, fold change=2) (Fig. 10b). The increased clearance through faecal matter in colonized mice shows that the PFNA accumulation by gut bacteria also happens in vivo. Together with in vitro results showing a varying degree of PFNA accumulation by different gut bacterial species, the mouse data indicates that the gut microbiota composition is a critical factor determining PFAS toxicokinetics.

In summary, the inventors discovered bacterial species with remarkably high capacity to intracellularly accumulate PFAS pollutants by gut bacteria. The specificity and genetic tractability of PFAS bioaccumulation offer possibilities for removing PFAS from, e.g., human body, using commensal bacterial strains.

Mouse faecal samples were analysed for their microbiota composition using 16S-sequencing. 17 out of 20 human gut bacterial strains colonized the gut of Com20 inoculated germ-free mice. No difference in composition was observed between control (DMSO) and PFNA (10 mg/kg body weight) treated mice colonized with Com20 (Table 1). In both DMSO and PFNA treated groups the percentage of bacteria classified as high-PFNA-accumulating was around 70 % (Table 2). Table 1. Relative abundance of gut bacterial species in control (DMSO) and PFNA (10 mg/kg body weight) treated germ-free (GF) and mice colonized with human gut bacteria. Columns: time points (0, 0.125, 1-, 2- and 3-days post-treatment) and collection sites (faecal pellets, colon, small intestine) of samples.

Table 2. Relative abundance of high- and low-accumulating strains in Com20 colonized control mice, Com20 colonized PFNA treated mice and germ-free PFNA treated mice. Column: time points (0, 0.125, 1-, 2- and 3-days post treatment) and collection sites (faecal pellets, colon, small intestine) of samples.

Methods

Bacterial strains and cultivation

Strains were selected to represent prevalent and abundant members of the healthy human gut microbiota. E. coli mutants were obtained from Typas lab (EMBL Heidelberg, BW25113 wild-type, BW25113 AtolC) and Luisi lab (University of Cambridge, C43 (DE3) wild-type, C43 (DE3) AacrAB-tolC). All bacterial experiments were performed in an anaerobic chamber (Coy Laboratory Products) filled with 2 % hydrogen and 12 % carbon dioxide in nitrogen. The chamber was equipped with a palladium catalyst system for oxygen removal, a dehumidifier, and a hydrogen sulfide removal system. Bacteria were grown at 37 °C in modified Gifu anaerobic medium (mGAM, HyServe, Germany, produced by Nissui Pharmaceuticals), prepared according to the instructions from the manufacturer and sterilised by autoclaving. Bacteria for starting cultures were grown for one or two days (depending on growth rate) in 10ml of media in 15 ml plastic tubes, which were inoculated directly from frozen glycerol stocks. Cultures were then diluted 100-fold and incubated again for the same amount of time before starting the experiments. Unless otherwise specified, the screening plates/tubes with cultivation medium were prepared the day prior at 2x compound concentration (2 % DMSO) and placed into the chamber over night to ensure anaerobic conditions for inoculation. Inoculation was performed 1:1 with a bacterial culture and plates were sealed with AlumaSeal II film (A2350-100EA) to avoid evaporation during incubation. Community-based screening approach (Fig. 11a, c)

On the day of the screen communities were assembled by pooling together second passages of individual strains according to their OD600 values. Assembled communities were then centrifuged at 25 °C and 4000 rpm for 15 min and the pellet was resuspended in PBS buffer to create a community with OD600 of 7.5. Each well was inoculated 1:1 with community in PBS to reach a starting OD600 of 3.75. For compound control wells bacteria-free PBS was added to the respective wells. Plates were incubated at 37 °C for 4 h, after which they were centrifuged at 21 °C and 4000 rpm for 15 min. The supernatant and compound controls were transferred to fresh 96-well plates and stored at -80 °C until extraction.

Single strain-xenobiotic screen (Fig. 7a,c,b)

Ten compounds identified by the 'Community-based screening approach' were tested for sequestration by individual strains. On the day of the screen each well was inoculated 1:1 with a second passage culture to reach a starting OD600 of 0.05. For compound control wells bacteria-free mGAM was added to the respective wells. Plates were incubated at 37 °C for 24 h, after which they were removed from the anaerobic chamber for sample collection. Whole culture, supernatant and compound control samples were collected and stored at -80 °C until extraction.

PFAS bioaccumulation analysis

Resting cell assay (Fig. 7d,2c; Fig. 12b, 3)

On the day of the screen each well was inoculated 1:1 with a culture in PBS to reach a starting OD600 of 3.75. For compound control wells bacteria-free PBS was added to the respective wells. Samples were incubated at 37 °C for 4 h, after which they were removed from the anaerobic chamber for sample collection. Whole culture, supernatant and compound control samples were collected and stored at -80 °C until extraction.

Growth assay (Fig. 7g; Fig. 12a)

On the day of the screen each well was inoculated 1:1 with a second passage culture to reach a starting OD600 of 0.05. For compound control wells bacteria-free mGAM was added to the respective wells. Samples were incubated at 37 °C for 24 h, after which they were removed from the anaerobic chamber for sample collection. Whole culture, supernatant and compound control samples were collected and stored at -80 °C until extraction. PFAS time-course experiments

PFAS time-course experiment with growing B. uniformis culture (Fig. 7e)

On the day of the screen each tube was inoculated 1:1 with a second passage culture to reach a starting OD600 of 0.05. Samples were incubated at 37 °C for 11 h. OD600 was measured every hour and supernatant samples for PFNA analysis were collected also every hour and stored at -80 °C until extraction.

PFAS time-course experiment with stationary phase B. uniformis culture (Fig. 7f)

1.5 ml of stationary second passage cultures of B. uniformis or pure mGAM were spiked with 15 pl of 2 mM PFNA in DMSO. Whole culture, supernatant and compound control samples were collected at 0, 15, 30 and 60 min and stored at -80 °C until extraction.

PFAS accumulation in live, heat-inactivated, and lysed bacterial cultures (Fig. 8a, Fig. 14)

On the day of the screen each well was inoculated 1:1 with an alive, heat-inactivated or lysed culture in PBS to reach a starting OD600 of 3.75. Second passage cultures were spun down, and the pellet was resuspended in PBS to an OD600 of 7.5. Each culture was split up into 3 aliquots: alive, heat- inactivated, lysed cultures. Live cultures were used as is. Bacteria were heat-inactivated at 70 °C for 40 min and lysed cultures were additionally freeze-thawed three times and sonicated for 3 min. After adding the respective cultures or bacteria-free PBS to the respective wells the plates were sealed and incubated at 37 °C. After 4 h whole culture, supernatant and compound control samples were collected and stored at -80 °C until extraction.

Bacterial samples

For the 'Community-based screening approach' 70 pl supernatant was extracted with 140 pl of ice- cold methanokacetonitrile (1:1) containing the internal standard (caffeine, ibuprofen) and incubated at 4 °C for 30 min. For the 'Single strain-xenobiotic screen' and all other PFAS bioaccumulation screens 50 pl sample was extracted with 200 pl ice-cold methanokacetonitrile (1:1) containing internal standard (caffeine, ibuprofen) and incubated at 4 °C for 15 min. Plates were then centrifuged at 4 °C and 4000 rpm for 10 min. Supernatants were transferred to 96-well plates for LC-MS analysis. Samples for concentration calibration and bacteria-free compound controls were processed in the same way. Mouse faecal samples

Frozen faecal samples were weighed out into beaded tubes and 250 pl extraction buffer (methanol + 0.05 KOH + 15 pM caffeine) was added. Tubes were then homogenised at 1500 rpm for 10 min followed by centrifugation at 14,000 rpm and 4 °C for 5 min. 20 pl supernatant was added to 80 pl water + 0.1 % formic acid, vortexed, incubated at 4 °C for 15 min and centrifuged at 14,000 rpm and 4 °C for 5 min. The supernatant was transferred to LCMS vials with inserts. Samples for concentration calibration were processed in the same way.

LC-MS/MS (QTOF) xenobiotic measurements

QTOF param eters

In short, LC-MS analysis was performed on an Agilent 1290 Infinity II LC system coupled with an Agilent 6546 LC/Q-TOF (Agilent). The QTOF MS scan was operated in positive or negative scanning mode (30- 1500 m/z), depending on the xenobiotic targeted for measurement. The source parameters were as follows: gas temperature: 200 °C, drying gas: 9 L/min, nebulizer: 20 psi, sheath gas temperature: 400 °C, sheath gas flow: 12 L/min, VCap: 3000 V, nozzle voltage: 0 V, fragmentor: 110 V, skimmer 45 V, Oct RF Vpp: 750 V. The online mass calibration was performed using a reference solution (positive: 121.05 and 922.01 m/z; negative: 112.99 and 1033.99 m/z). Collision energies used were 0 V, 10 V, 20 V, 40 V. The compounds were identified based on their retention time, accurate mass and fragmentation patterns. For all measured compounds pure standards were obtained from Sigma Aldrich (Merck KGaA, Darmstadt, Germany) and used for method development, compound identification and calibration.

Five different LC-methods were applied:

15 min reverse-phase LC-method used with QTOF in positive ionization mode (Fig. 11c)

The separation was performed using a ZORBAX RRHD Eclipse Plus column (C18, 2.1 x 100 mm, 1.8 pm; Agilent) with a ZOBRAX Eclipse Plus (C18, 2.1 x 5 mm, 1.8 pm; Agilent) guard column at 40 °C. The multisampler was kept at a temperature of 4 °C. The injection volume was 1 pL and the flow rate was 0.4 mL/min. The mobile phases consisted of A: water + 0.1 % formic acid + 5 mM ammonium formate; B: methanol + 0.1 % formic acid + 5 mM ammonium formate. The 15 min gradient started with 5 % solvent B, which was increased to 30 % by 1 min and then further increased to 100 % by 7 min and held for 3 min, before returning to 5 % solvent B for a 5 min re-equilibration.

10 min reverse-phase dual-pump LC-method used with QTOF in positive ionization mode (Fig. 7a, c) The separation was performed using two ZORBAX RRHD Eclipse Plus column (C18, 2.1 x 100 mm, 1.8 pm; Agilent) with the ZOBRAX Eclipse Plus (C18, 2.1 x 5 mm, 1.8 pm; Agilent) guard columns at 40 °C. The multisampler was kept at a temperature of 4 °C. The injection volume was 1 pL and the flow rate was 0.4 mL/min. The mobile phases consisted of A: water + 0.1 % formic acid + 5 mM ammonium formate; B: methanol + 0.1 % formic acid + 5 mM ammonium formate. The 10 min gradient started with 5 % solvent B, which was increased to 30 % by 1 min and then further increased to 100 % by 7 min and held for 1.7 min, before returning to 5 % solvent B at 8.8 min, which was held until 10 min. The re-equilibration gradient started with 5 % solvent B, which was then ramped up to 100 % solvent B at 0.1 min and held until 4 min before returning to the starting condition of 5 % solvent B at 4.1 min.

13 min reverse-phase LC-method used with QTOF in negative ionization mode ('Community-based screening approach' (Fig. 11c))

The separation was performed using a ZORBAX RRHD Eclipse Plus column (C18, 2.1 x 100 mm, 1.8 pm; Agilent) with a ZOBRAX Eclipse Plus (C18, 2.1 x 5 mm, 1.8 pm; Agilent) guard columns at 40 °C. The multisampler was kept at a temperature of 4 °C. The injection volume was 1 pL and the flow rate was 0.4 mL/min. The mobile phases consisted of A: water; B: methanol. The 13 min gradient started with 35 % solvent B, which was increased to 100 % by 9 min and held for 1 min, before returning to 35 % solvent B for a 3 min re-equilibration.

10 min reverse-phase dual-pump LC-method used with QTOF in negative ionization mode (Fig. 7a,b,d; Fig. 8a, c; Fig. 12; Fig. 13; Fig. 14; Fig. 16)

The separation was performed using two ZORBAX RRHD Eclipse Plus column (C18, 2.1 x 100 mm, 1.8 pm; Agilent) with the ZOBRAX Eclipse Plus (C18, 2.1 x 5 mm, 1.8 pm; Agilent) guard columns at 40 °C. The multisampler was kept at a temperature of 4 °C. The injection volume was 1 pL and the flow rate was 0.4 mL/min. The mobile phases consisted of A: water + 5 mM ammonium acetate + 0.03 % acetic acid; B: methanol + 5 mM ammonium acetate + 0.03 % acetic acid. The 10 min gradient started with 35 % solvent B, which was increased to 100 % by 7 min and held for 1.7 min, before returning to 35 % solvent B at 8.8 min, which was held until 10 min. The re-equilibration gradient started with 35 % solvent B, which was then ramped up to 95 % solvent B at 0.1 min and held until 4 min before returning to the starting condition of 35 % solvent B at 4.1 min.

2 min reverse-phase LC-method used with QTOF in negative ionization mode (Fig. 9c, d)

The separation was performed using a ZORBAX RRHD Eclipse Plus column (C18, 3.0 x 50 mm, 1.8 pm; Agilent) with a ZOBRAX Eclipse Plus (C18, 2.1 x 5 mm, 1.8 pm; Agilent) guard columns at 40 °C. The multisampler was kept at a temperature of 4 °C. The injection volume was 1 pL and the flow rate was 0.8 mL/min. The mobile phases consisted of A: water + 5 mM ammonium acetate + 0.03 % acetic acid; B: methanol + 5 mM ammonium acetate + 0.03 % acetic acid. The 2 min gradient started with 30 % solvent B, which was increased to 100 % by 0.5 min and held until 1 min, before returning to 30 % solvent B at 1.1 min until 2 min.

LC-MS/MS (QQQ) PFNA measurements

QQQ parameters

In short, LC-MS/MS analysis was performed on an Agilent 1290 Infinity II LC system coupled with an Agilent 6570 LC/TQ (Agilent). The QQQ was operated in Dynamic MRM mode. The source parameters were as follows: gas temperature: 300 °C, gas flow: 910 L/min, nebulizer: 50 psi, sheath gas temperature: 300 °C, sheath gas flow: 11 L/min, VCap: 3500 V (positive mode) or 3000 V (negative mode), nozzle voltage: 2000 V (positive mode) or 500 V (negative mode). For PFNA detection the scan segments were the following: precursor ion: 463, product ions: 418.9 and 294.1, fragmentor: 64 and 80 V, collision energy: 8 V. Pure standards were obtained from Sigma Aldrich (Merck KGaA, Darmstadt, Germany) and used for method development, compound identification and calibration.

2 min reverse-phase LC-method used with QQQ (Fig. 7e,fg)

The separation was performed using a ZORBAX RRHD Eclipse Plus column (C18, 3 x 50 mm, 1.8 pm; Agilent) with a ZOBRAX Eclipse Plus (C18, 2.1 x 5 mm, 1.8 pm; Agilent) guard column at 40 °C. The multisampler was kept at a temperature of 4 °C. The injection volume was 1 pL and the flow rate was 0.8 mL/min. The mobile phases consisted of A: water + 0.1 % formic acid; B: methanol + 0.1 % formic acid. The 2 min gradient started with 30 % solvent B, which was increased to 100 % by 0.5 min and held until 1 min, before returning to 30 % solvent B at 1.05 min and held until 2 min.

10 min reverse-phase LC-method used with QQQ for mouse faecal samples (Fig. 10)

The separation was performed using a ZORBAX RRHD Eclipse Plus column (C18, 2.1 x 100 mm, 1.8 pm; Agilent) with a ZOBRAX Eclipse Plus (C18, 2.1 x 5 mm, 1.8 pm; Agilent) guard column at 40 °C. The multisampler was kept at a temperature of 4 °C. The injection volume was 2 pL and the flow rate was 0.4 mL/min. The mobile phases consisted of A: water + 0.1 % formic acid; B: methanol + 0.1 % formic acid. The 10 min gradient started with 5 % solvent B, which was increased to 90 % by 5 min and further increased to 100 % solvent B by 7 min, before returning to 5 % solvent B at 7.1 min and held until 10 min. LC-MS/MS data analysis

The Agilent MassHunter Qualitative Analysis 10.0 software was used to qualify the selected xenobiotic standards. TIC, EIC and ElC-fragment graphs were extracted for each compound. The Agilent MassHunter TOF Quantitative Analysis (Version 10.1) or Agilent MassHunter QQQ Quantitative Analysis software (Version 10.1) was used to quantify the xenobiotic compounds in each sample.

Calibration curves based on pure compound standards were used to estimate the concentrations of target compounds. Data analysis was performed in RStudio Version 1.3.1093. The median of each sample group was compared to the median of the compound controls and an appropriate reduction was chosen as cut-off, to ensure relevant reduction compared to the compound control distribution ('Community-based screening approach' median reduction > 25 %; 'Single strain-xenobiotic screen' > 20 %, all other PFAS bioaccumulation screens > 20%). In addition, statistical comparison was performed using ttest (two-sided), and p-values were FDR corrected for the number of compounds ('Single strain-xenobiotic screen') tested (when a t-test was performed, the adjusted p-values are given in the respective figures). An adjusted p-value of < 0.05 was considered significant.

In the 'Single strain-xenobiotic screen' bioaccumulation was defined as compound sequestration to at least 20 % and adjusted p<0.05 from the supernatant but not from the whole culture sample, whereas biodegradation was defined as both supernatant and whole culture sample showing more than 20 % compound sequestration and adjusted p<0.05.

PFAS-bacteria growth screens (Fig. 7h, 8d)

Plates (Corning 3795) containing 50 pl mGAM with 2x PFAS (2 % DMSO) concentration were prepared the evening prior and placed into the anaerobic chamber over night to ensure anaerobe conditions for inoculation. On the day of the screen each well was inoculated with 50 pl of a second passage culture to reach a starting QD600 of 0.05. Plates were sealed with a gas-permeable membrane (Breath-Easy, Merck, Cat# Z380059), which was additionally pierced with a syringe to prevent gas build-up. Plates were stacked without lids and incubated at 37 °C for 24 h in a stacker-incubator system (Biostack 4, Agilent BioTek) connected to a plate reader (Epoch 2, Agilent BioTek) to record the QD600 every hour.

Growth curve analysis was performed in RStudio Version 1.3.1093. First, for each growth curve the minimum OD value was set to 0. Then the raw AUC was calculated for each well using 'bayestestR' package and area_under_curve() function. Further processing of growth curves was done by plate. AUC values were normalised by median AUC of all control wells (DMSO controls) on the respective plate to determine percent growth inhibition. Conventional ultrathin-section transmission electron microscopy (Fig. 8e-n)

On the day of the experiment each tube containing 2x concentration of PFAS was inoculated 1:1 with a second passage culture to reach a starting OD600 of 0.05. Samples were incubated at 37 °C for 24 h, after which bacterial cultures were spun down and the supernatant was removed. The bacteria were then fixed with a half Karnovsky fixative as 2.5% glutaraldehyde and 2 % paraformaldehyde in 0.1 M sodium cacodylate buffer (pH 7.4 with NaOH) for a few hours at room temperature. For conventional transmission electron microscopy, the postfixation was performed with a mixture of 1% osmium tetroxide and 1% potassium ferrocyanide in the cacodylate buffer. After stained en bloc with 5% aqueous uranyl acetate solution. The dehydration with a series of ethanol and the resin infiltration were completed for the plastic embedding in TER (TAAB Epoxy Resin). After the polymerization at 65 °C for a few days, the ultrathin-sections (~60 nm) were cut by using an ultramicrotome (Leica EM UCT/UC7/Artos-3D, Vienna Austria), mounted on formvar-carbon films on EM copper grids, and stained with lead citrate. The bacterial ultrastructure was observed using FEI Talos F200C 200kV transmission electron microscope (Thermo Fischer Scientific, Einthoven Netherlands) with Ceta-16M CMOS-based camera (4kx4k pixels under 16bit dynamic range) and JEM-1400 Flash TMP (JEOL Ltd., Tokyo Japan) with TVIPS TemCam-XF416 CMOS (Tietz Video and Image Processing Systems GmbH, Germany) as described in Amelio, I. et al. P Natl Acad Sci USA 117, 15694-15701, (2020).

Adaptive laboratory evolution (Fig. 9; Fig. 15)

B. uniformis, B. thetaiomicron, O. splanchnicus, P. merdae and E. coll BW25113 AtolC were evolved for 20 days in 500 pM PFHpA, 500 pM PFOA, 250 pM PFNA, and 125 pM PFDeA. DMSO was used as a control. For each PFAS compound 4 and for the DMSO controls 8 replicate lineages were evolved in parallel. 2 ml deep-well stock plates containing a lOOx stock of each PFAS in DMSO were prepared prior to the experiment and stored at -80 °C until use. On day 0 each well was inoculated with a second passage culture to reach a starting OD600 of 0.05. On each of the following day 50 pl of grown culture was transferred to a fresh compound plate containing PFAS/DMSO in mGAM. Every 5 days the growth of the strains in presence of PFAS was measured by transferring 100 pl starting culture to a clearbottom plate and measured and analysed it as described in the section 'PFAS-bacteria growth screens'. On day 20 glycerol stocks were prepared from each lineage and stored at -80 °C.

Mouse experiments (Fig. 10)

Animal experiments were approved by the local authorities (Regierungsprasidium Tubingen, H 02/20 G). Germfree C57BL/6J mice were bred in house (Gnotobiotic Mouse Facility, Tubingen). Mice were housed under germfree conditions in flexible film isolators (Zoonlab) and transferred to the Isocage P system (Tecniplast) to perform the experiments. Mice were supplied with autoclaved drinking water and y-irradiated maintenance chow for mice (Altromin) ad libitum. Female (n = 3) and male (n = 15) mice between 5 - 6 weeks were used and animals were randomly assigned to experimental groups. Mice were kept in groups of 3 mice per cage during the experiment. All animals were scored daily for their health status.

Preparation and inoculation of the Com20 bacterial community

The Com20 community included Phocaeicola vulgatus, Bacteroides uniformis, Bacteroides fragilis, Bacteroides thetaiotaomicron, Erysipelatoclostridium ramosum, Agathobacter rectalis, Roseburia intestinalis, Veillonella parvula, Eggerthella lenta, Fusobacterium nucleatum, Enterocloster bolteae, Clostridium perfringens, Lacrimispora saccharolytica, Streptococcus salivarius, Ruminococcus gnavus, Bariatricus comes, Parabacteroides merdae, Streptococcus parasanguinis, Collinsella aerofaciens, and Dorea formicigenerans. The Com20 community was prepared under anaerobic conditions (Coy Laboratory Products Inc., 2 % H2, 12 % CO2, rest N2). Consumables, glassware and media were prereduced at least 2 days before inoculation of bacteria. Each strain was grown in monoculture overnight in 5 ml of their respective growth medium at 37 °C. The next day, bacteria were sub-cultured (1:100) in 5 ml fresh medium and incubated for 16 h at 37 °C, except Eggerthella lenta, which was grown for 2 days. Optical density (OD) at 578 nm was determined and bacteria were mixed together in equal ratios to a total OD of 0.5 (OD of 0.025 of each of the 20 strains) in a final volume of 10 ml. After adding 2.5 ml 50 % glycerol (with a few crystals of palladium black (Sigma-Aldrich)), 200 pl aliquots were prepared in glass vials (2 ml, Supelco, Ref. 29056-U) and directly frozen at -80 °C. Frozen vials were used within 3 months.

For inoculation of germfree mice, cages were transferred to an ISOcage Biosafety Station (IBS) (Tecniplast) through a 2 % Virkon S disinfectant solution (Lanxess) dipping bath. Glycerol stocks of the frozen Com20 community (one per mouse) were kept on dry ice before being thawed during transfer into the IBS. Mixtures were used directly after thawing with a minimal exposure time to oxygen of maximum 3 min. Mice were inoculated by oral gavage (50 pl) and inoculation was repeated after 48 h using the same protocol. The germfree control group was left untreated. The IBS was sterilized with 3 % perchloracetic acid (Wofasteril, Kesla Hygiene AG).

10 days after the second inoculation with Com20, mice were orally gavaged either with PFNA (10 mg/kg in 25 % DMSO) in a volume of 50 pl. Fresh faecal samples were collected before treatment, 3 h, 1 day and 2 days after treatment in sterile weighed 1.5 ml Eppendorf-tubes and immediately frozen at -80 °C. On day 3 after treatment, mice were euthanized by CO ₂ and cervical dislocation, dissected and intestinal contents were taken from colon and the small intestine and collected in the same way. Data analysis and replicates

All data analysis was performed using open-source packages accessed from RStudio (Version 1.3.1093). The Agilent MassHunter TOF Quantitative Analysis (Version 10.1) or Agilent MassHunter QQQ Quantitative Analysis software (Version 10.1) was used to quantify the xenobiotic compounds in each sample. All t-tests are two-sided. Biological replicates refer to different inoculation cultures, while technical replicates refer to experiments starting with the same inoculation culture. None of the data points correspond to repeated measurements of the same sample. Fold changes refer to the ratio of medians.

Example 5

The inventors performed proteomics analysis in Bacteroides uniformis following PFNA (20 uM) and DMSO (control) treatment to determine the difference in proteome induced in response to perfluorononanoic acid (PFNA). Figure 17 and Table 3 show the proteins that are differentially expressed in the PFNA treated cells in comparison to DMSO.

In B. uniformis, 37 proteins were significantly upregulated following PFNA treatment and 19 were significantly downregulated. Of the upregulated proteins more than half were related to plasma membrane and transporter function with the top three upregulated proteins being efflux-pumps (R9I2M9 efflux transporter RND family: 5.39 fold-change; R9I2L8 hydrophobe/amphiphile efflux-1 (HAE1) family RND transporter: 5.03 fold-change; R9I2R1 NodT family efflux transporter: 4.79 foldchange). These results support efflux pump involvement in PFNA bioaccumulation.

The involvement of transporters is further supported by genetic analysis in another bioaccumulating gut bacterium Parabacteroides merdae. Vne inventors used a pool of transposon mutants of P. merdae (hereafter 'transposon library') to identify mutants with increased or decreased fitness due to PFNA exposure. The composition of the transposon library was determined after treatment with PFNA (500 uM) and DMSO (control). Homologues of the three most upregulated proteins (efflux-pumps) in B. uniformis showed reduced fitness in PFNA as compared to DMSO (Figure 18, Table 4). This suggests that lack of these efflux pumps makes strains less resistant to PFNA exposure supporting their role as PFNA exporters.

Together, proteomics and genetic analyses indicate the involvement of efflux pumps in export of PFNA out of the cells, which influences amount of PFNA bioaccumulation. Knock-out of genes encoding for efflux pumps is therefore expected to increase PFNA-accumulation capacity of bacterial cells. Table 3. Proteins that are differentially abundant between B. uniformis treated with PFNA (20 uM) and DMSO in comparison, with a Iog2 abundance ratio of 1 and an adjusted p-value of less than 0.05. Table 4. Proteomics results in B. uniformis compared to homologues in the P. merdae transposon library screen. Accession numbers are UniProt accession numbers.

Example 6

Several bacterial and yeast strains were tested for their ability to accumulate perfluorononanoic acid (PFNA). To enable cross-species comparison, tests were conducted using resting cell assay, i.e., bacterial cells suspended in PBS buffer at a density of OD ₆ oo of circa 3.75. Percentage of total PFNA removed from the supernatant, i.e., cell-free buffer, is shown in Figure 19A. Distribution of PFNA bioaccumulation showed a bimodal distribution (Figure 19B). When plotting the distribution grouped by gram-stain, gram-negative bacteria showed higher PFNA accumulation compared to gram-positive strains (Figure 19C). In addition, PFNA accumulation is linked to phylum. Most bacteria belonging to Actinomycetota, Bacillota and Pseudomonadota showed lower accumulation (< 20 %), while all bacteria belonging to Fusobacteriota and Bacteroidota accumulated more than 20 % PFNA. Using a gaussian mixture model with two compartments, strains were classified into low- or high-PFNA accumulating strains with a cut-off of 25 % PFNA accumulation, resulting in 40 high-accumulating strains (Table 5).

When comparing these results with the results shown for 14 bacterial strains under growth conditions in modified Gifu anaerobic medium (mGAM), a strong positive correlation was found (Pearson rank correlation: r = 0.76, p-value = 0.002; Spearman rank correlation: p = 0.64, p-value = 0.015) (Figure 19E). While there is some variation due to differences in experimental setup this comparison shows the applicability of both screening approaches and highlights that PFNA accumulation can occur in very different environmental settings (PBS = nutrient-devoid environment; mGAM = nutrient-rich environment). Table 5. Bacterial strains classified by PFNA accumulation capacity. Methods

Bacterial strains and in vitro incubation

The microbial strains spanned all major phyla of gut bacteria, including within species differences, several probiotic strains as well as strains isolated from kefir, including three yeasts. The experiments for the kefir strains were performed under aerobic conditions, while all other bacterial experiments were performed in an anaerobic chamber (Coy Laboratory Products) filled with 2 % hydrogen and 12 % carbon dioxide in nitrogen. The chamber was equipped with a palladium catalyst system for oxygen removal, a dehumidifier, and a hydrogen sulfide removal system. Bacteria were grown at 37 °C in either mGAM, MRS or YPD medium. Bacteria/yeasts for starting cultures were grown for one or two days (depending on their growth rate) in 10ml of growth media in 15 ml plastic tubes, which were inoculated directly from frozen glycerol stocks. Cultures were then diluted 100-fold and incubated again for the same amount of time before starting the experiments. The screening plates with 40 uM PFNA in PBS (2 % DMSO) were placed into the chamber overnight to ensure anaerobic conditions for inoculation. Inoculation was performed 1:1 with a bacterial culture in PBS to reach an OD of 3.75 and PFNA concentration of 20 uM. Plates were sealed with AlumaSeal II film (A2350-100EA) to avoid evaporation during incubation. Plates were incubated at 37 °C for 4 h, after which they were removed from the anaerobic chamber for sample collection. Whole culture, supernatant and compound control samples were collected and stored at -80 °C until extraction.

Sample extraction

50 pl sample was extracted with 200 pl ice-cold methanokacetonitrile (1:1) containing internal standard (caffeine, ibuprofen) and incubated at 4 °C for 15 min. Plates were then centrifuged at 4 °C and 4000 rpm for 10 min. Supernatants were transferred to 96-well plates for LC-MS analysis. Samples for concentration calibration and bacteria-free compound controls were processed in the same way.

QTOF parameters (Batch 1-2)

In short, LC-MS analysis was performed on an Agilent 1290 Infinity II LC system coupled with an Agilent 6546 LC/Q-TOF (Agilent). The QTOF MS scan was operated in negative scanning mode (30-1500 m/z). The source parameters were as follows: gas temperature: 200 °C, drying gas: 9 L/min, nebulizer: 20 psi, sheath gas temperature: 400 °C, sheath gas flow: 12 L/min, VCap: 3000 V, nozzle voltage: 0 V, fragmentor: 110 V, skimmer 45 V, Oct RF Vpp: 750 V. The online mass calibration was performed using a reference solution (112.99 and 1033.99 m/z). Collision energies used were 0 V, 10 V, 20 V, 40 V. The compounds were identified based on their retention time, accurate mass and fragmentation patterns. For all measured compounds pure standards were obtained from Sigma Aldrich (Merck KGaA, Darmstadt, Germany) and used for method development, compound identification and calibration.

10 min reverse-phase dual-pump LC-method used with QTOF (Batch 1-2)

The separation was performed using two ZORBAX RRHD Eclipse Plus column (C18, 2.1 x 100 mm, 1.8 pm; Agilent) with the ZOBRAX Eclipse Plus (C18, 2.1 x 5 mm, 1.8 pm; Agilent) guard columns at 40 °C. The multisampler was kept at a temperature of 4 °C. The injection volume was 1 pL and the flow rate was 0.4 mL/min. The mobile phases consisted of A: water + 5 mM ammonium acetate + 0.03 % acetic acid; B: methanol + 5 mM ammonium acetate + 0.03 % acetic acid. The 10 min gradient started with 35 % solvent B, which was increased to 100 % by 7 min and held for 1.7 min, before returning to 35 % solvent B at 8.8 min, which was held until 10 min. The re-equilibration gradient started with 35 % solvent B, which was then ramped up to 95 % solvent B at 0.1 min and held until 4 min before returning to the starting condition of 35 % solvent B at 4.1 min.

QQQ parameters (Batch 3-8)

In short, LC-MS/MS analysis was performed on an Agilent 1290 Infinity II LC system coupled with an Agilent 6570 LC/TQ (Agilent). The QQQ was operated in Dynamic MRM mode. The source parameters were as follows: gas temperature: 300 °C, gas flow: 910 L/min, nebulizer: 50 psi, sheath gas temperature: 300 °C, sheath gas flow: 11 L/min, VCap: 3500 V (positive mode) or 3000 V (negative mode), nozzle voltage: 2000 V (positive mode) or 500 V (negative mode). For PFNA detection the scan segments were the following: precursor ion: 463, product ions: 418.9 and 294.1, fragmentor: 64 and 80 V, collision energy: 8 V. Pure standards were obtained from Sigma Aldrich (Merck KGaA, Darmstadt, Germany) and used for method development, compound identification and calibration.

2 min reverse-phase LC-method used with QQQ (Batch 3-8)

The separation was performed using a ZORBAX RRHD Eclipse Plus column (C18, 3 x 50 mm, 1.8 pm; Agilent) with a ZOBRAX Eclipse Plus (C18, 2.1 x 5 mm, 1.8 pm; Agilent) guard column at 40 °C. The multisampler was kept at a temperature of 4 °C. The injection volume was 1 pL and the flow rate was 0.8 mL/min. The mobile phases consisted of A: water + 0.1 % formic acid; B: methanol + 0.1 % formic acid. The 2 min gradient started with 30 % solvent B, which was increased to 100 % by 0.5 min and held until 1 min, before returning to 30 % solvent B at 1.05 min and held until 2 min.

LC-MS/MS data analysis

The Agilent MassHunter Qualitative Analysis 10.0 software was used to qualify the selected xenobiotic standards. TIC, EIC and ElC-fragment graphs were extracted for each compound. The Agilent MassHunter TOF Quantitative Analysis (Version 10.1) or Agilent MassHunter QQQ Quantitative Analysis software (Version 10.1) was used to quantify the xenobiotic compounds in each sample.

Calibration curves based on pure compound standards were used to estimate the linearity of concentrations of target compounds within each batch. Data analysis was performed in RStudio Version 1.3.1093. The median of each sample group was compared to the median of the compound controls within each batch and the percent in PFNA-accumulation was calculated. In addition, statistical comparison was performed using t-test (two-sided), and p-values were FDR corrected for the number of strains tested. An adjusted p-value of < 0.05 was considered significant. A gaussian mixture model with 2 compartments was used to determine whether a strain was high- or low- accumulating.