The present invention relates to bacterial metabolites, and in particular those which may be conjugated to sulfur-containing moieties by a mammal, for use in methods of classifying individuals. It also relates to assessing cresol metabolites and sulfur concentrations for use in methods of classifying individuals.

The listing or discussion of a prior-published document in this specification should not necessarily be taken as an acknowledgement that the document is part of the state of the art or is common general knowledge.

Information-rich data that reflects the functional status of a complex biological system resides in the quantitative and qualitative pattern of metabolites in body fluids. Aspects of the biochemical composition of intracellular fluids are reflected in the extracellular tissue fluid and in the circulating blood that contacts tissue. Alterations in blood composition may in turn translate into altered urinary composition. Thus, abnormal cellular metabolic processes are likely to be reflected in altered composition of biofluids, such that these fluids provide diagnostic insights into the state of the body. Further, the measurement of metabolites has proved to be an important tool in a number of areas including drug development, early disease detection, patient stratification for treatment, and information on disease processes.

Surprisingly and unexpectedly, the inventors have now found that levels of urinary para- cresol sulfate (PCS), are correlated to the metabolic fate of the analgesic acetaminophen, and are also a marker for autism in children. Para-cresol is produced from tyrosine through the action of colonic bacteria, and is almost entirely converted to PCS in man by the sulfotransferase, SULT1A1 (Morinaga et a/, 2004, Legal Medicine 6:32-40; Gamage et al, 2006, Toxicol Sci 90:5-22). Thus, the inventors' observations reveal the importance of sulfur metabolism, and in particular the effect of gut bacteria on sulfur metabolism, in drug pharmacokinetics and disease processes.

The inventors therefore consider that assessing one or more of bacterial metabolites conjugated to a sulfur-containing moiety, cresol metabolites and sulfur concentrations in an individual, may be used to classify individuals, for example according to disease status or the ability to metabolise a particular drug. Jackson et a/ (Arch Biochem Biophys 2003, 418 119-124) and Cao et al {Arch Biochem Biophys 2000, 377 9-21) have identified PCS as the dominant component of urinary myelin basic like material (MBPLM), a known marker of progressive multiple sclerosis However, the origin of PCS and its role in the pathobiology in multiple sclerosis are said to be unclear

A first aspect of the invention provides a method for classifying an individual, the method comprising assessing any one or more of (a) a bacterial metabolite conjugated to a sulfur-containing moiety (b) a cresol metabolite and (c) overall sulfur concentration and/or distribution of sulfur atoms in molecules, in a sample taken from the individual, wherein when assessing para-cresol sulfate alone the individual is not classified according to whether or not the individual has multiple sclerosis, when assessing hippurate and/or 4-hydroxyhιppurate alone the individual is not classified according to whether or not the individual has autism spectrum disorder, and when assessing para- cresol the individual is not classified according to whether or not the individual has hyperactivity

The individual may be a human or mammalian individual, such as a horse, dog, pig, cow, sheep, rat, mouse, guinea pig or primate Preferably, the individual is a human individual

By 'classifying an individual ¹ , we include the meaning that the individual is classified as having a particular phenotype For example, the individual may be classified as one who has a particular ability to conjugate a molecule with a sulfur-containing moiety, or the individual may be classified according to whether it is appropriate to administer to the individual a molecule that undergoes conjugation to a sulfur-containing moiety in an individual either directly or indirectly Alternatively, the individual may be classified as one who has, or has an increased risk of developing, a pathological condition or the individual may be classified according to the degree to which an exogenous molecule will be conjugated to a sulfur-containing moiety either directly or indirectly in the individual As will become clear below, the method may conveniently be used to classify an individual as one who has an increased risk of developing an autism spectrum disorder (ASD) Other classifications may include classifying an individual according to whether or not the individual is likely to respond to a particular therapy, or according to whether or not the individual is suitable for participation in a clinical trial of a particular therapy, or according to whether or not a particular nutritional intervention is suitable for the individual; or according to how well the individual is responding to a treatment. It is appreciated that such classifications may be used to direct a subject to a therapy to which the patient is likely to respond; select a patient to participate in a clinical trial of a therapy; exclude a patient from receiving a therapy to which the patient is unlikely to respond, or likely to respond negatively; exclude a patient from participating in a clinical trial; direct a patient towards a nutritional intervention; or monitor a patient's response to an intervention (e.g. following PCS levels to monitor efficacy of Clostridia-directed therapy). It is appreciated that the individual may be classified as having any particular phenotype for which assessing any one or more of (a) a bacterial metabolite conjugated to a sulfur- containing moiety, (b) a cresol metabolite, and (c) overall sulfur concentration and/or distribution of sulfur atoms in molecules is predictive. Thus, any phenotype whose class representation variables are correlated with any one or more of (a) a bacterial metabolite conjugated to a sulfur-containing moiety, (b) a cresol metabolite and (c) overall sulfur concentration and/or distribution of sulfur atoms in molecules, may be characterised by assessing the respective any one or more of (a) a bacterial metabolite conjugated to a sulfur-containing moiety, (b) a cresol metabolite, and (c) overall sulfur concentration and/or distribution of sulfur atoms in molecules. Methods for identifying such phenotypes are standard practice in the art and are described in, for example, US 7,373,256 and WO 03/107270.

Examples of phenotypes for which assessing any one or more of (a) a bacterial metabolite conjugated to a sulfur-containing moiety, (b) a cresol metabolite, and (c) overall sulfur concentration and/or distribution of sulfur atoms in molecules is predictive, may include a medically relevant state or outcome such as the risk of developing a particular disease or condition, or the diagnosis of a particular existing disease or condition, or the prognosis of a particular existing disease or condition, or an individual's response to an intervention (e.g. drug, diet, etc).

It is also appreciated that by assessing any one or more of (a) a bacterial metabolite conjugated to a sulfur-containing moiety, (b) a cresol metabolite, and (c) overall sulfur concentration and/or distribution of sulfur atoms in molecules, in a sample taken from the individual, the individual may be classified as having any one or more particular phenotypes. The inventors believe that bacterial metabolites conjugated to a sulfur- containing moiety, cresol metabolites and overall sulfur concentration and/or distribution of sulfur atoms in molecules are correlated to a broad array of phenotypes such as disease status and metabolic fate of drugs, and so assessing such metabolites can provide a wealth of information about an individual.

For the avoidance of doubt, by assessing any one or more of (a) a bacterial metabolite conjugated to a sulfur-containing moiety, (b) a cresol metabolite, and (c) overall sulfur concentration and/or distribution of sulfur atoms in molecules, we include assessing any two or all three of (a) a bacterial metabolite conjugated to a sulfur-containing moiety, (b) a cresol metabolite, and (c) overall sulfur concentration and/or distribution of sulfur atoms in molecules.

The bacterial metabolite conjugated to a sulfur-containing moiety may be any bacterial metabolite that can be conjugated to a sulfur-containing moiety. For instance, the metabolite may be an aromatic (eg a phenolic) metabolite or a heterocyclic metabolite. Typically, the metabolite has a hydroxyl group or an N-oxide group to which a sulfur- containing moiety can be conjugated. Examples of bacterial hydroxylated aromatic molecules that may be conjugated to a sulfur-containing moiety include 4- hydroxybenzoic acid, 4-hydroxylactic acid, 4-hydroxyphenylacetic acid and 3- hydroxyphenylpropionic acid. Other examples may include a phenyl derivative (eg a phenyl such as p-cresol), a flavone (eg an equol such as methyl equol) and an indole derivative (eg an indoxyl) (Wikoff et a/, PNAS, March 2009, vol 106 (10), 3698-3703). It is preferred that the bacterial metabolite is a phenolic metabolite, particularly a cresol.

By 'conjugated' we include the meaning that the metabolite is attached to a sulfur- containing moiety through a covalent bond. Typically, the moiety is attached to the metabolite via a sulfur atom.

By 'sulfur-containing moiety' we include the meaning of a molecular moiety that comprises a sulfur atom, such as a sulfate moiety, a sulfonate moiety, a glutathione moiety, an N-acetyl cysteine moiety, a sulfide moiety, a sulfoxide moiety, a sulfone moiety, a sulfite moiety or a taurine moiety. Other less-preferred sulfur-containing moieties include thiosulfates and thiocyanates. It is preferred that the sulfur-containing moiety is a sulfate moiety or a sulfonate moiety.

It is appreciated that the bacterial metabolite conjugated to a sulfur-containing moiety may be the direct product of the mammalian host conjugating a bacterial metabolite to a sulfur-containing moiety or an indirect product, wherein a bacterial metabolite that has been conjugated to a sulfur-containing moiety in a mammal is further modified either by the mammalian host or bacteria.

Thus, examples of a bacterial metabolite conjugated to a sulfur-containing moiety include para-cresol sulfate (PCS), meta cresol sulfate (MCS), ortho cresol sulfate (OCS), indoxyl sulfate, a tauro-bile acid, phenyl sulfate, equol sulfate and methyl equol sulphate. In a preferred embodiment, the bacterial metabolite conjugated to a sulfur-containing moiety is PCS. Thus, the invention provides a method for classifying an individual, the method comprising assessing PCS in a sample taken from the individual, wherein the individual is not classified according to whether the individual has multiple sclerosis or not when assessing PCS alone.

By 'cresol metabolite' we include the meaning of cresol (eg 4-cresol, 3-cresol or 2- cresol), and any metabolite produced either directly or indirectly from cresol. By 'metabolite' we include the meaning of any substance produced by an enzymatic reaction in a living organism or a living cell system. Thus, a cresol metabolite may be one that has undergone Phase 1 and/or Phase 2 metabolism by the mammal. Phase 1 metabolism adds extra functionality (eg additional OH groups) and Phase 2 metabolism forms chemically bonded conjugates to these functional groups, which may increase water solubility or aid excretion.

It is appreciated that the pathway of cresol metabolism can involve various degrees of oxidation at the cresol methyl (-CH3), for example to give -CH ₂ OH and -COOH, and any such oxidation product is encompassed by the term cresol metabolite. Further, each of these oxidation products can be conjugated by the mammalian system at the OH on the aromatic ring by, for example, sulfate, glucuronic acid or glutathione moieties. Cresol itself may also be conjugated at the OH on the aromatic ring. For example, cresol may be conjugated to glucuronide to form cresol glucuronide, or to glutathione groups to form quinone methides. Where there is an additional OH group (eg when the methyl group has been oxidised to -CH ₂ OH), the product can be conjugated in the same way at the additional OH group. The -COOH group (eg when the cresol methyl is oxidised to give - COOH) can similarly be conjugated, for example as glycine (para-hydroxy-hippurate) or with glucuronic acid. All such modified derivatives of cresol are included in the meaning of cresol metabolite.

It is further appreciated that the aromatic ring of cresol can itself be substituted, typically with another -OH group (eg to give 3,4-hydroxy toluene and other isomers), although other aromatic substitutions are possible and all aromatic substitutions of cresol are included in the definition of a cresol metabolite.

Figure 5 illustrates possible metabolic fates of 4-cresol, and all such modifications are included in the definition of a cresol metabolite.

Thus, the cresol metabolite may be any of 4-cresol, 3-cresol, 2-cresol, p-tolyl sulfate, 4- methylenecyclohexa-2,5-dienone, 4-hydroxybenzoate, hippurate, 4-hydroxyhippurate, 4- cresol sulfate, cresol glucuronide, 4-hydroxybenzoic acid or 4-hydroxybenzoate, and hydroxyphenylacetic acid.

It is understood that the cresol metabolite may be a metabolite that is conjugated to a sulfur-containing moiety, and so includes cresol metabolites such as para-cresol sulfate. Alternatively, the cresol metabolite may be a metabolite that is not conjugated to a sulfur- containing moiety, and so includes cresol metabolites such as para-cresol. The cresol metabolite may also be a bacterial metabolite or a non-bacterial metabolite.

Any suitable method may be used to assess a given bacterial metabolite conjugated to a sulfur-containing moiety or a cresol metabolite in a sample taken from an individual, and it is appreciated that more than one method may be employed. For example, the bacterial metabolite or cresol metabolite may be assessed using NMR spectroscopy and/or mass spectrometry (eg by multiple ion monitoring). In this case, the sample may be fractionated prior to analysis, for example by liquid chromatography. For bacterial metabolites conjugated to a sulfur-containing moiety and for those cresol metabolites comprising a sulfur-containing moiety, liquid-chromatography inductively coupled plasma time of flight mass spectrometry (LC-ICPS/ToF-MS) may be used, wherein the LC- ICPMS provides a profile of all sulfur-containing species (the chromatographic trace being referred to as a sulfatogram) and the ToF-MS is used to identify them. The use of ¹ H NMR to assess PCS is described in Example 1 , and the measurement of metabolites using NMR or mass spectrometry is discussed in US patent no. US 7,373,256 and in WO 03/107270. Additionally or alternatively, the bacterial metabolite conjugated to a sulfur- containing moiety or cresol metabolite may be assessed by making use of a binding partner that binds selectively to the metabolite, or by making use of an enzyme linked assay (eg enzyme linked immunosorbent assay, ELISA or an assay in which the conjugated bacterial metabolite or the cresol metabolite is converted (either directly or indirectly) into a molecule which can readily be detected) that can be used to quantify the metabolite. By 'overall sulfur concentration' in the sample taken from the individual we include the meaning of the total concentration of sulfur atoms in the sample. The total concentration of sulfur atoms in a sample can be measured by any suitable method in the art including LC-ICPS/ToF-MS as mentioned above and as described in Corcoran et al (Rapid Commun in Mass Spec 14: 2377-2384 (2000)), X-ray fluorescence and atomic absorption spectroscopy. Other techniques for quantitative sulfur analysis include those based on UV-fluorescence, thermal combustion or inductively coupled plasma, as used routinely in the oil and mineral industries

By assessing 'distribution of sulfur atoms in molecules' in the sample taken from the individual, we include the meaning of assessing whether or the extent to which molecules in the sample taken from the individual are conjugated to a sulfur-containing moiety.

Thus, it is appreciated that assessing a distribution of sulfur atoms in molecules may involve assessing the degree to which a molecule that may be conjugated to a sulfur- containing moiety, is actually conjugated to a sulfur-containing moiety. For example, the concentration of para-cresol sulfate relative to para-cresol may be assessed.

Some molecules may be conjugated to more than one sulfur-containing moiety, and so also included in assessing a distribution of sulfur atoms in molecules, is the extent to which a given molecule is conjugated to sulfur-containing moieties. For example, the proportions of such molecules that are conjugated to one sulfur-containing moiety, two sulfur-containing moieties, and so on, may be assessed.

Further, assessing a distribution of sulfur atoms in molecules may involve assessing the distribution of sulfur atoms between two or more different molecules that may be, or are, conjugated to a sulfur-containing moiety.

By assessing 'distribution of sulfur atoms in molecules' in the sample taken from the individual, we also include the meaning of assessing the distribution of sulfur atoms in molecules of which one or more sulfur atoms form an integral part (eg glutathione, methionine, and taurine). Thus assessing the distribution of sulfur atoms in molecules may comprise assessing the concentration of one or more molecules in which one or more sulfur atoms form an integral part. It is also appreciated that assessing the distribution of sulfur atoms in molecules may involve assessing the pattern of sulfur atom distribution in either all, or a subset of, molecules that contain a sulfur atom (eg molecules that are conjugated to a sulfur- containing moiety such as a drug sulfate, a bile acid sulfate and a carbohydrate sulfate, or molecules in which a sulfur atom is an integral part such as glutathione, taurine and methionine) in a sample taken from the individual. In this way, the identity of the molecules that contain the sulfur atom need not be known, but the distribution of sulfur atoms in such molecules can be determined using any sulfur detection techniques known in the art such as LC-ICPS/ToF-MS described above.

For the avoidance of doubt, either or both of the overall sulfur concentration and distribution of sulfur atoms between molecules may be assessed in a sample taken from the individual. The sample taken from the individual may be any suitable sample. For example, the sample may be any of urine, blood, blood plasma, blood serum, saliva, sweat, tears, breath, breath condensate, fecal sample, cerebrospinal fluid, a tissue extract or hair. Preferably, the sample is a urine sample. When the sample is a urine sample, assessing the overall sulfur concentration or distribution of sulfur atoms in molecules may involve assessing the excretion of sulfur containing compounds. For example, an individual's sulfate excretion may be measured, for instance by measuring excretion of sulfate/sulphite/thiosulfate/thiocyanate in urine. Conveniently, assessing whether individual bel ongs to a particular class, comprises comparing the amount of any one or more of (a) the bacterial metabolite conjugated to a sulfur-containing moiety, (b) a cresol metabolite and (c) overall sulfur concentration and/or distribution of sulfur atoms in molecules in the sample to the amount of any one or more of a respective (a) bacterial metabolite conjugated to a sulfur-containing moiety, (b) a cresol metabolite and (c) overall sulfur concentration and/or distribution of sulfur atoms in molecules in a standard sample obtained from one or more individuals of known class membership. For example, to classify an individual according to ability to conjugate a molecule with a sulfur-containing moiety, the amount of any one or more of (a) bacterial metabolite conjugated to a sulfur-containing moiety, (b) a cresol metabolite and (c) overall sulfur concentration and/or distribution of sulfur atoms in molecules in the sample taken from the individual may be compared to the amount of any one or more of a respective (a) a bacterial metabolite conjugated to a sulfur-containing moiety, (b) a cresol metabolite and (c) overall sulfur concentration and/or distribution of sulfur atoms in molecules in a standard sample obtained from one or more individuals with known ability to conjugate a molecule with a sulfur-containing moiety. In an embodiment, in addition to assessing any one or more of (a) a bacterial metabolite conjugated to a sulfur-containing moiety (b) a cresol metabolite and (c) overall sulfur concentration and/or distribution of sulfur atoms in molecules in a sample taken from the individual, the method comprises assessing at least one further biological parameter. The at least one further biological parameter may be assessed either in the same sample or in a different sample taken from the individual. Further biological parameters that may be assessed include any one or more of genetic information, proteomic information, and metabolomic information. Thus, the at least one further biological parameter may be any of the sequence of a particular region of a chromosome (eg a genetic defect), the status of a specific protein (eg interleukin-13 or cysteine deoxygenase) or the status of a particular metabolite or a combination thereof. Additionally or alternatively, the at least one further biological parameter may be a phenotypic trait of an individual such as behaviour. In any case, it will be appreciated that the outcome of the assessment may depend on both the assessment of any one or more of (a) a bacterial metabolite conjugated to a sulfur-containing moiety, (b) a cresol metabolite and (c) overall sulfur concentration and/or distribution of sulfur atoms in molecules, and on the assessment of the at least one further biological parameter.

Thus, in one embodiment the at least one further biological parameter to be assessed comprises the status of a gene encoding factors related to sulfation, or sulfation deficiency. Analysis of the status of the gene applies to any attributes inherent in a gene, for instance its nucleotide sequence, chromosomal position, or copy number. Variations of these factors may expose differences from a sequence normal to a population, specifically a single nucleotide polymorphism (SNPs), multiple site polymorphisms or other mutations, differences in normal gene copy number (copy number variations, CNVs), and variance in epigenetic regulation, including factors regulating expression such as DNA methylation. In a preferred embodiment, the analysis of the gene measures an SNP that relates to a defect in the gene-encoded factor. The at least one further biological parameter may be known to be discriminatory for a particular class of individual (eg ability to conjugate a molecule with a sulfur-containing moiety or risk of developing a pathological condition), such that it would be beneficial to assess at least one further biological parameter in addition to any one or more of (a) the bacterial metabolite conjugated to a sulfur-containing moiety (b) a cresol metabolite and (c) overall sulfur concentration and/or distribution of sulfur atoms in molecules. For example, where the individual is classified according to risk of developing a particular pathological condition, the at least one further biological parameter may be a further risk factor for that condition.

In a preferred embodiment, when assessing whether the individual has, or has an increased risk of developing ASD (eg by assessing PCS), the at least one further biological parameter is an abnormal SNP that signals a defect in the intestinal NaS1 transporter.

In another embodiment, the at least one further biological parameter is a transferase detoxification enzyme, specifically a sulfotransferase or glutathione transferase. In another embodiment, when assessing whether the individual has, or has an increased risk of developing ASD (eg by assessing PCS), the at least one further biological parameter is a polymorphism in a phenol sulfotransferase such as SULT1A1 or SULT1A2. In another embodiment, the at least one further biological parameter is a polymorphism that confers hyper-activity to a competing sulfotransferase or glutathione transferase (by competing transferase it is meant a transferase enzyme that conducts sulfation reactions on metabolites or xenobiotics other than the phenolic compounds of interest of this invention). Examples of said competing sulfotransferases are those for androsterones (SULT2A1), estrogens (SULT1 E1), dopaminergics (SULT1A3), cholesterol (SULT2B1), and tyrosyl protein sulfotransferases (TS-PSTs), among others. Specific glutathione-s- transferases (GSTs) for polymorphism analysis include GST P1 haplotype, and GST Mu1. In another embodiment, the at least one further biological parameter is an enzyme involved in the generation of sulfur-containing compounds that can be used as substrates for Phase Il detoxification reactions. Such enzymes include cysteine dioxygenase, cystathione B-synthase, methionine synthetase, and Organic Anion Transporter B (OATPB).

By phase Il detoxification system or phase Il detoxification reactions, we include the meaning of a group of reactions whereby cells add another substance (e.g. cysteine, glycine or a sulfur molecule) to a toxic chemical or drug, to render it less harmful. This makes the toxin or drug water-soluble, so it can then be excreted from the body via watery fluids such as bile or urine. There are essentially six phase Il detoxification pathways (glutathione conjugation, amino acid conjugation, methylation, sulfation, acetylation, and glucuronidation). For efficient phase Il detoxification, cells require sulfur- containing amino acids such as taurine and cysteine. If the phase Il detoxification systems are not working adequately, the ensuing toxic intermediates can cause substantial damage. In another embodiment when assessing whether the individual has, or has an increased risk of developing ASD (eg by assessing PCS), the at least one further biological parameter is a gene known in the art to be associated with ASD such as a neuronal cell adhesion/synapse formation marker (PCDH10, CDH10, CDH9, NRXN1 , CNTN4), a copy number variation on NLGN1, ASTN2, a factor in the ubiquitin pathway displaying copy number variations (for example, UB3A, PARK2, RFWD2, FBXO40), dopamine B- hydroxylase, or MET kinase. Alternatively, the at least one further biological parameter may include genes correlated to pathologies associated with ASD such as Rett syndrome (genes MECP2 and CDKL5), Angelman syndrome (genes SLC9A6 and UBE3A), and Fragile X (gene FMR4).

The at least one further biological parameter may be a further bacterial metabolite conjugated to a sulfur-containing moiety or a further cresol metabolite. Thus, the method may comprise assessing more than one bacterial metabolite conjugated to a sulfur- containing moiety (eg two, three, four or five) in a sample taken from the individual and/or more than one cresol metabolite (eg two, three, four or five). Assessing more than one bacterial metabolite conjugated to a sulfur-containing moiety and/or more than one cresol metabolite may provide for more powerful discriminatory models for classifying individuals, than when only one such metabolite is assessed. Methods for generating models to classify individuals based on multiple variables that may be used in the context of the present invention, include those described in US patent no. 7,373,256 and WO 03/107270.

Alternatively, the at least one further biological parameter may be an endogenous sulfur- containing metabolite such as methionine or taurine or any taurine-conjugated molecule. Assessing endogenous sulfur-containing metabolites will provide further information on the status of sulfur metabolism in the individual and so may provide for better classifications. For example, the at least one further biological parameter may be a sulfur-containing metabolite or metabolite ratio indicative of a subject's sulfation capacity Said metabolites and metabolite ratios may comprise sulfate, sulfite, thiosulfate, thiocyanate, transulfurated androgens, plasma glutathione (GSH)/ glutathione disulfide (GSSG) ratio, cysteine, taurine, sulfate, free sulfate, GSSG, or paracetamol sulfate/paracetamol glucuronide ratio

Alternatively, the at least one further biological parameter may be a sulfate transporter Sulfate transporters are critical to regulation of free sulfate available to cells Defects in said sulfate transporters are correlated with decreased salvage of dietary or systemic sulfate necessary for detoxification and other processes The sulfate transporters measured in this method may be any of the group comprising the renal transporters NaS1 , sat-1 , or CFEX, the intestinal transporters NaS1 Dystrophic Dysplasia Sulfate Transporter (DTDST), or Down Regulated in Adenoma transporter (DRA) , and the mucosal sodium sulfate symporter

In one embodiment, the at least one further biological parameter to be assessed is at least one further metabolite that improves class discrimination when assessed in combination with any one or more of (a) a bacterial metabolite conjugated to a sulfur- containing moiety, (b) a cresol metabolite and (c) overall sulfur concentration and/or distribution of sulfur atoms in molecules For example, the assessment may comprise assessing any one or more of (a) a bacterial metabolite conjugated to a sulfur-containing moiety, (b) a cresol metabolite and (c) overall sulfur concentration and/or distribution of sulfur atoms in molecules, expressed as a ratio with at least one other metabolite such as any of creatinine, creatine, glycine, hippurate, NMNA, NMND, PAG, succinate and taurine

In one embodiment, the at least one further biological parameter is at least one further metabolite that is assessed in the same or a different sample Samples containing said at least one further biological parameter may be collected in blood (blood, cells, plasma or serum), tissue homogenates, faecal samples, urine samples, hair samples, interstitial fluid, cerebrospinal fluid, synovial fluid, or saliva samples that allow for the preservation of genetic information The preservation of these samples is conducted by common methods in the art, including the use of gentle homogenization and collection, and frozen storage until use The detection and measurement of genetic information is accomplished by any of the techniques common in the art, including RNA or DNA microarrays, fluorescence hybridizations such as fluorescence in situ hybridization (FISH), or genomic sequencing techniques.

It is appreciated that it may be necessary to express the measure of any one or more of (a) a bacterial metabolite conjugated to a sulfur-containing moiety, (b) a cresol metabolite and (c) overall sulfur concentration and/or distribution of sulfur atoms in molecules as a ratio with at least one other metabolite, for example to account for bulk mass differences between samples. In this way, the measure of any one or more of (a) a bacterial metabolite conjugated to a sulfur-containing moiety, (b) a cresol metabolite and (c) overall sulfur concentration and/or distribution of sulfur atoms in molecules is normalised. Preferably, the measure of any one or more of (a) bacterial metabolite conjugated to a sulfur-containing moiety, (b) a cresol metabolite and (c) overall sulfur concentration and/or distribution of sulfur atoms in molecules is expressed as a ratio with at least one other metabolite, typically in the same sample, whose excretion is relatively stable (millimoles/day/kg body weight). For example, the excretion of creatinine in urine is relatively constant and so the measure of any one or more of (a) bacterial metabolite conjugated to a sulfur-containing moiety, (b) a cresol metabolite and (c) overall sulfur concentration and/or distribution of sulfur atoms in molecules may be expressed as a ratio with creatinine.

Other ways of normalising to account for changes in dilution of samples that may be used in the present invention include lyophilisation and reconstitution into a specific volume, or, for spectroscopic data, normalising to the total sum of the residual spectra. In another embodiment, the at least one further biological parameter is at least one protein assessed in the same or different sample. For example, interleukin-13 or cysteine deoxygenase may be assessed in a sample taken from the individual.

A second aspect of the invention provides a use of a means for assessing any one or more of (a) a bacterial metabolite conjugated to a sulfur-containing moiety, (b) a cresol metabolite and (c) overall sulfur concentration and/or distribution of sulfur atoms in molecules in a sample taken from an individual, in classifying an individual, wherein when assessing para-cresol sulfate alone the individual is not classified according to whether the individual has multiple sclerosis or not, when assessing hippurate and/or 4- hydroxyhippurate alone the individual is not classified according to whether the individual has autism spectrum disorder or not, and when assessing para-cresol alone the individual is not classified according to whether the individual has hyperactivity or not. Preferences for assessing any one or more of (a) a bacterial metabolite conjugated to a sulfur-containing moiety, (b) a cresol metabolite and (c) overall sulfur concentration and/or distribution of sulfur atoms in molecules; the sulfur-containing moiety, the sample taken from the individual and for classifying the individual, for the second and all subsequent aspects of the invention, are described above with respect to the first aspect of the invention.

In a preferred embodiment, the use of a means for assessing a bacterial metabolite conjugated to a sulfur-containing moiety in a sample taken from an individual is for assessing PCS. Thus, the invention provides the use of a means for assessing PCS in classifying an individual, wherein the individual is not classified according to whether the individual has multiple sclerosis or not. Any suitable means may be used to assess any one or more of (a) a bacterial metabolite conjugated to a sulfur-containing moiety, (b) a cresol metabolite and (c) overall sulfur concentration and/or distribution of sulfur atoms in molecules in a sample taken from an individual. For example, the means may be a NMR spectrometer or a mass spectrometer arranged to detect the bacterial metabolite conjugated to a sulfur- containing moiety or cresol metabolite in a sample. Example 1 describes the use of an NMR spectrometer to detect and quantify PCS in a sample. Similarly, a mass spectrometer may be used in multiple ion monitoring mode, such that only certain ion fragments are entered into the instrument and detected by the mass spectrometer. Thus, if assessing PCS, the mass spectrometer can be programmed to detect only those ion fragments derived from PCS. It is appreciated that to confirm NMR and mass spectrometry assignments, the spectra may also be compared to known reference standards of the particular bacterial metabolite conjugated to a sulfur-containing moiety of interest. Means for assessing overall sulfur concentration and/or distribution of sulfur atoms in molecules include any suitable sulfur detection means such as LC-ICPS/ToF- MS described above.

Other means of assessing a bacterial metabolite conjugated to a sulfur-containing moiety or a cresol metabolite in a sample taken from an individual include an enzyme linked assay (eg ELISA or an assay in which the conjugated bacterial metabolite or the cresol metabolite is converted (either directly or indirectly) into a molecule which can be readily detected) or an agent that binds to a bacterial metabolite conjugated to a sulfur- containing moiety or a cresol metabolite. For example, the bacterial metabolite conjugated to a sulfur-containing moiety or cresol metabolite may be a substrate in an enzymatic reaction, such that an enzyme linked assay can be used to quantify the metabolite Enzyme linked assays are standard practice in the art and typically involve colorimetric, fluorescent, or chemiluminescent detection An agent that binds to a bacterial metabolite conjugated to a sulfur-containing moiety or cresol metabolite may be used to assess the metabolite by measuring the binding between the agent and the metabolite Conveniently, the agent is detectably labelled so that the presence of the metabolite can readily be detected Examples of labels include peptide labels, chemical labels, fluorescent labels or radio labels

As discussed in Example 1 , the inventors have connected pre-dose PCS with the metabolic fate of acetaminophen Specifically, the inventors found a high pre-dose level of PCS was associated with a low acetoaminophen sulfate (S)/acetoamιnophen glucuronide (G) ratio Since para-cresol and acetaminophen are structural analogues (Figure 6) and are substrates for the same human cytosolic sulfotransferase, SULT1A1 and can therefore compete for enzyme binding sites as well as for the sulfonate donor 3'- phosphoadenosme 5'-phosphosulfate (PAPS), without wishing to be bound by any theory the inventors believe that an individual's capacity to sulfonate acetaminophen will be reduced by ongoing presentation of endogenous para-cresol produced by colonic bacteria On this basis, the inventors have realised that the presence of any bacterial metabolite conjugated to a sulfur-containing moiety, or any cresol metabolite may affect the ability of an individual to conjugate a molecule with a sulfur-containing moiety Further, the inventors have rationalised that since PCS is influencing endogenous sulfur metabolism, its impact on overall sulfur concentration and/or distribution of sulfur atoms in molecules, may also affect the ability of an individual to conjugate a molecule with a sulfur-containing moiety For example, sulfate conjugation of drugs, environmental toxins, or dietary compounds which require sulfate conjugation for detoxification, may be impaired by the presence of PCS, thereby resulting in toxicity to the host Accordingly, a third aspect of the invention provides a method for assessing the ability of an individual to conjugate a molecule with a sulfur-containing moiety, the method comprising assessing any one or more of (a) a bacterial metabolite conjugated to a sulfur-containing moiety, (b) a cresol metabolite and (c) overall sulfur concentration and/or distribution of sulfur atoms in molecules, in a sample taken from the individual

In a preferred embodiment, the method comprises assessing PCS in a sample taken from the individual Thus, the invention provides a method for assessing the ability of an individual to conjugate a molecule with a sulfur-containing moiety, the method comprising assessing PCS in a sample taken from the individual.

Many endogenous and exogenous molecules become conjugated to sulfur-containing moieties.

In one embodiment, the molecule is an exogenous molecule. In this way the method may be used to assess the ability of an individual to conjugate a drug or xenobiotic with a sulfur-containing moiety, which is important in drug-induced responses and excretion. Preferably, the exogenous molecule is one that undergoes conjugation to a sulfur- containing moiety in an individual either directly or indirectly. Thus, the molecule may itself be conjugated with a sulfur-containing moiety, or the molecule may be modified first, for example as part of Phase I metabolism, and subsequently conjugated to a sulfur-containing moiety, for example as part of Phase Il metabolism. In any event, the molecule is preferably one that ultimately undergoes conjugation to a sulfur-containing moiety within the individual. For example, the molecule may undergo sulfation or sulfonation in the individual or it may become conjugated to glutathione or N-acetyl cysteine. Examples of exogenous molecules that undergo conjugation to sulfur-containing moieties include those that comprise a hydroxyl moiety such as a phenol moiety, or those that comprise an N-oxide group or those that comprise an hydroxylamine moiety.

Particular examples of what the exogenous molecule may be include any known exogenous substrate of an enzyme that conjugates a sulfur-containing moiety to a molecule. For example, the molecule may be any of the exogenous substrates of sulfotransferases cited in Gamage et al, 2006 (Toxicol Sc/, 90(1):5-22; see for example

Table 2), Kauffman, 2004 (Drug Metabol Rev, 36(3&4):823-843; see for example Table

1) or Klassen and Boles, 1997 (The FASEB Jour, 11:404-418), all of which are incorporated herein by reference. Thus, the exogenous molecule may be any of apomorphine; butesonide; ethinylestradiol; monoixidil; tamoxifen; curcumin; a flavonoid; epicatachin; an aliphatic alcohol; a benzylic alcohol; an alkenylbenzene; an aromatic amine or amide; a heterocyclic amine; a polynuclear aromatic hydrocarbon; a perflorocarboxylic acid; manganese; a si mple phenolic compound such as acetaminophen, p-nitrophenol, m-nitrophenol, p-ethylphenol, p-cresol, N-hydroxy-PhlP,

N-hydroxy-2-AAF, 1-hydroxymethylpyrene and 1-naphthol; an estrogen such as 17- ethinyl-E2, equilenin, 2-hydroxyestrone, 2-hydroxyestradiol, 4-hydroxyestrone and 4- hydroxyestradiol; 1-hydroxymethylpyrene; 6-hydroxymethylbenzo[a]-pyrene; and hycanthone. Other examples of exogenous molecules that may be conjugated to sulfur- containing moieties are provided in Drug Metabolism Towards the Next Millennium, Gooderham N J (eds) (see, for example, chapter by Coughtrie et al), incorporated herein by reference.

In another embodiment, the molecule is an endogenous molecule. Thus, the method of the invention may be used to assess the general sulfation status of the individual or to assess the defensive capability of the individual (eg many sulfur-containing compounds such as glutathione function in the body's defence mechanisms).

The endogenous molecule may be any endogenous molecule that is ultimately conjugated to a sulfur-containing moiety in an individual, such as a polypeptide, a lipid, a carbohydrate, a biogenic amine, a hydroxy organic acid or a steroid. It is appreciated that the endogenous molecule may be any substrate of an enzyme that conjugates a sulfur-containing moiety to a molecule. For example, the molecule may be any of the endogenous substrates of sulfotransferases cited in Gamage et al, 2006 (Toxicol Sci, 90(1):5-22; see for example Table 2), Kauffman, 2004 {Drug Metabol Rev, 36(3&4):823- 843; see for example Table 1) or Klassen and Boles, 1997 {The FASEB Jour, 11:404- 418), all of which are incorporated herein by reference. Thus, the endogenous molecule may be any of a catecholamine such as dopamine and norepinephrine; an iodothyronine such as 3,3'diiodothyronine (T2), 3,3',5-triiodothyronine (T3), 3,3',5'-reverse triiodothyronine (r-T3) and thyroxine; ascorbic acid; vitamin D; an estrogen such as β- estradiol (E2) and estrone (E1); dihydroepiandrosterone; an androgen; a steroid such as cholesterol, an oxysterol and DHEA; pregnenolone; and a bile acid. Further examples of endogenous molecules that undergo sulfonation include those cited in Strott, 2002 (Endocrin Rev 23(5):703-732) which is incorporated herein by reference.

Thus, the method of the third aspect of the invention may be used to assess or predict the ability of the individual to sulfonate any of the above listed molecules which are substrates of a sulfotransferase.

In a further embodiment of the method of the third aspect of the invention, the assessing comprises comparing the measure of the any one or more of (a) a bacterial metabolite conjugated to a sulfur-containing moiety, (b) a cresol metabolite and (c) overall sulfur concentration and/or distribution of sulfur atoms in molecules in the sample to the measure of a respective any one of more of (a) bacterial metabolite conjugated to a sulfur-containing moiety, (b) a cresol metabolite and (c) overall sulfur concentration and/or distribution of sulfur atoms in molecules in a standard sample obtained from one or more individuals with known ability to conjugate a molecule with a sulfur-containing moiety Generally, the higher the amount of the bacterial metabolite conjugated to a sulfur-containing moiety or cresol metabolite, the less the ability to conjugate a molecule with a sulfur-containing moiety Similarly, a reduced overall sulfur concentration or a change in the distribution of sulfur atoms in molecules may be indicative of a reduced ability to conjugate a molecule with a sulfur-containing moiety Methods for assessing any one or more of (a) a bacterial metabolite conjugated to a sulfur-containing moiety in a sample, (b) a cresol metabolite and (c) overall sulfur concentration and/or distribution of sulfur atoms in molecules are provided above with respect to the first aspect of the invention

A fourth aspect of the invention provides a use of a means for assessing any one or more of (a) a bacterial metabolite conjugated to a sulfur-containing moiety, (b) a cresol metabolite and (c) overall sulfur concentration and/or distribution of sulfur atoms in molecules, in a sample taken from an individual, in assessing the ability of an individual to conjugate a molecule with a sulfur-containing moiety In a preferred embodiment, the use is a use of a means for assessing PCS Thus, the invention provides the use of a means for assessing PCS in assessing the ability of an individual to conjugate a molecule with a sulfur-containing moiety

Preferences for the means for assessing any one or more of (a) a bacterial metabolite conjugated to a sulfur-containing moiety, (b) a cresol metabolite and (c) overall sulfur concentration and/or distribution of sulfur atoms in molecules in a sample taken from an individual and for the molecule that is conjugated to a sulfur-containing moiety, for the fourth and all subsequent aspects of the invention are defined above with respect to the second and third aspects of the invention

As discussed above and in Example 1 , the inventors believe that an individual's capacity to sulfonate acetaminophen will be reduced by ongoing presentation of endogenous para-cresol On this basis, the inventors have realised that the presence of any bacterial metabolite conjugated to a sulfur-containing moiety, or a cresol metabolite, or overall sulfur concentration and/or distribution of sulfur atoms in molecules may affect the ability of an individual to conjugate a molecule with a sulfur-containing moiety Since conjugation to sulfur-containing moieties (eg sulfonation) is important in the excretion of many compounds including drugs, assessing any one or more of (a) a bacterial metabolite conjugated to a sulfur-containing moiety, (b) a cresol metabolite and (c) overall sulfur concentration and/or distribution of sulfur atoms in molecules may be used to determine whether it is appropriate to administer to an individual a molecule that undergoes conjugation to a sulfur-containing moiety. For example, if an individual has reduced ability to conjugate a molecule such as a drug with a sulfur-containing moiety, excretion of that molecule may be impaired such that it accumulates in the body and becomes toxic. Accordingly, a fifth aspect of the invention provides a method for determining whether it is appropriate to administer to an individual a molecule that undergoes conjugation to a sulfur-containing moiety in an individual either directly or indirectly, the method comprising assessing any one or more of (a) a bacterial metabolite conjugated to a sulfur-containing moiety, (b) a cresol metabolite and (c) overall sulfur concentration and/or distribution of sulfur atoms in molecules in a sample taken from the individual.

In a preferred embodiment, the method comprises assessing PCS in a sample taken from the individual. Thus, the invention provides a method for determining whether it is appropriate to administer to an individual a molecule that undergoes conjugation to a sulfur-containing moiety in an individual either directly or indirectly, the method comprising assessing PCS in a sample taken from the individual.

It is appreciated that the method may be used to determine whether it is appropriate to administer any of the endogenous or exogenous molecules described above, such as one that undergoes sulfation or sulfonation in the individual or one that becomes conjugated to glutathione or N-acetyl cysteine, either directly or indirectly. The molecule may be any of a polypeptide, a lipid, a carbohydrate, a biogenic amine, a hydroxy organic acid, a steroid or a xenobiotic. In a further embodiment of the fifth aspect of the invention, the assessing comprises comparing the measure of any one or more of (a) a bacterial metabolite conjugated to a sulfur-containing moiety, (b) a cresol metabolite and (c) overall sulfur concentration and/or distribution of sulfur atoms in molecules in the sample to the measure of a respective any one or more of (a) a bacterial metabolite conjugated to a sulfur-containing moiety, (b) a cresol metabolite and (c) overall sulfur concentration and/or distribution of sulfur atoms in molecules in a standard sample obtained from one or more individuals for whom it is known to be appropriate or inappropriate to administer a molecule that undergoes conjugation to a sulfur-containing moiety. Generally, the higher the amount of the bacterial metabolite conjugated to a sulfur-containing moiety, or the higher the amount of the cresol metabolite, or a lower overall concentration of sulfur and/or a changed distribution of sulfur atoms in molecules in the sample taken from the individual, the less the ability to conjugate a molecule with a sulfur-containing moiety and the less appropriate it may be to administer the molecule.

A sixth aspect of the invention provides a use of a means for assessing any one or more of (a) a bacterial metabolite conjugated to a sulfur-containing moiety, (b) a cresol metabolite and (c) overall sulfur concentration and/or distribution of sulfur atoms in molecules in a sample taken from an individual in determining whether it is appropriate to administer to an individual a molecule that undergoes conjugation to a sulfur-containing moiety in an individual. In a particularly preferred embodiment, the use is a use of a means for assessing PCS. Thus, the invention provides the use of a means for assessing PCS in determining whether it is appropriate to administer to an individual a molecule that undergoes conjugation to a sulfur-containing moiety in an individual. As discussed above and in Example 1 , the inventors have associated PCS with the ability to conjugate molecules to sulfur-containing moieties within the body. In addition to being important in compound excretion, conjugation of molecules to sulfur-containing moieties is crucial to the structure and properties of macromolecules such as chondroitin sulfate. Further, conjugation to sulfur-containing moieties is known to have a role in modulating the action of hormones and neurotransmitters, and appears to be especially important during early human development. Thus, without wishing to be bound by any theory the inventors believe that assessing any one or more of (a) a bacterial metabolite conjugated to a sulfur-containing moiety, (b) a cresol metabolite and (c) overall sulfur concentration and/or distribution of sulfur atoms in molecules can be used as a marker for perturbed sulfur metabolism and so may be used to determine the risk of an individual developing a pathological condition.

Accordingly, a seventh aspect of the invention provides a method for determining whether an individual has, or has an increased risk of developing a pathological condition, the method comprising assessing any one or more of (a) a bacterial metabolite conjugated to a sulfur-containing moiety, (b) a cresol metabolite and (c) overall sulfur concentration and/or distribution of sulfur atoms in molecules in a sample taken from the individual, wherein when assessing para-cresol sulfate alone the pathological condition is not multiple sclerosis, when assessing hippurate and/or 4-hydroxyhippurate alone the pathological condition is not autism spectrum disorder, and when assessing para-cresol alone the pathological condition is not hyperactivity.

In a preferred embodiment, the method comprises assessing PCS in a sample taken from the individual. Thus, the invention provides a method for determining whether an individual has, or has an increased risk of developing a pathological condition, the method comprising assessing PCS in a sample taken from the individual, wherein the pathological condition is not Multiple Sclerosis.

The pathological condition may be a toxicologically-induced condition, an infection- induced condition, a lifestyle-induced condition, a genetically-induced condition or a degenerative condition. Preferably, the pathological condition is one whose aetiology involves conjugation of a molecule to a sulfur-containing moiety, for example sulfation or sulfonation of a molecule, or conjugation of a molecule to N-acetyl cysteine or glutathione. Since bacterial metabolites conjugated to a sulfur-containing moiety can be used as a measure of an individual's capacity to defend against free radical induced damage or oxidative damage, then it will be appreciated that the pathological condition may be any condition that is related to this capacity, such as atherosclerosis and cancer. Further examples of particular pathological conditions include autism spectrum disorder (ASD), rheumatoid arthritis, pre-eclampsia, post-operative sepsis, gut dysbiosis, cystic fibrosis, obesity, diabetes, Huntington's disease, muscular dystrophy and Alzheimer's disease. Additional conditions and diseases related to sulfation deficiency include schizophrenia, Parkinson's disease, motor neuron disease, cirrhosis of the liver, and migraine headaches.

ASD consists of a continuum or spectrum of complex neurodevelopmental disorders with a serious lifelong impact on individuals from all ethnic and socioeconomic backgrounds (Minschew, 2007 The Summer Institute of Neurodevelopmental Disorders. UC Davis M. I.N, D. Institute, Sacramento, California, August 2-3; Pessah, 2006 Understanding immunological and neurobiologic susceptibilities contributing to autism risk. UC Davis MIND Institute Conference, Sacramento, California, November 2-3; Grandjean & Landigan, 2006 www.thelancet.com Published online November 8, 2006 DOI: 10.1016/S0140-6736 (06) 69665-7; London & Etzel, 2000 Environmental Health Perspectives Supplements 108). Both genetic and environmental factors appear to contribute to the development of ASD (Herbert, 2006 Clinical implications of environmental toxicology for children's neurodevelopment in autism. UC Davis MIND Institute Conference, Sacramento, California, November 2-3; Hertz-Picciotto, Croen, Hansen, Jones, Van de Water, & Pessah, 2006 Environmental Health Perspectives 114, from www.ehponline.org; London & Etzel, 2000 Environmental Health Perspectives Supplements 108). ASD includes five disorders: Autistic Disorder, Pervasive developmental disorder not otherwise specified (PDD, NOS), Asperger's Disorder, Retts Disorder, and Childhood Disintegrative Disorder. Common manifestations include challenges in communication, imaginative play, and socializing with others. Preoccupation with unusual interests and a restricted/repetitive pattern of interests and/or behaviours are typical. Routines, inflexibility, and difficulty with new situations characterize many individuals. Intellectual disabilities are common.

In a preferred embodiment, the pathological condition is an autism spectrum disorder, so the invention includes a method for determining whether an individual has an increased risk of developing an autism spectrum disorder, the method comprising assessing any one or more of (a) a bacterial metabolite conjugated to a sulfur-containing moiety, (b) a cresol metabolite and (c) overall sulfur concentration and/or distribution of sulfur atoms in molecules, wherein the method does not comprise assessing hippurate and/or 4- hydroxyhippurate alone in a sample taken from the individual. Preferably, the method comprises assessing PCS in a sample taken from the individual, and more preferably the PCS to glycine, PCS to NMNA, PCS to NMND or PCS to succinate ratios. It is appreciated that assessing such ratios of PCS in urine samples provides for an effective normalisation of PCS levels to account for bulk mass differences between samples. Since sulfonation is known to be important during early human development, it may be that a reduced capacity to conjugate a molecule to a sulfur-containing moiety is more commonly expressed in pathological conditions early in development. Thus in one embodiment, the individual is a pre-pubescent individual. By pre-pubescent individual we include the meaning of a neonate up to the age at which the brain is fully developed. For example, the individual may be aged 18 years or less, 17 years or less, 16 years or less, 15 years or less, 14 years or less, 13 years or less, 12 years or less, 11 years or less, 10 years or less, 9 years or less, 8 years or less, 7 years or less, 6 years or less, 5 years or less, 4 years or less, 3 years or less, 2 years or less or 1 year or less. In a further embodiment of the seventh aspect of the invention, the assessing comprises comparing the measure of a respective any one or more of (a) a bacterial metabolite conjugated to a sulfur-containing moiety, (b) a cresol metabolite and (c) overall sulfur concentration and/or distribution of sulfur atoms in molecules in the sample to the measure of any one or more of (a) a bacterial metabolite conjugated to a sulfur- containing moiety, (b) a cresol metabolite and (c) overall sulfur concentration and/or distribution of sulfur atoms in molecules in a standard sample obtained from one or more individuals which are known to have or have a known risk of developing a pathological condition. .

An eighth aspect of the invention provides the use of a means for assessing any one or more of (a) a bacterial metabolite conjugated to a sulfur-containing moiety, (b) a cresol metabolite and (c) overall sulfur concentration and/or distribution of sulfur atoms in molecules in a sample taken from an individual in determining whether an individual has, or has an increased risk of developing a pathological condition, wherein when assessing para-cresol sulfate alone the pathological condition is not multiple sclerosis, when assessing hippurate and/or 4-hydroxyhippurate alone the pathological condition is not autism spectrum disorder, and when assessing para-cresol alone the pathological condition is not hyperactivity.

In a particularly preferred embodiment, the means is a means for assessing PCS in a sample taken from an individual and so the invention includes the use of a means for assessing PCS in a sample taken from an individual in determining whether an individual has, or has an increased risk of developing a pathological condition, other than Multiple Sclerosis.

Preferences for the pathological condition and the individual are defined above with respect to the seventh aspect of the invention.

It will be appreciated that once an individual has been identified as having, or having an increased risk of developing a pathological condition by assessing the any one or more of (a) a bacterial metabolite conjugated to a sulfur-containing moiety, (b) a cresol metabolite and (c) overall sulfur concentration and/or distribution of sulfur atoms in molecules as described above, the individual may be amenable to treatment aimed at modifying the capacity to conjugate a molecule to a sulfur-containing moiety.

Accordingly, a ninth aspect of the invention provides a method of combating a pathological condition in an individual, the method comprising:

a) determining whether the individual has, or has an increased risk of developing a pathological condition by assessing any one or more of (a) a bacterial metabolite conjugated to a sulfur-containing moiety, (b) a cresol metabolite and (c) overall sulfur concentration and/or distribution of sulfur atoms in molecules in a sample taken from the individual; and

b) depending upon the outcome of the assessment, administering to the individual an agent that reduces a bacterial metabolite conjugated to a sulfur- containing moiety in the individual or an agent that reduces a cresol metabolite in the individual or an agent that increases the availability of sulfur-containing moieties in the individual, or applying a dietary and/or pharmacological regime to the individual.

By "combating" we include the meaning that the method can be used to alleviate symptoms of the disorder (ie the method is used palliatively), or to treat the disorder, or to prevent the disorder (ie the method is used prophylactic ally). The invention includes the use of an agent that reduces a bacterial metabolite conjugated to a sulfur-containing moiety in an individual or an agent that reduces a cresol metabolite in an individual or an agent that increases the availability of sulfur- containing moieties in an individual, in the manufacture of a medicament for combating a pathological condition in an individual, which individual has been found to have, or have an increased risk of developing a pathological condition by assessing any one or more of (a) a bacterial metabolite conjugated to a sulfur-containing moiety, (b) a cresol metabolite and (c) overall sulfur concentration and/or distribution of sulfur atoms in molecules in a sample taken from the individual. The invention includes an agent that reduces a bacterial metabolite conjugated to a sulfur-containing moiety in an individual or an agent that reduces a cresol metabolite in an individual or an agent that increases the availability of sulfur-containing moieties in an individual for use in combating a pathological condition in an individual, which individual has been found to have, or have an increased risk of developing a pathological condition by assessing any one or more of (a) a bacterial metabolite conjugated to a sulfur- containing moiety, (b) a cresol metabolite and (c) overall sulfur concentration and/or distribution of sulfur atoms in molecules in a sample taken from the individual.

In a preferred embodiment, the pathological condition is an autism spectrum disorder.

Thus, the invention provides a method of combating an autism spectrum disorder in an individual, the method comprising: a) determining whether the individual has, or has an increased risk of developing an autism spectrum disorder by assessing any one or more of (a) a bacterial metabolite conjugated to a sulfur-containing moiety, (b) a cresol metabolite and (c) overall sulfur concentration and/or distribution of sulfur atoms in molecules in a sample taken from the individual, and

b) depending upon the outcome of the assessment, administering to the individual an agent that reduces a bacterial metabolite conjugated to a sulfur- containing moiety in the individual or an agent that reduces a cresol metabolite in the individual or an agent that increases the availability of sulfur-containing moieties in the individual

Similarly, the invention includes the use of an agent that reduces a bacterial metabolite conjugated to a sulfur-containing moiety in an individual or an agent that reduces a cresol metabolite in the individual or an agent that increases the availability of sulfur- containing moieties in an individual, in the manufacture of a medicament for combating an autism spectrum disorder in an individual, which individual has been found to have or have an increased risk of developing ASD by assessing any one or more of (a) a bacterial metabolite conjugated to a sulfur-containing moiety, (b) a cresol metabolite and (c) overall sulfur concentration and/or distribution of sulfur atoms in molecules in a sample taken from the individual

Likewise, the invention includes an agent that reduces a bacterial metabolite conjugated to a sulfur-containing moiety in an individual or an agent that reduces a cresol metabolite in the individual or an agent that increases the availability of sulfur-containing moieties in an individual for use in combating an autism spectrum disorder in an individual, which individual has been found to have an increased risk of developing ASD by assessing any one or more of (a) a bacterial metabolite conjugated to a sulfur-containing moiety, (b) a cresol metabolite and (c) overall sulfur concentration and/or distribution of sulfur atoms in molecules in a sample taken from the individual

In a preferred embodiment, the method comprises assessing PCS in a sample taken from the individual

Thus, the invention also provides a method of combating a pathological condition in an individual, the method comprising a) determining whether the individual has an increased risk of developing a pathological condition by assessing PCS in a sample taken from the individual; and

b) depending upon the outcome of the assessment, administering to the individual an agent that reduces a bacterial metabolite conjugated to a sulfur- containing moiety in the individual or an agent that reduces a cresol metabolite or an agent that increases the availability of sulfur-containing moieties in the individual, or applying a dietary or pharmacological regime to the individual. Similarly, it will be appreciated that the invention includes the use of an agent that reduces a bacterial metabolite conjugated to a sulfur-containing moiety in an individual or an agent that reduces a cresol metabolite in an individual or an agent that increases the availability of sulfur-containing moieties in an individual, in the manufacture of a medicament for combating a pathological condition in an individual, which individual has been found to have an increased risk of developing a pathological condition by assessing PCS in a sample taken from the individual.

Likewise, it will be appreciated that the invention includes an agent that reduces a bacterial metabolite conjugated to a sulfur-containing moiety in an individual or an agent that reduces a cresol metabolite in an individual or an agent that increases the availability of sulfur-containing moieties in an individual for use in combating a pathological condition in an individual, which individual has been found to have an increased risk of developing a pathological condition by assessing PCS in a sample taken from the individual.

Preferences for the pathological condition and the individual are defined above with respect to the seventh aspect of the invention.

It is appreciated that the agent that reduces the amount of a bacterial metabolite conjugated to a sulfur-containing moiety or a cresol metabolite may be one that does so directly or indirectly. Thus the agent may be one that reduces the amount of the bacterial metabolite itself and therefore also the bacterial metabolite conjugated to a sulfur-containing moiety. The agent that reduces the amount of a bacterial metabolite conjugated to a sulfur- containing moiety in the individual may be one that modulates the bacteria producing the metabolite that is conjugated to a sulfur-containing moiety. For example, when the bacterial metabolite conjugated to a sulfur-containing moiety is PCS, the agent that reduces the amount of PCS may be one that modulates the bacteria producing para- cresol. Thus, it is appreciated that the agent that reduces the amount of a bacterial metabolite conjugated to a sulfur-containing moiety may be an antibiotic (which kills the para-cresol containing bacteria) or a probiotic (which change the gut microflora to reduce the proportion of para-cresol producing bacteria).

Alternatively, the agent that reduces the amount of a bacterial metabolite conjugated to a sulfur-containing moiety in the individual may be one that modulates the metabolism of the bacterial metabolite that is conjugated to a sulfur-containing moiety. For example, when the bacterial metabolite conjugated to a sulfur-containing moiety is PCS, the agent that reduces the amount of PCS may be one that decreases production of para-cresol. Para-cresol is produced from tyrosine as illustrated in Figure 4, and so the agent that reduces the amount of a bacterial metabolite conjugated to a sulfur-containing moiety may be a tyrosine oxidase inhibitor. Similarly, the agent that reduces the amount of a cresol metabolite in the individual may be one that modulates the metabolism of the particular cresol metabolite.

It is appreciated that there are various agents that reduce a cresol metabolite in an individual, which agents work by different methods including targeting the cresol (eg para-cresol) biosynthesis pathway, targeting the cresol (eg para-cresol) metabolism pathway, targeting the microbes responsible for producing cresol (eg para-cresol), and directly removing cresol (eg para-cresol) from the host. By 'cresol biosynthesis and metabolism pathway', we include the meaning of the series of reactions occurring within a mammal, such as a human, wherein several chemicals are modified by reactions catalyzed by enzymes, which often require dietary minerals, vitamins, and other cofactors in order to function properly, and which yield cresol molecules as a product, as well as the series of chemical reactions wherein cresol is converted to any of its metabolites, for example, p-cresol sulfate.

Thus in one embodiment, the agent that reduces a cresol metabolite in an individual is one that targets molecules of the cresol biosynthesis and metabolic pathway to reduce levels of a cresol metabolite. Suitable target molecules of the pathway of cresol biosynthesis and metabolism include enzymes, other proteins, co-factors, nucleic acids, or small molecule metabolites that play a role in cresol biosynthesis or metabolism. It is appreciated that, while individual agents are listed in the embodiments herein as a means to exemplify various aspects of targeting molecules of the cresol biosynthesis and metabolism pathways, combinations of multiple agents contemplated herein are additionally useful to elicit the desired therapeutic effect (eg combinations of at least 2, 3, 4 or 5 agents). In one embodiment, the agent is one that inhibits an enzyme in the cresol biosynthetic pathway responsible for production of cresol. For example, without wishing to be bound by any theory, para-cresol can be produced by various routes through the degradation of the amino acid tyrosine or exogenous toxins. It is an object of this invention to inhibit any or all of these routes via administering agents that target specific components of said routes. The routes include, but are not limited to, metabolism of tolerated endogenous metabolites such as tyrosine and tyramine, and degradation of toxins such as toluene.

In a preferred embodiment, the agent targets an enzyme the products of which are in subsequent steps converted to para-cresol. Target enzymes comprise those in the sequential formation of 4-hydroxyphenylacetic acid from tyrosine, through tyramine- independent and -dependent pathways (Lis et al, 1976 Clin. Chem 22; Curtius et al, 1976 J. Chromatogr. 126). The former comprise 2-oxoglutarate aminotransferase (EC 2.6.1.1/2.6.1.9) and 4-hydroxyphenylpyruvate oxidase (EC 1.2.3.13). The latter comprise tyrosine decarboxylase (EC 4.1.1.25), tyramine oxidase (EC 1.4.3.4), and A- hydroxyphenylacetaldehyde dehydrogenase (EC 1.2.1.53). Said enzyme targets catalyzing the formation of precursors of cresol (eg para cresol) may be expressed by human cells, by microbes such as Clostridia sp., Pseudomonas sp., or E. coli, which populate a number of anatomical locations of said humans such as the gut, or in certain cases may be expressed by both human and microbial cells. Suitable inhibitors of said targets include compounds that decrease the target enzyme activity, either in an in vitro enzymatic assay or cellular system. Additionally, inhibitors can exert an effect either by direct inhibition or indirect inhibition of enzyme activity. Suitable inhibitors are also those known in the art to exert specific inhibition of enzyme activity. Examples of 2- oxoglutarate aminotransferase and tyrosine transaminase inhibitors include A- methylsulfonyl biphenyl derivatives, ketoconazole, tolyfluanid, phenelzine, aspartate, and glutamate (Johansson et al, 2005 Basic Clin. Pharmacol. Toxicol. 96; Dyck and Dewar, 1986. J. Neurochem 46; Miller and Litwack, 1971 J. Biol. Chem. 246). Examples of tyrosine decarboxylase inhibitors include lactic acid, citric acid, hydroxylamine, 2- mercaptoethanol, 3-hydroxybenzyl hydrazine, and related derivatives, A- deoxypyroxidone, trihydroxybenzylhydrazines, phenylpropionates, and derivatives thereof, and Monoamine Oxidase (MAO) inhibitors, such as phenelzine, pargyline, and clorgyline (Morena-Arribas and Lonvaud-Funel, 1999 FEMS Microbiol. Lett.180; Kezmarsky et al, 2005 Biochim Biophys Acta 1722, Nagy and Hiripi, 2002 Neυrochem lnt 41 ; Dyck and Dewar, 1986, J Neurochem 46, Cho et al 1996, Eur J Pharmacol 314, Beiler and Martin, 1947 J Biol Chem, David et al, 1974 Proc Nat Acad Sci USA 71 , Chappie et al, 1985 Planta 167)

In a particularly preferred embodiment, the agents inhibit a tyrosine decarboxylase enzyme The agent may be selected from 4-deoxypyroxιdone, a tπhydroxybenzylhydrazine, a phenylpropionate, and derivatives thereof For a characterization of the mechanism and inactivation of tyrosine decarboxylase from specific bacterial and plant species, see Beiler and Martin, 1947 J Biol Chem, David et al, 1974 Proc Nat Acad Sci USA 71, and Chappie et al, 1985 Planta 167

In another embodiment, the enzyme targeted by the agent directly forms cresol (eg para- cresol) as a reaction product, and therefore inhibition of said enzyme directly inhibits cresol (eg para-cresol) production A preferred example of said enzyme is p- hydroxyphenylacetate (pHPA) decarboxylase (EC 4 1 1 83), which directly forms para- cresol from the precursor p-hydroxyphenylacetate, and is present in Clostridia species In a preferred embodiment, the inhibitors used to target pHPA decarboxylase are small molecule competitive inhibitors, including p-hydroxyphenylacetamide, p- hydroxymandelate, or related compounds or derivatives thereof (see Selmer and Andrei, 2001 Eur J Biochem 268) Other suitable forms of inhibition or decrease of the activity of pHPA comprise the use of agents that inhibit the hetero-octamer assembly necessary to activate the enzyme In one embodiment, the agent inhibits the serine protein kinase reaction required for pHPA decarboxylase oligomeπzation and activity A range of serine kinase inhibitors that can selectively inhibit said reactions are exemplified in Thorp et al, 1994 Immunology 81 Additionally, acceptable agents are those that activate serine phosphatases required for the reversible removal of the phosphoryl group, such as serine phosphatase activators, or addition of exogenous serine phosphatase enzyme (naturally isolated or produced in a recombinant host), or agents that directly inhibit hetero-octamer assembly, including peptides, proteins, or hydrophobic agents favoring the dissociation of the octamer (See Yu et al, 2006 Biochem 45)

In another embodiment, the agent that reduces a cresol metabolite in an individual is one that promotes the consumption or metabolism of cresol (eg para-cresol) The agent may be an enzyme that modulates the cresol (eg para-cresol) metabolic pathway, or the agent may target an enzyme that modulates the cresol (eg para-cresol) metabolic pathway, thereby effectively removing cresol by conversion to degradation products or downstream metabolites. In a preferred embodiment, the agent is para-cresol methyl hydroxylase (PCMH, EC 1.17.99.1), which converts para-cresol to p-hydroxybenzyl alcohol. PCMH may be naturally isolated or produced in a recombinant host such as a bacterial cell or a mammalian cell using methods known in the art. The agent may also be a gene encoding PCMH, which is delivered to a host using methods known in the art, such as cloning into a suitable vector for delivery to a subject, such as a virus or a bacteriophage. In another embodiment, the agent is a small molecule or peptide that allosterically increases the activity of PCMH, or a cofactor of PCMH that is necessary for optimal enzyme activity. In another embodiment, an agent comprising PCMH, or a gene encoding PCMH, or an allosteric modulator of PCMH, or a cofactor of PCMH, is preferentially administered to a population of Clostridia organisms of a subject. The agent may be formulated into any of a number of delivery systems known in the art to enable delivery to the distal gut of a subject, as discussed in more detail below. In one embodiment, the cresol metabolite may be toxic. Thus, the agent that reduces a cresol metabolite in an individual may be one that reduces the conversion of cresol (eg para-cresol) into a toxic downstream metabolite. Conversion of cresol into toxic downstream metabolites may occur preferentially when a subject's ability to detoxify cresol (eg para-cresol) through sulfation reactions has been impaired (for example, because of a lack of sulfate substrate) and may involve conjugation of para-cresol to other moieties, such as glucuronide or glutathione. In one embodiment, the toxic downstream metabolite is any of quinone methide, hydroxyl-benzaldehyde, 4-methyl- ortho-benzoquinone, and 4-hydroxyhippuric acid. In a preferred embodiment, the agent that reduces a cresol metabolite in an individual is one that inhibits an enzyme that converts cresol (eg para-cresol) into such a toxic metabolite. Examples of the enzymes that transform cresol (eg para-cresol) into said toxic metabolites comprise human variants of Cytochrome P450 (CYP450), such as enzymes which efficiently produce quinone methide (CYP2D6, CYP2C19, CYP1A2), hydroxyl-benzaldehyde (CYP1A2, CYP2D6), 4-methyl-orthe-benzoquinone (CYP2E1 , CYP1A1), and 4-hydroxyhippuric acid (CYP2E1 , CYP1A1) (Yan et al, 2005 Drug Metab. Disp. 33). In a preferred embodiment, the agent that reduces the conversion of cresol into a toxic metabolite is a small molecule inhibitor of the specific CYP isoforms mentioned above. Examples of inhibitors of CYP2D6 include quinidine (Branch et al, 2000 CHn. Pharmacol. Then 68), fluoxetine (Brynne et al, 1999 Br. J. CHn. Pharmacol. 48), and bupropion (Kotylar et al, 2005 J. CHn. Psychopharmacol. 25). Examples of CYP2C19 inhibitors include ticlopidine (leiri et al, 2005 Pharmacogenet. Genomics 47), and proton pump inhibitors such as omeprazole (Ko et al, 1985 Er. J. CHn. Pharmacol. 28). Examples of inhibitors of CYP1A2 include ciprofloxacin (Granfors et al, 2004 Clin. Pharmacol. Ther. 76), fluvoxamine (Brosen et al, 1993 Biochem. Pharmacol. 45), and furafylline (Sesardic et al, 1990 Br. J. CHn. Pharmacol. 29). Examples of inhibitors of CYP2E1 include diethyl- dithiocarbamate (Guengerich et al, 1991 Chem Res Toxicol. 4) and disulfiram (Kharasch et al, 1993 Clin. Pharmacol. Ther. 53). Additional non-limiting examples of inhibitors with activity for one or more of the CYPs listed above comprise terbinafine, selective serotonin reuptake inhibitors (SSRIs), moclobemide, ketoconazole, verapamil, cimetidine, and fluoroquinones. The inhibitors are used in this invention to treat a pathological condition associated with a sulfation deficiency, where the indication is not depression, hypertension, heart arrhythmia, a fungal infection, or a bacterial infection. In another preferred embodiment, agents are used that target microbial enzymes that convert cresol (eg para-cresol) into the aforementioned potentially toxic metabolites. Preferably, the agents are inhibitors of hydroxybenzyl alcohol dehydrogenase or bacterial decarboxylase enzymes. In a preferred embodiment, p-chloromercuribenzoate is used to inhibit hydroxyl-benzaldehyde production from hydroxybenzyl alcohol dehydrogenase.

In another embodiment, the agent that reduces a cresol metabolite in an individual, is a modulator of a cresol (eg para-cresol) producing microbe. For example, the agent that reduces a cresol metabolite may be one that modulates a microbe, including but not limited, to any of Clostridia sp., Bacteroides sp., Fusobacteria sp., Bifidobacteria sp., and Staphylococcus albus, which produce para-cresol directly from tyrosine or from products of tyrosine fermentation (Bone et al, 1976 Am. J. CHn. Nutr.29). Additionally, certain species of these microbes may produce para-cresol more readily or more robustly than others, such as C. difficile and C. butyricum (Bone et al). In this sense, the invention also provides for the specific modulation of individual species that are recognized as producing high levels of cresol (eg para-cresol). Modulators of said microbial species may comprise probiotics, prebiotics, synbiotics (combinations or prebiotics and probiotics), and antibiotics. In one embodiment, the cresol producing microbe belongs to the Clostridia class. In this embodiment, the agent that reduces a cresol metabolite in an individual is one that modulates Clostridia in the individual. Preferably, the modulators of p-cresol-producing Clostridia comprise prebiotics that have a detrimental effect on Clostridia, such as a galactooligosaccharide (GOS), lactulose, inulin, or a fructooligosaccharide (FOS). In another embodiment, the modulators of p-cresol-producing Clostridia comprise probiotics that specifically decrease Clostridial colonization in the individual, such as L. plantarum, L acidophilus, or Sacchromyces boulardii. In a further embodiment, the agent is a synbiotic (or combinations of pre- and pro-biotics discussed above) administered to decrease Clostridia growth and treat a condition. In another embodiment, the agent is an antibiotic administered to decrease the growth of Clostridia, including but not limited to metronidazole, vancomycin, linezolid, ramoplanin, and cholestyramine. Alternative classes of small molecule compounds suitable for inhibition of Clostridia are discussed in Reddy et al, 1982 App. Environ. Microb. 43, incorporated herein by reference.

In another embodiment, the agent that reduces a cresol metabolite in an individual, is a sorbent that adsorbs bacterial co-metabolites, thereby sequestering said metabolites and reducing the formation of cresol metabolites. An example of such a sorbent is AST-120, a spherical carbon preparation that adsorbs uremic toxins in the gut, including indoxyl sulfate and p-cresol (Niwa et al, 1993 Nephron 65). In one embodiment, the sorbent is formulated into a formulation suitable for oral administration and targeted delivery to the gut, and preferentially adsorbs toxins in the gut. In a preferred embodiment, the sorbent is modified with immobilized enzymes that possess catalytic activity to degrade p-cresol or a p-cresol metabolite. The enzymes immobilized comprise those mentioned above that sufficiently metabolize p-cresol or its metabolites.

In another embodiment, the agent that reduces a cresol metabolite in an individual, is an agent that inhibits, blocks, or otherwise disrupts a microbial transporter of tyrosine or tyrosine metabolites. Said transporters have been presumed or characterized in various microbial species (Wolken et al, 2006 J. Bacteriol. 188; Allende et al, 1992 Arch.

Biochem. Biophys. 292; Prieto et al, 1997 FEBS Lett. 414). Specific examples of transporters targeted in the invention are the tyrosine transporter, as well as transporters of metabolites in the cresol biosynthesis pathway, downstream of tyrosine and upstream of para-cresol. Preferably, the transporter targeted is a 4-hydroxyphenylacetate (4-HPA) transporter from a microbe.

In one embodiment, the agent that reduces a cresol metabolite in an individual is an inhibitor of the microbial 4-HPA transporter, which may cause a decrease of para-cresol production due to a lack of sufficient 4-HPA substrate. Suitable inhibitors comprise phenol derivatives of the natural substrate, 4-HPA, including modifications of the hydroxyl position and dihydroxy derivatives (See Prieto et al, 1997 FEBS Lett. 414). In a preferred embodiment, the inhibitors are specific for microbial transporters (and thus do not act on human transporters). In a particularly preferred embodiment, said transporter targeted is the 4-HPA transporter from Clostridium difficile. In addition to reducing the levels of a bacterial metabolite conjugated to a sulfur- containing moiety and a cresol metabolite in an individual, the invention also provides for combating a pathological condition in an individual by the use of agents to promote the availability of sulfur-containing moieties in the individual, i.e. increase the availability of sulfur resources or increase the sulfur pool in an individual. By sulfur resources or sulfur pool, we include the meaning of sulfur-containing compounds, such as sulfate and glutathione, that the body naturally conjugates to compounds such as xenobiotics or drugs in order to facilitate their excretion and elimination through the urine, and which may also participate in processes of drug activation, cellular signalling, or hormone activation and inactivation, as well as the precursors of said conjugates, such as taurine, cysteine, or methionine.

Agents that promote the availability of sulfur-containing moieties in the individual include small molecules, proteins, peptides, amino acids, carbohydrates, nucleotides, and lipids, and may increase the capacity of an individual to sulfate metabolites, proteins, peptides, amino acids, carbohydrates, nucleotides, and lipids, such as hormones, neurotransmitters, glycosaminoglycans, drugs, or toxins. It is appreciated that such agents can be used alone, or in combination with an agent to reduce a bacterial metabolite conjugated to a sulfur-containing moiety in an individual and/or an agent to reduce a cresol metabolite in an individual.

The agent that promotes availability of sulfur-containing moieties in the individual may be any donor of a sulfur-containing moiety such as 3'-Phosphoadenosine-5'-phosphosulfate (PAPS), N-acetyl cysteine, methionine, glutathione or taurine.

In one embodiment, it may be necessary to assess any of PCS levels, PCS/PCG ratio, bacterial faecal composition, or combinations of the above, of an individual to determine whether said individual should be treated with agents to modulate said individual's sulfur pool.

In one embodiment, the agent that promotes availability of sulfur-containing moieties in the individual is a modulator of a host sulfate transporter. Sulfate transporters, widely expressed in the kidney and intestines, control the influx and efflux of free sulfate, and thus their modulation is a key step in controlling sulfate availability for the variety of sulfur-conjugating reactions necessary for detoxification and general cellular maintenance. There are various sulfate transporters characterized in mammals, including the sodium/sulfate cotransporter NaS1 , sulfate/bicarbonate antiporter SAT1, sulfate/chloride antiporter Diastrophic Dysplasia Sulfate Transporter (DTDST), sulfate hydroxide antiporter Down Regulated in Adenoma transporter (DRA), and the sulfate/chloride-oxalate antiporter CFEX. Relevant characterization of normal function, distribution, and expression of these transporters is made clear in various studies (Markovich and Aronson, 2007 Annu. Rev. Physiol. 69 ; Morris and Murer, 2001 J. Membrane Biol. 181).

In a preferred embodiment, the agent that promotes availability of sulfur-containing moieties in the individual is an activator of a sulfate "importer" (by sulfate "importer" it is meant a transporter that makes sulfate available for sulfur-conjugating reactions). Examples of importers include NaS1 and SAT1. Agents that are suitable as activators of sulfate importers comprise compounds that are known to induce protein or genetic upregulation of said transporters. Examples of activators of NaS1 , include Vitamin D, thyroid hormone, and growth hormone (Sagawa et al, 1999 J. Pharmacol. Exp. Ther. 290; Fernandes et al, 1997 J. Clin. Invest. 100; Sagawa et al, 1999 An. J. Physiol. 276). Additionally, agents that induce or upregulate said natural transporter activators are suitable for use, as are agents that upregulate cofactors or protein adaptors necessary for transporter activity. In one embodiment, when the individual has, or has an increased risk of developing a condition associated with a sulfation deficiency (eg the aetiology of the condition involves conjugation of a molecule to a sulfur-containing moiety), the agent is a Vitamin D supplement to treat said sulfation deficiency by activating NaS1 import of free sulfate. In a preferred embodiment, said individual has an autism spectrum disorder and further has a mutation of a NaS1 transporter. Additionally, the agent that increases availability of sulfur-containing moieties in an individual may be an activator of renal and intestinal sulfate transporters also include the natural ions involved in co-transport or antiport. For example, sodium is the first binder and activator of the NaS1 transporter, and therefore sodium ion-containing compositions may be administered alone or in combination with sulfate ion-containing compositions to activate sulfate import in individuals with a condition associated with a sulfation deficit. As an alternative example, the agent that increases availability of sulfur-containing moieties may be a scavengers of bicarbonate or foods that promote reduced levels of systemic bicarbonate to activate (or more appropriately, reverse the inhibition of) the sat-1 sulfate/bicarbonate antiporter.

In another preferred embodiment, the agent that promotes availability of sulfur-containing moieties is an inhibitor of a sulfate "exporter" (by sulfate "exporter" it is meant a transporter that excretes sulfate and thus reduces sulfate availability for sulfur- conjugating reactions). An example of this type of transporter is the renal cell-to-lumen sulfate transporter CFEX. Molecules which inhibit the transporter, such as selenate or stilbene derivatives, as well as the use of high levels of ions against the gradient formed by CFEX, are suitable as antagonists. Additionally, CFEX is known to be activated by binding to proteins at its PDZ-domain binding motif (Markovich and Aronson, 2007 Annu, Rev. Physiol. 69), and thus inhibitors of this PDZ-mediated binding, for instance the binding of PDZK1 , are also preferable as therapeutic antagonists of sulfate transport. In one embodiment, said agents to modulate an individual's sulfur pool are supplements that contain either inorganic sulfur or sulfate, including sulfur-containing amino acids and their derivatives, such as cysteine, N-acetylcysteine, methionine, and taurine; sulfate salts, such as sodium sulfate, magnesium sulfate, and calcium sulfate; glutathione supplements and glutathione-rich diets such as diets comprising fresh fruits, vegetables, cooked fish and meat; sulfate-rich foods such as bread, dried fruits, vegetables, nuts, and fermented beverages; and formulations of sulfate donors, such as 3'- phosphoadenosine 5'phosphosulfate (PAPS), or a combination of an aryl sulfate and aryl sulfotransferase. The invention also provides for delivery systems that increase the oral bioavailability of said modulators (as will become clearer below).

In another embodiment, the agent that promotes availability of sulfur-containing moieties is a cofactor or otherwise factor used by enzymes conducting Phase Il detoxification reactions. In a preferred embodiment, the cofactors are used by enzymes conducting Phase Il sulfation reactions, and may comprise molybdenum, cysteine, methionine, Vitamin B-12, folic acid, methyl donors, magnesium, vitamin B-6/ pyridoxil-5-phosphate (P-5-P), and glucosamine chondroitin (MSM). In another preferred embodiment, the cofactors support glutathione synthesis and conjugation, and may comprise glutathione precursors such as cysteine, glycine, and glutamic acid; essential fatty acids such as black currant seed oil, flax seed oil, eicosapentaenoic acid (EPA); glutathione synthesis stimulators comprising Vitamin C, Mega H, N-acetyl cysteine, and botanical medicines such as silymarin; and parathyroid tissue. In yet another preferred embodiment, the cofactors support methylation, and comprise methionine, choline, magnesium, folic acid, Vitamin B-12, S-adenosylmethionine (SAM) and other methyl donors.

In another embodiment, the agent that promotes availability of sulfur-containing moieties is a direct inducer of phase Il detoxification enzymes, including inducers of sulfation enzymes such as cysteine, methionine, and taurine; inducers of glutathione-conjugating enzymes such as brassica vegetables and limonene-containing foods, such as citrus peel, dill weed oil, and caraway oil; and inducers of methylation enzymes, such as lipotropic nutrients such as choline, methionine, betaine, folic acid, and Vitamin B-12. In another embodiment, the agent that promotes availability of sulfur-containing moieties specifically targets sulfite oxidase (SOX; SOX is a crucial detoxification enzyme which metabolizes sulfites to sulfates, which can then be safely excreted in the urine; mutations of SOX are known to increase an individual's sensitivity to sulfur-containing drugs). In a preferred embodiment, the agent is Molybdenum.

In another embodiment, the agent that promotes availability of sulfur-containing moieties is a sulfotransferase enzyme. Said enzyme may have been naturally isolated or produced in a recombinant host. In a preferred embodiment, the agent is a phenol sulfotransferase (PST) selected from SULT1A1 and SULT1A2, for example when the individual has, or has an increased risk of developing ASD.

It is appreciated that when the agent is a sulfotransferase enzyme, it can be used preventative^, for example when the individual has been assessed has having an increased risk of developing a condition whose aetiology involves conjugation to a sulfur- containing moiety. In a preferred embodiment, one or more markers selected from PCS levels, PCS/PCG ratio, bacterial faecal composition, bacterial faecal Clostridia levels, an NaS1 transporter gene polymorphism, or combinations of the above, of an individual (eg infant), are measured, and used to determine whether the individual is at increased risk of developing an ASD and to determine whether the individual should be directed to a preventative treatment course of PST. Additionally, the treatment course may also include any of the agents discussed above to reduce a bacterial metabolite conjugated to a sulfur-containing moiety in an individual, to reduce a cresol metabolite in an individual or to increase the availability of sulfur-containing moieties in an individual.

Alternatively, where the agent is a sulfotransferase enzyme it may be used therapeutically. In this case, the individual is one who has, or has an increased risk of developing a pathological condition whose aetiology involves conjugation to a sulfur- containing moiety.

In another embodiment, the agent that promotes the availability of sulfur-containing moieties in an individual is a compound that activates a sulfotransferase enzyme. In a preferred embodiment, said compound activates phenol sulfotransferases (PSTs). Said compound may be a PPARα agonist (PPARα agonists upregulate the sulfotransferase enzymes at the genetic level, leading to increased expression and protein synthesis, and therefore increased systemic sulfotransferase activity). In an alternative preferred embodiment, the agent that promotes the availability of sulfur-containing moieties in an individual is an allosteric activator that binds directly to the sulfotransferase enzyme and enhances its catalytic activity. In another embodiment, the agent that promotes the availability of sulfur-containing moieties in an individual is an activator of a PAPS synthase enzyme (PAPS synthase generates PAPS, which is the main substrate for enzyme-driven sulfation reactions) selected from a PPARα agonist and a direct allosteric modulator. In yet another embodiment, the agent that promotes the availability of sulfur-containing moieties in an individual is an activator of a sulfatase (sulfatase enzymes remove sulfate groups from sulfated metabolites, and thus perform the opposite reaction of a sulfotransferase enzyme). As there are various metabolites and proteins in the body competing for sulfur, the activation of a sulfatase in a pathway not related to the specific sulfation pathway relevant to the condition of interest would increase the sulfate available to said pathway relevant to the condition of interest). In a preferred embodiment, the agent is an activator of estrogen sulfatases, for example when the individual has, or has an increased risk of developing an autistic spectrum disorder. In another embodiment, the agent that promotes the availability of sulfur-containing moieties in an individual is an activator of enzymes that make sulfur available for Phase Il detoxification reactions in the form of cysteine, glutathione, taurine, and sulfate. Said enzymes comprise enzymes responsible for steps along the conversion of homocysteine to said sulfur carriers. The activation of any of said enzymes, either directly or indirectly, accelerates or increases the transformation of homocysteine to readily usable sulfur resources for Phase Il detoxification reactions. In a preferred embodiment, the enzymes targeted for activation are those responsible for converting cysteine to glutathione, specifically glutathione synthase (EC 6.3.2.2) and glutamylcysteine ligase (EC 6.3.2.2). Examples of activators of glutathione synthase activity comprise butylhydroquinone derivatives, thioacetamide, diethyl maleate, and agents leading to the depletion of folic acid and Vitamin E (Huang et al, 2000 Biochim. Biophys. Acta 1493; Tchantchou et al, 2004 J. Neurosci. Res. 15). Examples of activators of glutamylcysteine ligase comprise a wide variety of agents that upregulate expression of enzyme subunits, such as erythropoietin, adriamycin, apocynin, quercetin, and 6-hydroxydopamine (Wild and Milcahy, 2000 Free Radic. Res. 32). In a preferred embodiment, the agent is adriamcyin, for example when the individual has, or has increased risk of developing a condition whose aetiology involves conjugation to a sulfur-containing moiety (eg ASD). In yet another embodiment, the agent that promotes the availability of sulfur-containing moieties in an individual may be cystathione β-synthase (CSB) or an activator of cystathione β-synthase (an enzyme which catalyzes the committing transformation of homocysteine to cysteine, beginning the generation of taurine, glutathione, and sulfate). In a preferred embodiment, the agent that promotes the availability of sulfur-containing moieties in an individual is a testosterone antagonist, (testosterone is known to downregulate CSB; the testosterone antagonist prevents said downregulation thus activating the synthesis of taurine, glutathione, and sulfate).

In another embodiment, the agent that promotes the availability of sulfur-containing moieties in an individual is an activator of PAPS translocase, allowing greater import of the PAPS substrate. The PAPS translocase enzyme is an exchanger of the universal sulfate donator PAPS with the used product of the reaction, PAP (Ozeran et al, 1996, Biochemistry 35). In one embodiment, small molecule inducers of PAPS, either direct activators of PAPS transport activity or inducers of gene expression, are administered to an individual with a condition with a sulfation-deficiency component.

In another embodiment, the agent that promotes the availability of sulfur-containing moieties in an individual is an agent that reduces populations of sulfate-reducing bacteria

(SRB). As these bacteria consume free sulfate necessary for sulfation reactions, inhibition of said bacteria increases availability of free sulfate to the host. Prebiotics or probiotics may be administered to reduce populations of SRB. In a preferred embodiment, the agent is methanogenic bacteria, for example when the individual has or has an increased risk of developing a pathological condition whose aetiology involves conjugation of a sulfur-containing moiety. Said methanogenic bacteria may out-compete the SRB and thus increase free sulfate. In another embodiment, the agent is molybdenum, preferably in the form of sodium molybdate, which decreases the burden of

SRB, specifically SRB of the genus Desulfovibrio. In another embodiment, the levels of SRB (which can be measured by methods known in the art such as genome sequencing or FISH), are used as a diagnostic for a sulfation deficiency condition.

In one embodiment, a specific dietary and/or pharmacological regime is applied to the individual, for example when the individual is diagnosed with a condition whose aetiology involves conjugation of a molecule to a sulfur-containing moiety (eg ASD). By applying a specific dietary and/or pharmacological regime, we include the meaning that the individual is directed to avoid specific dietary or pharmacological compounds. Said direction or recommendation is based on the fact that the sulfate transporters discussed above, are also downregulated or inhibited by certain foods and agents, and thus the avoidance of said foods and agents can restore the individual's sulfation capacity. Examples of inhibitors of NaS1 include activators of protein kinase C, activators of protein kinase A, glucocorticoids, and non-steroidal anti-inflammatory drugs (NSAIDS) (Lee et at, 2000 Proc. Soc. Exp. Biol. Med. 225; Sagawa et al, 1999 J. Pharmacol. Exp. Ther. 287; Markovich, 2001 Physiol. Rev. 81). Examples of inhibitors of the transporter SAT1 include molybdate, selenate, and thiosulfate (Markovich and Aronson, 2007 Annu. Rev. Physiol. 69). In a preferred embodiment, a diagnosis of autism to a patient by (a) a bacterial metabolite conjugated to a sulfur-containing moiety, (b) a cresol metabolite and (c) overall sulfur concentration and/or distribution of sulfur atoms in molecules is used to direct said patient to avoid high doses of NSAIDS in the management of inflammation and pain. In a further embodiment, one or more markers selected from PCS levels, PCS/PCG ratio, bacterial faecal composition, or combinations of the above, of an individual, are measured, and used to determine whether said individual should be directed to avoid specific foods or agents that are inhibitors of phase Il conjugating enzymes.

In a preferred embodiment, an individual that shows levels of said markers which indicate a sulfation deficiency is directed to avoid non-steroidal anti-inflammatory drugs (NSAIDS) such as aspirin. In another preferred embodiment, said individual is directed towards avoiding tartrazine.

As mentioned above and in Examples 1 and 2, the inventors have associated PCS to the capacity of an individual to conjugate a molecule to a sulfur-containing moiety and to disease. Thus, the inventors believe that pathological conditions whose aetiology involves conjugation of a molecule to a sulfur-containing moiety may be amenable to treatment aimed at modifying the capacity to conjugate a molecule to a sulfur-containing moiety by administering an agent that reduces the amount of a bacterial metabolite conjugated to a sulfur-containing moiety in the individual.

Accordingly, a tenth aspect of the invention provides a method of combating in an individual a pathological condition whose aetiology involves conjugation of a molecule to a sulfur-containing moiety, the method comprising administering an agent to reduce the amount of a bacterial metabolite conjugated to a sulfur-containing moiety in the individual or an agent that reduces the amount of a cresol metabolite in the individual or an agent that increases the availability of sulfur-containing moieties in the individual. The invention provides the use of an agent to reduce the amount of a bacterial metabolite conjugated to a sulfur-containing moiety in an individual or an agent that reduces the amount of a cresol metabolite in the individual or an agent that increases the availability of sulfur-containing moieties in the individual in the manufacture of a medicament for combating in an individual a pathological condition whose aetiology involves conjugation of a molecule to a sulfur-containing moiety

The invention provides an agent to reduce the amount of a bacterial metabolite conjugated to a sulfur-containing moiety in an individual or an agent that reduces the amount of a cresol metabolite in the individual or an agent that increases the availability of sulfur-containing moieties in the individual for use in combating in an individual a pathological condition whose aetiology involves conjugation of a molecule to a sulfur- containmg moiety

Preferences for the pathological condition, the individual and the agent that reduces the amount of a bacterial metabolite conjugated to a sulfur-containing moiety or the agent that reduces the amount of a cresol metabolite or an agent that increases the availability of sulfur-containing moieties in the individual are defined above with respect to the ninth aspect of the invention

In a preferred embodiment, the pathological condition is an autism spectrum disorder

The invention provides a method of combating an autism spectrum disorder in an individual, the method comprising administering an agent to reduce the amount of a bacterial metabolite conjugated to a sulfur-containing moiety in the individual or an agent to reduce the amount of a cresol metabolite in the individual or an agent that increases the availability of sulfur-containing moieties in the individual Preferably, the agent in respect of the tenth aspect of the invention is one that reduces the amount of PCS in the individual

Similarly, the invention provides the use of an agent to reduce the amount of a bacterial metabolite conjugated to a sulfur-containing moiety in an individual or an agent that reduces the amount of a cresol metabolite or an agent that increases the availability of sulfur-containing moieties in the individual in the manufacture of a medicament for combating an autism spectrum disorder in an individual. Preferably, the agent is one that reduces the amount of PCS in the individual.

Likewise, the invention provides an agent to reduce the amount of a bacterial metabolite conjugated to a sulfur-containing moiety in an individual or an agent that reduces the amount of a cresol metabolite for use in combating an autism spectrum disorder in an individual or an agent that increases the availability of sulfur-containing moieties in the individual. Preferably, the agent is one that reduces the amount of PCS in the individual. As discussed above, the capacity of an individual to conjugate a molecule to a sulfur- containing moiety is central to various biological processes including the body's defence mechanisms against free radicals and oxidative damage. Thus, without wishing to be bound by any theory, the inventors believe that the capacity to conjugate a molecule to a sulfur-containing moiety as evidenced by assessing any one or more of (a) a bacterial metabolite conjugated to a sulfur-containing moiety, (b) a cresol metabolite and (c) overall sulfur concentration and/or distribution of sulfur atoms in molecules, may be used as a marker for general well-being. Consequently, assessing any one or more of (a) a bacterial metabolite conjugated to a sulfur-containing moiety, (b) a cresol metabolite and (c) overall sulfur concentration and/or distribution of sulfur atoms in molecules will be of value in assessing the efficacy of a treatment regime for general well-being.

Accordingly, an eleventh aspect of the invention provides a method for assessing the efficacy of a treatment regime for general well-being, the method comprising:

a) assessing any one or more of (a) a bacterial metabolite conjugated to a sulfur-containing moiety, (b) a cresol metabolite and (c) overall sulfur concentration and/or distribution of sulfur atoms in molecules in a sample taken from an individual;

b) administering a treatment regime for general well-being;

c) assessing the any one or more of (a) a bacterial metabolite conjugated to a sulfur-containing moiety, (b) a cresol metabolite and (c) overall sulfur concentration and/or distribution of sulfur atoms in molecules in a sample taken from the individual subsequent to administering the treatment regime; and d) comparing the assessment made in step (a) with that made in step (c) to assess the efficacy of the treatment regime. In a particularly preferred embodiment, the method comprises assessing PCS in a sample taken from an individual. Thus, the invention includes a method for assessing the efficacy of a treatment regime for general well-being, the method comprising:

a) assessing PCS in a sample taken from an individual ;

b) administering a treatment regime for general well-being;

c) assessing PCS in a sample taken from the individual subsequent to administering the treatment regime; and

d) comparing the assessment made in step (a) with that made in step (c) to assess the efficacy of the treatment regime.

It is appreciated that the method may be employed in the context of a clinical trial of a candidate treatment regime, eg a drug, for general well-being. In this case the method is typically performed on a population of individuals. Thus, the method may be carried out on at least 10, 50, 100, 200, 300, 400, 500 individuals, or at least 1000 individuals, or at least 5000 individuals or more.

As is well known in the art, to control for the 'placebo effect', it may be desirable to substitute the treatment regime for a placebo in a proportion of the individuals undergoing the clinical trial.

The treatment regime may be one based upon diet, exercise or other lifestyle change, or pharmaceutical intervention, or a combination thereof.

A pharmaceutical intervention may comprise an agent that reduces the amount of a bacterial metabolite conjugated to a sulfur-containing moiety in the individual or an agent that reduces the amount of cresol metabolite in the individual or an agent that increases the availability of sulfur-containing moieties in the individual, including those described above. Thus, the treatment regime may be an agent that reduces the amount of a bacterial metabolite conjugated to a sulfur-containing moiety in the individual or an agent that reduces the amount of a cresol metabolite, such as any of an antibiotic, a probiotic or a tyrosine oxidase inhibitor, or an agent that increases the availability of sulfur- containing moieties in the individual such as 3'-phosphoadenosine-5'-phosphosulfate (PAPS), N-acetyl cysteine, methionine, glutathione and taurine. When the treatment regime is pharmaceutical agent, the treatment regime may be administered as an individual dose or in several doses over a period of 1 , 2, 3 or 4 weeks or months, or more. Typically, assessing any one or more of (a) a bacterial metabolite conjugated to a sulfur- containing moiety, (b) a cresol metabolite and (c) overall sulfur concentration and/or distribution of sulfur atoms in molecules in a sample taken from an individual is done immediately prior to the commencement of administering the treatment regime.

The assessment of the any one or more of (a) a bacterial metabolite conjugated to a sulfur-containing moiety, (b) a cresol metabolite and (c) overall sulfur concentration and/or distribution of sulfur atoms in molecules in a sample taken from the individual subsequent to the administration step may be determined multiple times, for example at regular intervals such as weekly or monthly in order to monitor efficacy of the treatment regime over time.

It is appreciated that it may be desirable to compare the effects of a treatment regime for general well-being with an alternative treatment for general well-being.

A twelfth aspect of the invention provides a use of a means for assessing any one or more of (a) a bacterial metabolite conjugated to a sulfur-containing moiety, (b) a cresol metabolite and (c) overall sulfur concentration and/or distribution of sulfur atoms in molecules in a sample taken from an individual in assessing the efficacy of a treatment regime for general well-being.

In a particularly preferred embodiment, the means for assessing any one or more of (a) a bacterial metabolite conjugated to a sulfur-containing moiety, (b) a cresol metabolite and (c) overall sulfur concentration and/or distribution of sulfur atoms in molecules is a means for assessing PCS in a sample taken from an individual and so the invention includes a use of a means for assessing PCS in a sample taken from an individual in assessing the efficacy of a treatment regime for general well-being. The inventors have associated the capacity of an individual to conjugate a molecule to a sulfur-containing moiety, with PCS present in a sample taken from the individual. Thus, without wishing to be bound by any theory the inventors have realised that an agent that modulates the amount of a bacterial metabolite conjugated to a sulfur-containing moiety or the amount of a cresol metabolite may be used to modulate the ability of an individual to conjugate a molecule with a sulfur-containing moiety. Accordingly, a thirteenth aspect of the invention provides a method for modulating the ability of an individual to conjugate a molecule with a sulfur-containing moiety, comprising administering an agent that modulates the amount of a bacterial metabolite conjugated to a sulfur-containing moiety in the individual or an agent that modulates the amount of a cresol metabolite.

In a particularly preferred embodiment, the method comprises administering an agent that modulates the amount of PCS in the individual. Thus, the invention includes a method for modulating the ability of an individual to conjugate a molecule with a sulfur- containing moiety, comprising administering an agent that modulates the amount of PCS in the individual.

It is appreciated that the agent that modulates the amount of a bacterial metabolite conjugated to a sulfur-containing moiety or that modulates the amount of a cresol metabolite may be one that does so directly or indirectly. Thus the agent may be one that modulates the amount of the bacterial metabolite itself and therefore also the bacterial metabolite conjugated to a sulfur-containing moiety.

Agents that reduce the amount of a bacterial metabolite conjugated to a sulfur-containing moiety and agents that reduce the amount of a cresol metabolite include those defined above with respect to the ninth aspect of the invention. Thus, the agent may be any of an antibiotic, a probiotic that changes the gut microflora to reduce the proportion of bacteria producing the bacterial metabolite that is subsequently conjugated to a sulfur containing moiety, or a tyrosine oxidase inhibitor.

Agents that increase the amount of a bacterial metabolite conjugated to a sulfur- containing moiety include any donor of a sulfur-containing moiety such as 3'- Phosphoadenosine-5'-phosphosulfate (PAPS), N-acetyl cysteine, methionine glutathione or taurine, or a probiotic that changes the gut microflora to increase the proportion of bacteria producing the bacterial metabolite that is subsequently conjugated to a sulfur- containing moiety.

Agents that increase cresol metabolites include those that upregulate the cresol biosynthetic pathway and cresol metabolism pathways described above, including agents that target molecules of the cresol biosynthesis and metabolic pathway to increase a cresol metabolite, or agents that activate an enzyme in the cresol biosynthetic pathway responsible for production of cresol. Alternatively, the agent may be one that reduces the consumption or metabolism of cresol (eg para-cresol), or one that is a modulator of a cresol (eg para-cresol) producing microbe (eg an agent that promotes the growth of cresol producing bacteria such as Clostridia or an agent that reduces the growth of bacteria that do not produce cresol).

Whilst it is possible for the agent that reduces a bacterial metabolite conjugated to a sulfur-containing moiety in an individual or the agent that reduces a cresol metabolite in an individual or an agent that increases the availability of sulfur-containing moieties in an individual, to be administered alone, it is preferable to present it as a pharmaceutical formulation, together with one or more acceptable carriers. The carrier(s) must be "acceptable" in the sense of being compatible with the therapeutic agent and not deleterious to the recipients thereof. Typically, the carriers will be water or saline which will be sterile and pyrogen free. Where appropriate, the formulations may conveniently be presented in unit dosage form and may be prepared by any of the methods well known in the art of pharmacy. Such methods include the step of bringing into association the active ingredient (agent that reduces a bacterial metabolite conjugated to a sulfur-containing moiety in an individual or the agent that reduces a cresol metabolite in an individual or an agent that increases the availability of sulfur-containing moieties in an individual) with the carrier which constitutes one or more accessory ingredients. In general the formulations are prepared by uniformly and intimately bringing into association the active ingredient with liquid carriers or finely divided solid carriers or both, and then, if necessary, shaping the product. Formulations in accordance with the present invention suitable for oral administration may be presented as discrete units such as capsules, cachets or tablets, each containing a predetermined amount of the active ingredient; as a powder or granules; as a solution or a suspension in an aqueous liquid or a non-aqueous liquid; or as an oil-in-water liquid emulsion or a water-in-oil liquid emulsion. The active ingredient may also be presented as a bolus, electuary or paste.

A tablet may be made by compression or moulding, optionally with one or more accessory ingredients. Compressed tablets may be prepared by compressing in a suitable machine the active ingredient in a free-flowing form such as a powder or granules, optionally mixed with a binder (e.g. povidone, gelatin, hydroxypropylmethyl cellulose), lubricant, inert diluent, preservative, disintegrant (e.g. sodium starch glycolate, cross-linked povidone, cross-linked sodium carboxymethyl cellulose), surface-active or dispersing agent. Moulded tablets may be made by moulding in a suitable machine a mixture of the powdered compound moistened with an inert liquid diluent. The tablets may optionally be coated or scored and may be formulated so as to provide slow or controlled release of the active ingredient therein using, for example, hydroxypropylmethylcellulose in varying proportions to provide desired release profile.

Formulations suitable for topical administration in the mouth include lozenges comprising the active ingredient in a flavoured basis, usually sucrose and acacia or tragacanth; pastilles comprising the active ingredient in an inert basis such as gelatin and glycerin, or sucrose and acacia; and mouth-washes comprising the active ingredient in a suitable liquid carrier.

Preferred unit dosage formulations are those containing a daily dose or unit, daily sub-dose or an appropriate fraction thereof, of an active ingredient.

It should be understood that in addition to the ingredients particularly mentioned above (agent that reduces a bacterial metabolite conjugated to a sulfur-containing moiety in an individual or the agent that reduces a cresol metabolite in an individual or an agent that increases the availability of sulfur-containing moieties in an individual) the formulations of this invention may include other agents conventional in the art having regard to the type of formulation in question, for example those suitable for oral administration may include flavouring agents.

The amount of the agent which is administered to the individual is an amount effective to combat the particular individual's condition. The amount may be determined by the physician.

It is appreciated that agents herein described which act on the cresol biosyntheis and metabolic pathway may be formulated for enteral, parenteral, and topical delivery. The invention provides for formulations of said agents that enable targeted delivery to organs, particularly the gut, the liver, the kidneys, and the central nervous system. A particularly preferred embodiment comprises oral formulations of cresol biosynthesis inhibitors for targeted delivery to the colon, where said agents may be most effective in preventing cresol biosynthesis given that the colon is densely populated with Clostridia, which produce large amounts of cresol. Another particularly preferred embodiment comprises oral and intravenous formulations of cresol metabolism modulators for targeted delivery to the colon and the liver, where said agents may be most effective in stimulating cresol degradation or preventing formation of toxic cresol metabolites, given that the colon and liver are the major sites where said metabolism reactions occur.

Parenteral formulations of any of the cresol biosynthesis and metabolic pathway modulators described above may involve a number of dosage formulations known in the art for intravenous, intra-arterial, intramuscular, or subcutaneous injection, comprising solutions, suspensions, and lyophilized powders to be reconstituted in solution.

Parenteral formulations for protein-based agents, such as PCMH or a serine phosphatase enzyme may further comprise a number of excipients known in the that serve the purpose of enhancing protein stability in solution and lyophilized formulations, such as Triton X-100, Fibronectin, Pluronic F0127, Heparin, Cellobiose, trehalose, mannitol, or sucrose.

Enteral formulations of any of the cresol biosynthesis and metabolic pathway modulators described above may involve a number of dosage formulations known in the art for oral, gastric feeding tube, and rectal administration. In one embodiment, a suppository or enema form is used for rectal delivery of the agents of the invention. In a preferred embodiment, enteral formulations consisting of tablets, capsules, or drops are used for oral delivery targeted to the colon of a subject. Numerous methods have been described in the art to enable general delivery of drugs to the colon, and any of them can be combined with the agents of this invention to enable delivery to the colon. Such methods include pH-sensitive formulations (e.g. formulations coated with enteric polymers that release drug when the pH move towards a more alkaline range, after passage through the stomach), formulations that delay the release of the drug for a lag time of 3-5 hours, roughly equivalent to small intestinal transit time, thereby securing delivery to the colon, drugs coated with bioadhesive polymers that selectively provide adhesion to the colonic mucosa (e.g. see US Patent 6,368,586), and delivery systems that incorporate protease inhibitors to prevent proteolytic activity in the gastrointestinal tract from degrading biologic drug agents.

Formulations for oral delivery that enable general delivery of drugs to the colon may be prepared using a pharmaceutically acceptable "carrier" composed of materials that are considered safe and effective and may be administered to an individual without causing undesirable biological side effects or unwanted interactions. By "carrier", it is meant all components present in the pharmaceutical formulation other than the active agent or agents, such as diluents, binders, lubricants, disintegrators, fillers, and coating compositions, as well as components of the coating composition which may include plasticizers, pigments, colorants, stabilizing agents, and glidants.

The formulations may be delayed-release. Delayed release dosage formulations for oral delivery targeted to the colon may be prepared as described in references such as "Pharmaceutical dosage form tablets", eds. Liberman et. al. (New York, Marcel Dekker, Inc., 1989), "Remington - The science and practice of pharmacy", 20th ed., Lippincott Williams & Wilkins, Baltimore, MD, 2000, and "Pharmaceutical dosage forms and drug delivery systems", 6th Edition, Ansel et.al., (Media, PA: Williams and Wilkins, 1995) which provides information on carriers, materials, equipment and process for preparing tablets and capsules and delayed release dosage forms of tablets, capsules, and granules. Optional pharmaceutically acceptable excipients present in the drug- containing tablets, beads, granules or particles include, but are not limited to, diluents, binders, lubricants, disintegrants, colorants, stabilizers, and surfactants. As will be appreciated by those skilled in the art and as described in the pertinent texts and literature, a number of methods are available for preparing drug-containing tablets, beads, granules or particles that provide a variety of drug release profiles. Such methods include, but are not limited to, the following: coating a drug or drug-containing composition with an appropriate coating material, typically although not necessarily incorporating a polymeric material, increasing drug particle size, placing the drug within a matrix, and forming complexes of the drug with a suitable complexing agent. The delayed release dosage units may be coated with the delayed release polymer coating using conventional techniques, e.g., using a conventional coating pan, an airless spray technique, fluidized bed coating equipment (with or without a Wurster insert), or the like. For detailed information concerning materials, equipment and processes for preparing tablets and delayed release dosage forms, see Pharmaceutical Dosage Forms: Tablets, eds. Lieberman et al. (New York: Marcel Dekker, Inc., 1989), and Ansel et al., Pharmaceutical Dosage Forms and Drug Delivery Systems, θ.sup.th Ed. (Media, PA: Williams & Wilkins, 1995). Delayed release formulations are created by coating a solid dosage form with a film of a polymer which is insoluble in the acid environment of the stomach, and soluble in the neutral environment of small intestines. The delayed release dosage units can be prepared, for example, by coating a drug or a drug-containing composition with a selected coating material. The drug-containing composition may be, e.g., a tablet for incorporation into a capsule, a tablet for use as an inner core in a "coated core" dosage form, or a plurality of drug-containing beads, particles or granules, for incorporation into either a tablet or capsule. Formulations for oral delivery that enable general delivery of drugs to the colon may also comprise enteric coated capsules consisting of natural polymers or mixtures of natural polymers which are insoluble in the acidic pH of the stomach, polymers useful for surface coatings that are applied by spraying, brushing, or various industrial processes, and gelling agents that undergo a high degree of cross-linking or association when hydrated and dispersed in the dispersing medium, or when dissolved in the dispersing medium. Exemplary of natural polymers resistant to the acidic pH of the stomach include, but are not limited to, pectin and pectin-like polymers which typically consist mainly of galacturonic acid and galacturonic acid methyl ester units forming linear polysaccharide chains.

In a particularly preferred embodiment, the agents of the invention are directed to specific microbes, for example a Clostridia sp., by formulation with delivery systems targeting the microbe of interest. An example includes conjugation of the competitive pHPA-decarboxylase inhibitor with peptides targeting cell surface molecules selectively expressed in Clostridia species.

Liver-targeted formulations for systemic delivery of agents may comprise chemical conjugates of said agents to steroids and lipids (see Soutschek et al, Nature, 432, 173- 178, 2004, or Lorenz et al, Bioorg Med Chem Lett., 14, 4975-4977, 2004), for example chemical conjugates of said agents with cholesterol. Polymeric systems can also be used for gene delivery to the liver, for examples systems based on poly(L-lysine) conjugated to asialoorosomucoid for targeted delivery to liver hepatocytes (See Wu et al, J Biol Chem, 262, 4429-4432, 1987). Liver-targeted formulations may further comprise targeting ligands that enable tissue-specific gene expression, e.g. galactose for hepatocyte targeting (See Han et al, lnt J Pharm, 202, 151-160, 2000, or Perales et al, Proc Natl Acad Sci USA, 91 , 4086-4090, 1994).

Kidney-targeted formulations for systemic delivery of agents may comprise use of prodrugs of any of the sulfur pool modulators described above that exploit metabolic processes that are selectively localized to the kidney. Glutathione, γ-glutamyl derivatives, and β-lyase substrates have been investigated as renal-specific prodrugs based on the presence of certain transporters such as plasma membrane transport systems for glutathione and cysteine conjugates (se for example Lash et al, MoI Pharmacol, 28, 278- 282, 1985, or Scaheffer et al, MoI Pharmacol, 33, 293-298, 1987) and enzymes such renal specific transporter or enzymes such as γ-glutamyltransferase (EC 2.3.2.2), dipeptidase, and β-lyase (EC4.4.1.13). In a preferred embodiment, a sulfur pool modulator is conjugated to a moiety selected from a glutathione group, a γ-glutamyl derivative, and β-lyase substrate.

Central Nervous System (CNS)-targeted formulations for delivery of any of the sulfur pool modulator agents and cresol metabolism modulator agents described above may comprise formulations that circumvent poor solubility of active agents at the blood-brain barrier. Bolus cerebral injections, implanted biodegradable drug crystals (see

Routtenberg, Behav Biolo, 7:601-641 , 1972), implantable cannulas for chronic drug injections (see Ott et al, Pharmacol Biochem Behav, 2:715-718, 1974 or Crane et al, Pharmacol Biochem Behav, 10: 799-800, 1979), or osmotic pumps for local delivery to

CNS (Theeuwes et al, Ann Biomed Eng, 4:343:353, 1976) have been described in the art and may be used in conjunction with the agents of the invention.

Topical formulations for transdermal delivery of any of the sulfur pool modulator agents and cresol biosynthesis and metabolism modulators described above are also provided.

Said topical formulations may comprise chemical modulators of transdermal permeation described in the art, such as sulfoxides, azone, pyrrolidones, fatty acids, alcohols, fatty alcohols, glycols, surfactants, urea, terpenes, and phospholipids (See Williams, A.C.

"Transdermal and Topical Drug Delivery; from theory to clinical practice". Pharmaceutical Press, London. 2003"). Said topical formulations may also comprise physical or technical modulators of transdermal penetration which have been described in the art, such as liposomal drug carriers, non-ionic surfactant vesicles, or other strategies to physically or mechanically perturb the skin barrier, such as microneedles, laser ablation, electroporation, iontophoresis, ultrasound, magnetophoresis, and needleless injections with high-velocity particle jets.

Formulations for enteral or parenteral delivery of proteins, such as the enzymes listed above, may comprise features that protect said protein from proteolysis reactions in the gastrointestinal tract or in the blood, by a number of techniques known in the art, including PEGylation, conjugation to other macromolecules, formation of covalent bonds that stabilize the protein structure, such as disulfide bonds, or encapsulation within polymeric carrier systems, such as methacrylic acid, polyvinylalcohol, polyvinylpirrolidone, gelling polysaccharides, polyethylene oxide, or polyethylene glycol. For example, the cresol-degrading enzyme p-cresol methyl hydroxylase, may be administered in a PEGylated version, allowing for oral passage and proper bioavailability to the gut In one embodiment, the agent delivered is a small molecule delivered that has low oral bioavailability and acts on a microbial population of the host's gut. Low oral bioavailability is generally undesirable in drugs, since absorption through the intestine is an objective of most oral therapies. However, it is appreciated that in the context of modulating the intestinal flora, it may be a desirable attribute since the intended site of action is the contents of the gut. The permeability of the small molecule may be decreased by tethering the small molecule to a high molecular weight compound such as a polymer, thereby reducing the molecule's bioavailability. Highly hydrophilic or hydrophobic small molecules may be used, since both extremes are detrimental to oral bioavailability (highly hydrophilic molecules do not cross the epithelium, while highly hydrophobic molecules are not solubilised in aqueous media). In one embodiment, a prodrug is formed by chemically modifying a small molecule drug with highly charged groups that increase the hydrophilicity of said drug, thereby decreasing its oral bioavailability. As discussed above, some of the agents that promote the availability of sulfur-containing moieties in an individual, are themselves sulfur-containing compounds. However, the clinical efficacy of such sulfur-containing compounds may be limited by their low oral bioavailability (for example, glutathione and sulfate have particularly low oral bioavailability). Accordingly, the invention provides various methods to increase the oral bioavailability of sulfur-containing compounds, which methods comprise formulating the sulfur-containing compounds. The invention includes such formulations. It is appreciated that these methods may aid in the management of conditions associated with a sulfation deficiency, By 'oral bioavailability' we include the meaning of the degree to which or rate at which a substance is absorbed or becomes available at a site of physiological activity in an individual after oral administration. Oral bioavailability can be assessed using any suitable technique known in the art. If bioavailability of a sulfur-containing compound administered intravenously is 100%, preferably, the method of the invention increases oral bioavailability to at least 20%, 30%, 40%, or 50% and more preferably to at least 60%, 70%, 80%, 90% or 100%.

Sulfur-containing compounds may comprise sulfate salts (for example magnesium sulfate or its heptahydrate, MgSO ₄ -7H ₂ O, commonly called Epsom salt, iron sulfate, sodium sulfate, or calcium sulfate), 3'-Phosphoadenosine-5'-phosphosulfate (PAPS), N- acetyl cysteine, glutathione, taurine or methionine. Low bioavailability of sulfur- containing compounds may be a consequence of their low solubility or low permeability. Cosolvents known in the art may be used to increase solubility, such as dimethylsulfoxide (DMSO), propylene glycol, ethanol, polyethoxylated castor, or mixtures of the above. Low bioavailability of sulfur-containing compounds may also be a consequence of their low permeability, in which case mechanisms to increase flux of said compounds through the epithelial gut membrane may be exploited, including altering the membrane microenvironment or increasing the concentration gradient of the compound at the membrane interface. The membrane microenvironment may be altered through the use of absorption enhancers known in the art, such as Tween 80 or Pluronic. Increasing the concentration gradient of the compound at the membrane interface may be accomplished by formulating the drug in particles of mucoadhesive polymers known in the art, such as polyanhydrides, which attach to the gut epithelium, thereby increasing the local concentration gradient of said compound at the membrane interface. Said mucoadhesive polymers also extend the retention time of said compound in the intestine by adhering to the gut wall, further enhancing the bioavailability of said compound. Increasing the concentration gradient of the compound at the membrane interface may also be accomplished by formulating the drug into particles containing lectin moieties at the particle surface, which enable lectin binding to carbohydrates on the gut membrane and thus generate higher localized drug concentrations next to the gut membrane, thereby increasing absorption (See "Van de Waterbeemd et al, "Drug Bioavailability", in Methods and Principles in Medicinal Chemistry, Wiley, vol 40, 2009 ).

Increased permeability of sulfur-containing compounds may also be accomplished by formulation into lipid-based hydrophobic carriers, such as liposomes or chylomicrons. By chylomicrons, it is meant large lipoprotein particles comprised of bile acids that transport dietary lipids from the intestines to other locations in the body. Associations of lipophilic compounds with emulsions of chylomicrons have been described in the art (For example, see Gershkovich, Eur J Pharm Sci. 2005 Dec;26(5):394-404)

The bioavailability of the sulfur-containing compounds of the invention may also be enhanced by synthesis of prodrugs of said compounds. Prodrugs may be synthesized by conjugating chemical groups in reactive moieties of said compounds, such as carboxylic acid, alcohol, phenol, amine, or amide moieties. Said prodrugs may be cleaved by hydrolysis reactions after the prodrug has crossed the intestinal membrane, thereby enhancing said compound's bioavailability. In a preferred embodiment, said sulfur-containing compound is allithiamin (a natural compound that occurs for example in garlic), and said compound is modified into thiamine tetrahydrofurfuryl disulfide, a lipid- soluble prodrug form of allithiamin. The bioavailability of the sulfur-containing compounds of the invention may also be enhanced by conjugating said compounds to substrates of active transporters of the gastrointestinal tract, such as peptide transporters, nucleoside transporters, amino acid transporters, monosaccharide transporters, organic cation transporters, organic anion transporters, monocarboxylate transporters, ABC transporters, or bile acid transporters, thus facilitating the transport of said compounds through an active transport mechanism. In a preferred embodiment, the sulfur-containing compound may be conjugated to a Vitamin B-12 substrate.

A further application of the inventors' observations relating to the effect of PCS on the capacity of an individual to conjugate a molecule to a sulfur-containing moiety, is the prediction of the degree to which an exogenous molecule will be conjugated to a sulfur- containing moiety.

Accordingly, a fourteenth aspect of the invention provides a method of predicting the degree to which an exogenous molecule will be conjugated to a sulfur-containing moiety either directly or indirectly in an individual, the method comprising assessing any one or more of (a) a bacterial metabolite conjugated to a sulfur-containing moiety, (b) a cresol metabolite and (c) overall sulfur concentration and/or distribution of sulfur atoms in molecules in a sample taken from the individual. Thus the invention provides a method of predicting the degree to which an exogenous molecule will be sulfated or sulfonated, or the degree to which an exogenous molecule will be conjugated to glutathione or N- acetyl cysteine.

In a preferred embodiment, the method comprises assessing PCS in a sample taken from the individual. Thus, the invention includes a method of predicting the degree to which an exogenous molecule will be conjugated to a sulfur-containing moiety either directly or indirectly in an individual, the method comprising assessing PCS in a sample taken from the individual.

Preferences for the exogenous molecule are defined above with respect to the third aspect of the invention. A fifteenth aspect of the invention provides a use of a means for assessing any one or more of (a) a bacterial metabolite conjugated to a sulfur-containing moiety, (b) a cresol metabolite and (c) overall sulfur concentration and/or distribution of sulfur atoms in molecules in a sample taken from an individual in predicting the degree to which an exogenous molecule will be conjugated to a sulfur-containing moiety.

In a preferred embodiment, the means is a means for assessing PCS in a sample taken from an individual and so the invention includes a use of a means for assessing PCS in a sample taken from an individual in predicting the degree to which an exogenous molecule will be conjugated to a sulfur-containing moiety.

Preferences for the exogenous molecule are defined above with respect to the third aspect of the invention.

In a further embodiment of the fourteenth and fifteenth aspects of the invention, the assessing comprises comparing the measure of any one or more of (a) a bacterial metabolite conjugated to a sulfur-containing moiety, (b) a cresol metabolite and (c) overall sulfur concentration and/or distribution of sulfur atoms in molecules to the measure of a respective any one or more of (a) a bacterial metabolite conjugated to a sulfur-containing moiety, (b) a cresol metabolite and (c) overall sulfur concentration and/or distribution of sulfur atoms in molecules in a standard sample obtained from one or more individuals in whom the degree to which an exogenous molecule will be conjugated to a sulfur-containing moiety either directly or indirectly is known. Generally, the higher the amount of a bacterial metabolite conjugated to a sulfur-containing moiety in the sample or the higher the amount of a cresol metabolite in the sample or the lower the overall sulfur concentration and/or a changed distribution of sulfur atoms in molecules, the lower the degree to which the exogenous molecule will be conjugated to a sulfur-containing moiety.

It will be appreciated that having determined an individual's ability to conjugate an exogenous molecule with a sulfur-containing moiety, the method may be used to design an appropriate nutritional therapy or intervention for that individual. Thus, the invention provides a method for designing a nutritional therapy or intervention personalized to an individual's ability to conjugate a molecule with a sulfur-containing moiety, wherein said ability to conjugate a molecule with a sulfur-containing moiety is determined according to the fourteenth aspect of the invention, for example by assessing any one or more of (a) a bacterial metabolite conjugated to a sulfur-containing moiety, (b) a cresol metabolite and (c) overall sulfur concentration and/or distribution of sulfur atoms in molecules in a sample taken from the individual. Preferably, the degree to which an individual can conjugate an exogenous molecule to a sulfur-containing moiety further comprises genotyping any of the 'biological parameters' mentioned above and/or phenotyping any the metabolic parameters mentioned above (eg measures of cresol biosynthesis and metabolic pathway). Further, the method may further comprise analyzing associations between genetic, metabolic, and nutritional status of a subject using a computer supported algorithm, and using said associations to design a personalized nutritional therapy for a subject afflicted with a sulfation deficiency. The nutritional therapy may include combinations of elements selected from drinks rich in mineral sulfate salts, cysteine or methionine rich diets, and foods rich in sulfate such as bread, dried fruit, vegetables, nuts, and fermented beverages. Additionally, the nutritional therapy may provide for a recommendation to avoid a food, including avoidance of monoamine containing foods such as cheese, chocolate, or bananas, avoidance of flavonoids, and avoidance of phenolic compounds and salycilates.

A further aspect of the invention provides a method for identifying novel therapeutics for a condition whose aetiology involves conjugation to a sulfur-containing moiety, the method comprising the screening and identification of inhibitors of bacterial phenolic transporters. In one embodiment, the method entails screening of compounds against both bacterial transporters and equivalent human transporters, followed by the identification of inhibitors with greater than 2-fold, 5-fold, 10-fold, or 100-fold selectivity for the microbial transporters versus the human transporters. The screening may be conducted in an in vitro protein transport assay, a cellular uptake assay, or a computational virtual screen. In a preferred embodiment, a transporter from a specific subset of bacteria, for example the 4-HPA transporter from Clostridium difficile, is used in a screen to identify inhibitors with selectivity over both human and other homologous bacterial transporters.

A further aspect of the invention provides a method of identifying therapies for a condition whose aetiology involves conjugation to a sulfur-containing moiety (eg a sulfation deficiency), the method comprising screening for activators, allosteric or otherwise, of a protein involved in sulfation or glutathione detoxification. The key proteins relevant to the sulfation and glutathione pathways comprise, but are not limited to, sulfotransferases, PAPS synthases, CBS, cysteine oxidase, SOX ₁ and sulfate transporters. In one embodiment, the method entails screening of compounds against at least one protein selected from a sulfotransferase, a PAPS synthase, CBS, cysteine oxidase, SOX, and a sulfate transporter, and selecting one or more compounds that enhance the conversion of a substrate of said protein into a product of said protein or enhance the flux of a sulfur-containing substrate across a sulfate transporter. The screening may be conducted in an in vitro protein transport assay, a cellular enzymatic assay, or a computational virtual screen. In one embodiment, a library of small molecules is screened for activation of a phenol sulfotransferase enzyme in an in vitro high-throughput assay, and a subset of molecules are identified which enhance the catalytic activity of said enzyme, and said molecules are confirmed in vivo to have a beneficial effect on the sulfation capacity of an organism. In another embodiment, the three-dimensional structures of proteins selected from a sulfotransferases, a PAPS synthases, CBS ₁ cysteine oxidase, SOX, and a sulfate transporters, which are known in the art, are used to guide structure-based drug design on allosteric regions of the enzyme, lead compounds are synthesized, and then tested to confirm in vitro catalytic enhancement and in vivo efficacy.

A further aspect of the invention provides a method of predicting the degree to which an exogenous molecule will be conjugated to a sulfur-containing moiety either directly or indirectly in an individual, the method comprising administering paracetamol to the individual and determining a ratio of two or more paracetamol conjugates produced by Phase Il detoxification.

Preferably, the individual is administered a safe dose of paracetamol, such as 200 mg or less, 150 mg or less, or 100 mg or less. It is particularly preferred if the dose is 81 mg or less.

The paracetamol conjugates may be produced via any Phase Il detoxification route as is well known in the art. Preferably, the ratio of acetaminophen-sulfate to acetaminophen- glucuronide is determined (see Example 1). Typically, the method comprises comparing the ratio with a corresponding ratio obtained from a population of subjects with known degree to which an exogenous molecule will be conjugated to a sulfur-containing moiety either directly or indirectly. Optionally, the method may further comprise assessing the individual's excretion of sulfur-containing products in urine, such as sulfate, sulfite, thiosulfate, and thiocyanate, and together with the determined ratio of paracetamol conjugations, used to predict the degree to which an exogenous molecule will be conjugated to a sulfur-containing moiety either directly or indirectly in an individual The invention also provides safer versions of drugs that are conjugated to a sulfur moiety during Phase Il detoxification reactions. The present invention suggests that saturation of an individual's ability to conjugate molecules with a sulfur-containing moiety, caused by microbial co-metabolites, xenobiotics, a genetic predisposition, or combinations of the above, may be a causation factor in diseases such as autism. Therefore, it is another object of the invention to propose modified forms of existing drugs known to undergo sulfur conjugation reactions and drug regimens that do not require sulfur-conjugation reactions by the host, thus alleviating the host's needs for sulfur resources.

In one embodiment, a known drug is modified to avoid sulfur-conjugation while preserving the drug's efficacy in its new form. In a preferred embodiment, a compound that would normally be sulfated by Phase Il conjugation reactions formulated in a sulfated form, or in a form containing a sulfur masked by a capping groups such as an ester, which may be cleaved by cellular esterases. In one embodiment, said compound is acetaminophen, and the modified forms are acetaminophen sulfate and an acetaminophen-thioester.

The data produced from carrying out the methods of the invention may conveniently be recorded on a data carrier. Thus, the invention includes a method of recording data on the classification of an individual using any of the methods of the invention and recording the results on a data carrier. Typically, the data are recorded in an electronic form and the data carrier may be a computer, a disk drive, a memory stick, a CD or DVD or floppy disk or the like.

Information recorded on the data carrier may include the name, date of birth, age, sex, height and body mass index of the individual or individuals.

The invention will now be described in more details with the aid of the following Figures and Examples.

Figure 1. Representative ¹ H NMR spectra of urine samples provided before and after taking 1 g of acetaminophen, a, the δ 8.0 - 0.5 region of the pre-dose urine spectrum for a subject whose urine contained a relatively high level of p-cresol sulfate. The peak at ca. δ 2.35, which arises from the methyl group of p-cresol sulfate, is ringed in red on the relevant expansion, b, the corresponding 0-3 h post-dose urine spectrum, which shows a relatively low ratio of acetaminophen sulfate to acetaminophen glucuronide. c, the δ 8.0 - 0.5 region of the pre-dose urine spectrum for a subject whose urine did not contain a high level of p-cresol sulfate, d, the corresponding 0-3 h post-dose spectrum, which shows a relatively high ratio of acetaminophen sulfate to acetaminophen glucuronide. To facilitate their comparison, all these spectra were processed in the same way, without resolution enhancement and with a digital filter used to minimise the residual water features, which would otherwise be observed at ca. δ 4.7. Furthermore, each spectrum has been scaled so that the creatinine methylene peak at ca. δ 4.06 is just on scale (with the result that the corresponding creatinine methyl peak at ca. δ 3.03 is off scale in each case). The insets, which are expansions of selected spectral regions, are scaled to fill the available space. Key to numbered compounds: 1 , creatinine; 2, hippurate; 3, phenylacetylglutamine; 4, p-cresol sulfate; 5, citrate; 6, cluster of N-acetyl groups from acetaminophen-related compounds; 7, acetaminophen sulfate; 8, acetaminophen glucuronide; 9, other acetaminophen-related compounds. Figure 2. ¹ H NMR spectra of pre-dose urine samples with colour-coding according to post-dose behaviour. All the plots were produced in MATLAB with each individual spectrum being normalised to constant creatinine, a, an expansion of the δ 2.335 - 2.360 spectral region, which contains the methyl signal from p-cresol sulfate (PCS), with the individual spectra for the 25 subjects giving the highest post-dose 0-3h S/G ratios shown in blue and superimposed on the individual spectra for the 25 subjects giving the lowest post-dose 0-3h S/G ratios shown in red. b, is the same plot as above but with the further addition of the corresponding data for the other 49 subjects (shown in green), c, shows the same spectral region and the average spectra for the three different groups, with the same colour-coding, d, shows the same average pre-dose spectra, with the same colour-coding, over the region of δ 7.18 - 7.32, which contains the PCS aromatic signals (the pair of 'doublets' centred at ca. δ 7.21 and at ca. δ 7.29). In all plots 'a.u.' designates arbitrary units. In plots a and b, some spectra are obscured by the subsequently superimposed spectra. Figure 3. The observed relationship between the pre-dose urinary ratio of para-cresol sulfate (PCS) to creatinine and the post-dose urinary ratio of the major acetaminophen metabolites, acetaminophen sulfate (S) and acetaminophen glucuronide (G). a, shows the pre-dose urinary PCS/creatinine integral ratio for each subject plotted against the corresponding urinary S/G ratio obtained in the 0-3 h post-dose collection, b, the corresponding plot for the 3-6 h post-dose collection. I. R. designates the integral ratio of the peaks at ca. δ 2.35 and at ca. δ 4.06 in the ¹ H NMR spectrum recorded from the pre- dose urine sample. For equimolar PCS and creatinine, I. R. would be expected to approximate to 1.5 because of the number of protons contributing to each signal. No subjects were excluded from either plot, c, a table summarising our interpretation of the findings. Figure 4. Relevant metabolic pathways, a, the hydroxyl group of acetaminophen (1) may either be sulfonated to produce acetaminophen sulfate (2) or glucuronidated to produce acetaminophen glucuronide (3). b, step-wise production of para-cresol sulfate (8) from tyrosine (4) ²² . The green box highlights the highly analogous and potentially competitive sulfonation of acetaminophen and para-cresol; both reactions are known to proceed under the influence of the human cytosolic sulfotransferase SULT1 A1 , for which the two substrates may compete, and both also depend on the potentially limited availability of the sulfonate donor S'-phosphoadenosine-δ'-phosphosulfate, which is consumed as sulfonation proceeds ²⁴ ' ²⁵ . c, step-wise production of phenylacetylglutamine (12) from phenylalanine (9) ²² with the yellow box highlighting similarities with the metabolism of tyrosine. Key to compounds: 1, acetaminophen; 2, acetaminophen sulfate; 3, acetaminophen glucuronide; 4, tyrosine; 5, 4- hydroxyphenylpyruvic acid; 6, 4-hydroxyphenylacetic acid; 7, para-cresol; 8, para-cresol sulfate; 9, phenylalanine; 10, phenylpyruvic acid; 11, phenylacetic acid; 12, phenylacetylglutamine. In the body, compounds 2 and 8 would normally be expected to exist as ROSO ₃ ^" rather t han as ROSO ₃ H, where R designates the remainder of each molecule. Indicated references are those listed for Example 1.

Figure 5. Diagram illustrating 4-cresol metabolism Figure 6. Comparison of structures of acetaminophen sulfate and cresol sulfate showing that these are structural analogues.

Figure 7. Selected scores and loadings plots obtained from PCA data, (a) The scores plot for PC2 vs. PC5 where each point represents a different subject. The points for the 25 subjects showing the highest S/G 0-3 h ratios are coloured blue. The points for the 25 subjects showing the lowest S/G 0-3 h ratios are coloured red. The points for the 49 subjects showing intermediate S/G 0-3 h ratios are coloured green. The plot shows partial separation of the red and blue points on PC2; (b) The corresponding loadings plot, with each 0.04 ppm-wide segment identified by the chemical shift at its centre. The 7.36, 7.40 and 7.44 spectral segments relate mainly to phenylacetylglutamine (PAG). The 7.20, 7.24 and 7.28 spectral segments relate mainly to p-cresol sulfate (PCS). This plot indicates that the spectra producing high scores on PC2 have relatively high levels of PAG and/or PCS. Furthermore, it indicates that the blue coloured points in a, mainly have relatively low levels of PAG and PCS.

Example 1: Pharmacometabonomic identification of a novel host-microbiome interaction affecting human metabolism

Summary

We provide the first demonstration in man of the principle of pharmacometabonomics by showing a clear connection between an individual's metabolic phenotype, in the form of a pre-dose urinary metabolite profile, and the metabolic fate of a standard dose of the widely used analgesic acetaminophen. Pre- and post-dose urinary metabolite profiles were determined by ¹ H NMR spectroscopy. The pre-dose spectra were statistically analyzed in relation to drug metabolite excretion to detect pre-dose biomarkers of drug fate and a novel human-gut microbiome co-metabolite predictor was identified. Thus, we found that individuals having high pre-dose urinary levels of p-cresol sulfate had low post-dose urinary ratios of acetaminophen sulfate to acetaminophen glucuronide. We conclude that, in individuals with high bacterially-mediated p-cresol generation, competitive O-sulfonation of p-cresol reduces the effective systemic capacity to sulfonate acetaminophen. Given that acetaminophen is such a widely used and seemingly well understood drug, this novel finding provides a clear demonstration of the immense potential and power of the pharmacometabonomic approach. However, we expect many other sulfonation reactions to be similarly affected by competition with p-cresol and our finding also has important implications for certain diseases as well as for the variable responses induced by many different drugs and xenobiotics. We propose that assessing the effects of microbiome activity should be an integral part of pharmaceutical development and of personalized healthcare. Furthermore, we envisage that gut bacterial populations might be deliberately manipulated in order to improve drug efficacy and to reduce adverse drug reactions

Introduction

The effects of drug treatments can vary greatly between different individuals and pharmacogenomics has been widely advocated as a potential means of personalizing human drug treatments in order to increase drug efficacy and to decrease adverse reactions (1-6) However, environmental factors (such as nutritional status, gut bacterial activities, age, disease and other drug use) are also important determinants of individual metabolic phenotypes, which modulate drug metabolism, efficacy and toxicity, and such environmental complications, which may also alter gene expression, will tend to limit the usefulness of predictions of drug-induced responses that are based only on genomic differences (7, 8) For instance, for many classes of compound, enzyme induction state, which is environmentally determined, influences drug metabolism and toxicity and this is not captured in genomic data Recognizing this important limitation of pharmacogenomics, a different approach to personalized drug treatment has recently been proposed wherein pre-dose metabolite profiling would instead be used to predict a subject's responses to potential drug interventions (9) This new 'pharmacometabonomic' approach has a number of major advantages, which include the ready availability and relative ease of analysis of biofluids such as urine and blood plasma as well as the fact that the derived metabolite profiles are sensitive to both genomic and environmental influences affecting metabolism A further crucial advantage and feature of this metabonomic approach is its openness to finding unexpected biomarkers and biomarker combinations, as multiple analytes are quantified simultaneously without pre-specification of what those analytes should be (10) The only factors limiting which analytes are detected are the nature of the sample that is analyzed and the analytical platform employed Thus, pharmacometabonomic modelling need not be limited by prior understanding or hypothesis However, despite much support and enthusiasm for the concept (9, 11-13), there has, until now, to the best of our knowledge, been no convincing pharmacometabonomic demonstration in man

To test the feasibility of applying the pharmacometabonomic approach to man, we chose as our example the well known analgesic and antipyretic drug acetaminophen (N-acetyl- p-aminophenol, known as paracetamol in Europe) Acetaminophen is one of the most widely used non-prescription medicines in the world and its toxicology and metabolism have been extensively investigated over many years (14-20) However, we will show here that, even for this most familiar drug, pharmacometabonomic analysis will yield significantly increased understanding of its metabolic behaviour in man. These findings have considerable implications for personalized drug treatment in general and lead to new and testable hypotheses for a number of diseases.

Acetaminophen was chosen to exemplify the pharmacometabonomic principle for a variety of reasons, which included its common usage and its low toxicity at therapeutic doses, which was necessary to establish an ethically-approved clinical trial. It is also predominantly and relatively rapidly eliminated in the urine (15-19). Thus, by collecting post-dose urine samples we could study the manner of its excretion by each subject with such excretion being known to show considerable inter-subject variation (20). Precisely how a particular drug is metabolized and excreted by each individual can have a major influence on its safety and efficacy. Thus, for instance, greater or lesser production of a toxic metabolite might be expected to influence the extent of an adverse effect, whilst the rate of removal of the pharmacologically-active compound might be expected to influence the extent and duration of the desired pharmacological action. With this in mind, the aim of the present study was to determine if metabonomic analysis of pre-dose human urine would allow prediction of some aspect of acetaminophen metabolism or excretion at the individual subject level and, thereby, provide a proof-of-principle for the feasibility of pharmacometabonomics in man.

Our ethically-approved study (supplementary information, Sl) was based on 99 healthy male volunteers who were all non-smokers aged between 18 and 64 years old with a condition of their participation being that they had not taken any drugs in the preceding week. Each participant provided a pre-dose urine sample and then, after taking a standard therapeutic dose of acetaminophen (two 500 mg tablets), each was requested to collect all of his post-dose urine over two consecutive 3 h periods (0 - 3 h and 3 - 6 h after dosing). All the samples were analyzed by 600 MHz ¹ H NMR spectroscopy with the pre-dose spectra providing profiles of the detectable, naturally-occurring ('endogenous') metabolites and the post-dose spectra providing profiles of the acetaminophen-related compounds superimposed on an endogenous metabolite 'background'. Using these NMR spectra and the known urine volumes, the urinary acetaminophen-related compounds excreted by each subject over each post-dose collection period were quantified as acetaminophen sulfate (S), acetaminophen glucuronide (G) and 'other', with the 'other' components expected to be mainly the parent, cysteine conjugate and N- acetylcysteine conjugate (17-19). We then searched for components of inter-subject variation in the pre-dose spectra that would correlate with inter-subject variation in the post-dose data. However, from the outset, prediction of the S/G ratio was of particular interest as this ratio is known to show extensive inter-individual variation and is assumed to be indicative of the relative extent to which acetaminophen is metabolized via two major phase 2 conjugative processes (O-sulfonation and glucuronidation) that impact on the metabolism of many different drugs (20, 21). Additionally, we expected that the S/G ratio would be less susceptible to any sample collection errors than the absolute amounts of metabolites excreted.

Materials and Methods

Full details are provided in the supplementary information (Sl). The study volunteers were nominally healthy men aged between 18 and 64 years old. Urine samples were prepared for NMR analysis by mixing 440 μl of urine with 220 μl of phosphate buffer (pH ca. 7.4 to which sodium azide had been added as an antibacterial preservative) and the mixture centrifuged to remove suspended particulates. 550 μl of 'clear' buffered urine was transferred to a sample vial and 55 μl of a TSP/D ₂ O solution added to give a final TSP concentration of 1 mM. TSP (sodium 3-trimethylsilyl-[2,2,3,3- ² H ₄ ]-1 -propionate) is a chemical shift reference compound used in the NMR experiment and the D ₂ O provided a field/frequency lock for the NMR spectrometer. The ¹ H NMR spectra of the prepared urine samples were acquired at 303K on a Bruker Avance 600 NMR spectrometer equipped with a flow probe and operated by means of the 'Xwinnmr' software (all from Bruker Biospin, Rheinstetten, Germany). The 'noesypresat' pulse sequence was used to suppress the water signal and the spectral acquisitions automated using the 'lconnmr' software and a 'BEST' sample changer (both Bruker Biospin). Using Xwinnmr, 1 Hz line- broadening was applied to the pre-dose ¹ H NMR spectra by means of an exponential multiplication of the free induction decay signal and these time-domain data were Fourier-transformed to frequency-domain spectra with a single zero-filling and with a digital filter used to reduce the size of the residual water signal. The resulting spectra were manually phased to give an even baseline around the NMR signals and the baseline of each spectrum was manually adjusted to zero intensity using a straight-line baseline correction algorithm. The chemical shift scale was set by assigning the value of δ 0 to the signal from the added reference compound (TSP). The post-dose ¹ H NMR spectra were processed similarly, using 0.3 Hz line broadening, and subsequently with resolution enhancement, and a measure approximating to the mole ratio of acetaminophen sulfate (S) to acetaminophen glucuronide (G) determined by integration of the respective resolution-enhanced N-acetyl peaks. For each subject and collection period, relative measures of the amounts of S and G excreted were also determined, by reference to the added TSP and the mass of urine collected. Except where stated, these excretion values were not adjusted to excretion per unit body mass. Subsequently, the pre-dose spectra were loaded into Matlab using the 'MetaSpectra' routine provided by Dr O. Cloarec and normalised to a constant integral for the δ 4.07 to 4.05 region, which encompasses the creatinine methylene singlet. Average group spectra were then calculated in Matlab and various spectral plots were made and examined. After integrating each pre-dose spectrum over consecutive 0.04 ppm spectral segments and then normalising these integrals to a constant value for the integral for the δ 4.07 - 4.05 region, Principal Components Analysis of the integrals for the δ 9.1 - 6.9 region was performed in Pirouete 3.11 (from Infometrix, U.S.A.) using mean-centred variable scaling. The PCA was also repeated after normalisation to a constant integral for the δ 3.07 - 3.03 region, which encompasses the creatinine methyl singlet. Local baseline correction and integration of selected peaks in the pre-dose spectra was performed in Xwinnmr. Statistical significance of abstracted data was assessed using the Mann- Whitney U test.

Results and Discussion The urinary excretion of acetaminophen and its metabolites.

We found that approximately 30% of the 1 g dose was recovered (as any acetaminophen-related compound) in each of the two post-dose urine collection periods (0-3 h and 3-6 h), this being consistent with the findings of an earlier report (15) where the urinary excretion of acetaminophen, S and G was studied at 1.5 h intervals after a 12 mg/kg dose. Furthermore, we found that, in terms of moles, the combined excretion of S and G typically accounted for ca. 85% of the total amount of acetaminophen-related compounds recovered in each collection period, this being consistent with the 24 h urinary recoveries from 111 Caucasians given a 1.5 g dose (20). Representative pre- and post-dose spectra are shown in Figure 1.

The average S/G ratios for the two post-dose collections were found to be 0.71 (0-3 h) and 0.53 (3-6 h) respectively with the S/G value for each individual subject always being lower in the 3-6 h collection than in the 0-3 h collection and this change in the S/G ratios being consistent with what has been reported for the early excretion of a 20 mg/kg dose (17). Furthermore, there was a clear correlation between the S/G 0-3 h and S/G 3-6 h data (r = 0.927). However, a plot of these data indicated an outlier and it was judged, from this and from an unusually low excretion of paracetamol-related metabolites, that this subject had not fully collected his 0-3 h sample. With this subject excluded, the correlation between S/G 0-3 h and S/G 3-6 h was slightly improved (r = 0.948). A moderate (17%) decrease in average S excretion was observed between the two collection periods along with a largely compensating increase in average G excretion. Having excluded the 0-3 h data for the one subject who appeared not to have collected his 0-3 h urine properly, the correlation coefficients between the observed S/G ratios and the amount of S excreted were found to be 0.510 (0-3 h) and 0.835 (3-6 h) respectively. Likewise, the correlation coefficients between S/G and the amount of G excreted were found to be -0.738 (0-3 h) and -0.803 (3-6 h) respectively. Precautionary checks showed that the amounts of S and G excreted, and the S/G ratio, were not related to the age of the subjects or to their body mass or to the order in which the samples were analyzed (Sl). Examination of the pre-dose spectral profiles.

Our initial way of searching for relationships between the pre- and post-dose data was by means of the PLS (Projection to Latent Structure)-based pattern recognition methods that are typically used in metabonomic studies (22) and which had proved highly effective in our earlier animal-based work (9). However, in the present case, the PLS-based approach was relatively unproductive (Sl) and, subsequently, we made a detailed visual comparison of creatinine-normalized pre-dose spectra for subjects at the two ends of the S/G ratio distribution (25 subjects at each end). With this revised and relatively simple approach, we found two potentially discriminatory pre-dose metabolites, later identified as the microbial co-metabolites p-cresol sulfate (PCS) and phenylacetylglutamine (PAG), with higher levels of these metabolites being visually associated with lower S/G values (Figures 1 and 2). These findings were supported by Principal Components Analyses (PCA) focussed on the aromatic region (δ 9.1 - 6.9) of the pre-dose NMR spectra (Sl) and no other components of the pre-dose spectra were found to have such clear discriminatory potential in regard to the S/G ratios observed. On closer inspection, the pre-dose urinary levels of PCS and PAG were found to be broadly correlated (r = 0.75), which, in retrospect, was unsurprising because there is a significant degree of commonality in their origins prior to the final sulfate and glutamine conjugations; thus, p- cresol and phenylacetic acid are known to be produced from tyrosine and phenylalanine, respectively, with these conversions being largely analogous and dependent on the action of colonic bacteria (23) (Fig. 3). However, our further analysis, using integrated pre-dose spectral band intensities and numerical single variable discovery procedures, showed that, of all the individual spectral components, only PCS was likely to provide statistically significant discrimination with respect to S/G (Sl).

Further analyses focussed on PCS.

In order to get the best possible measure of the level of PCS relative to creatinine, which, for this study, was a suitable internal reference compound for the urinary quantitation (Sl), selected peaks in the pre-dose NMR spectra were then integrated after local baseline correction. The peaks chosen for integration were the PCS methyl singlet at ca. δ 2.35 and the creatinine methylene singlet at ca. δ 4.06 and the relevant integral ratios [designated I. R., where I. R. = (integral of PCS methyl)/(integral of creatinine methylene)] obtained for each of the 99 subjects are presented in Fig 4, where these pre-dose data are plotted against the corresponding S/G ratios for the two post-dose collections. From inspection of Fig. 4, it is readily apparent that a high pre-dose level of PCS (I. R. > 0.06) is associated with a low S/G ratio post-dose and use of the Mann-Whitney U test in conjunction with an appropriate Bonferroni correction (100; Sl) confirmed the statistical significance of the distribution of the high PCS group (25 subjects) with respect to the S/G ratios obtained in each post-dose collection. The Bonferroni correction was applied to counter the multiple hypothesis testing that results from the multivariate nature of metabonomic data (24). With a Bonferroni correction of 100, the P value for 95% confidence becomes 0.05/100 = 5 x 10" and the P values obtained from the Mann Whitney tests were 1.O x IO- ⁴ (for S/G 0-3 h) and 1.2 x 10 ^"4 (for S/G 3-6 h).

In the preceding analysis of the full data set (all subjects included), we found a high pre- dose level of PCS to be associated with a low S/G ratio post-dose and, clearly, that post- dose ratio could be affected by variation in the amounts of both S and G excreted. However, since the conversion of p-cresol to PCS is analogous to the conversion of acetaminophen to S (Fig. 3), the observed connection to pre-dose PCS strongly suggests that the relevant controlling factor is the amount of S excreted. Furthermore, with lower S/G ratios being observed in the 3-6 h collection than in the 0-3 h collection, it appears that 1g of acetaminophen represents a substantial challenge to the sulfonation capacity of the subjects studied. The extent to which any compound undergoes sulfonation can potentially be limited both by the availability of the sulfonate donor, 3'- phosphoadenosine δ'-phosphosulfate (PAPS), and by the characteristics and availability of the relevant sulfotransferase enzyme (25). Thus, in the present context, it is particularly notable that p-cresol and acetaminophen are both substrates for the same human cytosolic sulfotransferase, SULT1A1 (26), and can, therefore, compete for enzyme binding sites as well as for PAPS. Furthermore, in contrast to what has been reported for rats (27), recent literature suggests that p-cresol is almost entirely converted to PCS in man (28-30). Thus, we envisaged that an individual's capacity to sulfonate acetaminophen will be reduced by ongoing presentation of endogenous p-cresol and the potential competitive significance of the p-cresol challenge was confirmed by calculation (Sl). On the basis of this hypothesis, we examined, with the full data set, the post-dose excretion of both S and G and found that, in the 3-6 h collection, the high pre-dose PCS subjects (LR. > 0.06) were clearly associated with lower S excretion (P = 2.3 x 10 ^'5 with 95% confidence 3I P = S x IO ^"4 after Bonferroni correction). Furthermore, a similar and statistically significant (P = 1.6 x 10 ^"4 ) relationship was found when the S excretion values for this collection period were first corrected to unit body mass. Thus, these data are fully consistent with the hypothesis that substantial pre-dose production of endogenous p-cresol can reduce an individual's ability to sulfonate acetaminophen by acting as a competitive substrate (Fig. 3). As regards the site of this competitive sulfonation, the colonic mucosa is known to have significant sulfonation capacity (25, 31) and could potentially convert colonically-produced p-cresol to PCS. However, with acetaminophen being rapidly absorbed from the small intestine and with the liver being regarded as the principal site of its metabolism (17), we envisage that some colonically- produced p-cresol may escape further colonic modification and be sulfonated in the liver rather than in the gastrointestinal tract.

Influence of experimental variables.

To check for experimental variables that might be influencing these data, the distribution of the 25 high pre-dose PCS (I.R.> 0.06) subjects was also investigated with respect to analytical run order and with respect to the data obtained for subject age, height, body mass and body mass index (BMI) but no association was found. However, it was noticeable that the 10 subjects showing the highest pre-dose PCS levels (LR. > 0.09) tended to be older (P = 0.01) and shorter (P = 0.02) individuals who would, in general, be expected to have less muscle mass and to excrete less creatinine. However, the first of these findings also suggests that ageing might lead to increased p-cresol production and this could potentially be caused by age-related changes in the nature of the gut bacteria

(32). Furthermore, whereas, in the present study, we found no clear evidence of a relationship between the S/G ratio and age, an age-related decrease in acetaminophen sulfonation has been observed in male rats (33). To further check the basis of our findings we also re-examined the data after first excluding all those subjects where there was any known or suspected non-compliance with the study protocol, e.g. where a sensitive analysis of a subject's pre-dose urine sample suggested some prior use of acetaminophen or where there was evidence of recent alcohol consumption (Sl). With the remaining 78 subjects, the graphical relationship between pre-dose PCS and the S/G ratio was maintained [the P-values for the distribution of the high PCS subjects (I. R. > 0.06) with respect to S/G 0-3 h and S/G 3-6 being 7.4 x 10 ^" " and 9.9 x 10 ^"4 respectively] and statistical significance was still achieved for the distribution of the high pre-dose PCS subjects with respect to the absolute amount of S excreted during the 3-6 h collection (P = 3.4 x 10 ^"4 ). We, therefore, conclude that the perceived pre-to-post-dose connection is real.

Potential biomedical significance.

Although the potential significance of the gut bacteria in relation to human metabolism, disease and drug-induced reactions is becoming increasingly well recognized (34 -41), the present finding is believed to be entirely novel and could be of considerable importance if it can be proven to hold for the wider human population or for specific subsets of that population.

We envisage that, by depleting hepatic sulfonation capacity, continual exposure to colonically-produced p-cresol would leave the liver more vulnerable to acetaminophen- induced damage and that markedly increased p-cresol production could potentially explain the reported association between fasting and an increased likelihood of hepatotoxicity from acetaminophen (42, 43). However, in principle, sustained prior exposure to colonically-produced p-cresol could also potentially increase acetaminophen hepatotoxicity by other means, such as by enzyme induction or glutathione depletion

(44-46), and preliminary data (Sl) suggests that high p-cresol exposure might lead to a more generalised impairment of sulfur-dependent reactive metabolite detoxification, with

PAPS depletion possibly leading to depletion of both taurine and glutathione. However, it remains to be investigated if our present finding has any significance for adverse reactions to acetaminophen. Instead, the wider and more obvious significance of our finding lies in its potential consequences for sulfonation reactions in general and in suggesting a potentially causal link between certain diseases and the gut bacteria.

Many different compounds are substrates for sulfotransferase-catalysed sulfonation, which, by making them more hydrophilic, has a major role in modifying the physical properties of both small and large molecules. Thus, sulfonation facilitates the excretion of many compounds and is crucial to the structure and properties of macromolecules such as chondroitin sulfate (a component of cartilage). Notably, many drugs and/or their hydroxylated metabolites are phase Il conjugated via sulfonation. Amongst several other important functions, sulfonation is also known to have a role in modulating the action of hormones and neurotransmitters and appears to be especially important during early human development (21 , 25, 26, 47-51). There are various human sulfotransferases (both cytosolic and membrane-bound) but human cytosolic sulfotransferase SULT1A1 , which acts on acetaminophen, has a broad substrate range and is one of the most important sulfotransferases for xenobiotic sulfonation as well as acting on several endogenous substrates (26, 51). Additionally, a key feature of sulfonation in higher organisms is that all such reactions utilise PAPS as the universal sulfonate donor. Thus, we might reasonably expect that, by competing for PAPS or for one or more sulfotransferases, the flux of p-cresol through the system will affect the sulfonation of a wide range of drugs and endogenous compounds, thereby influencing normal bodily processes as well as drug metabolic fate, efficacy and toxicity. However, given what is known about gut bacterial production of p-cresol from protein residues (23, 43, 52-54) (Fig. 3), with Clostridium difficile being one of a number of p-cresol producers (53, 54)), our present results show that environmental factors can exert a dominant influence on the extent to which a compound becomes sulfonated in the human body. Thus, we would expect that, by altering the amount of p-cresol produced, variation in either the diet or the gut bacteria could potentially exert a major influence on drug-induced responses or diseases where sulfonation has an important role.

In its role as a hypotensive agent, minoxidil provides one example of a drug where sulfonation is considered to be important in producing the desired pharmacological effect (55). However, sulfonation is not always beneficial and tamoxifen, which is used in treating breast cancer, provides an example of a drug where sulfonation has been suggested to be important for the development of an associated adverse reaction, namely an increased incidence of endometrial cancer. Thus, it has been suggested that tamoxifen-DNA adducts are formed via O-sulfonation (56). As a further example, sulfonation phenotype could potentially influence both the efficacy and side effects of apomorphine, for which sulfonation is the major metabolic pathway in man (57). As regards known associations with disease, hyperactivity in children provides one example of a condition that has been associated with increased p-cresol levels and where the involvement of dietary factors is also suspected (58). Furthermore, an increased urinary level of PCS has been associated with the progression of multiple sclerosis (59, 60). Additionally, various other diseases (Parkinson's disease, motor neurone disease, rheumatoid arthritis and childhood autism) have been associated with a reduction in the S/G ratios obtained following acetaminophen dosing (61-63), which leads us to tentatively suggest, on the basis of our current finding, that excessive gut bacterial p- cresol production might also have some relevance to their aetiology, with further circumstantial evidence coming from additional associations with gastrointestinal abnormalities (64-67). However, it should be clearly recognised that these various associations do not prove a causal role for p-cresol in respect of these diseases and also that p-cresol can exert a variety of effects (27-30, 43, 45, 46, 68-73) such as blocking the conversion of the neurotransmitter dopamine to noradrenaline (68). Therefore, as far as we are aware, any involvement of p-cresol in respect of these diseases remains to be proven as well as the exact nature of any such involvement. However, one general hypothesis would be that where the diet or the profile of the gut bacteria are altered in favour of p-cresol production, impaired sulfonation and other effects can result such that, depending on a subject's individual characteristics and state of development, a variety of consequences would be expected.

Conclusions and future prospects.

In the population studied, in this, our first pharmacometabonomic study in man, we have found a clear and novel association between an individual's pre-dose urinary metabolite profile and the post-dose urinary fate of acetaminophen. Our findings strongly suggest that a person's capacity for acetaminophen sulfonation can be significantly reduced by competitive p-cresol sulfonation with p-cresol known to be produced from protein-derived tyrosine in reactions involving the gut bacteria. Given the range of substances for which sulfonation is important, this finding suggests a means by which the gut bacteria might influence both drug-induced responses and disease development. Furthermore, given that acetaminophen is so widely used and has been extensively studied over many years, our novel findings provide a remarkable demonstration of the power and potential of the pharmacometabonomic approach, which we hope will eventually be used to improve drug treatment outcomes. With a view to the future practicality of this exciting new approach, it is also encouraging that the present result was obtained without the potential advantage of a standard diet and using only 'snap-shot' pre-dose urine samples. Furthermore, we envisage that rapidly growing recognition of the multiple metabolic interactions between humans and their gut symbionts, and the potential significance of the latter in regard to drug efficacy and adverse drug reactions, will lead to a revolution in the way that drugs are developed. We also envisage that, in certain cases, the gut bacteria will be deliberately manipulated by some prior or co-treatment in order to improve drug treatment outcomes. SUPPLEMENTARY DATA

The NMR-derived data reported above are based on the original (2003) spectra with the post-dose data being derived using the resolution-enhancement approach. Unless otherwise specified, it is these data that are referred to. However, comparisons with results obtained by different methods and comparisons of the 2003 and 2007 data are also provided. Ethnicity of subjects

Ninety eight of the ninety nine subjects were described as 'white'. The remaining subject was described as 'white/Mexican'. Comparison of acetaminophen metabolite data for the two post-dose urine collections

1 g of acetaminophen is ca. 0.0066 moles (F.Wt. acetaminophen = 151.17). The average percentages of this dose recovered (as any acetaminophen-related compound) in the 0-3 h and 3-6 h urine collections were estimated as 27.9 and 27.4 respectively (i.e. approximately 0.0018 moles was recovered in each 3 h collection). Furthermore, taken together, S and G typically accounted for ca. 85% of the total number of moles of acetaminophen-related compounds recovered in each post-dose collection.

The average S/G ratios for the two post-dose collections were found to be 0.711 (0-3 h) and 0.532 (3-6 h) respectively with the S/G value for each individual subject always being lower in the 3-6 h collection than in the 0-3 h collection. Furthermore, there was a clear correlation between the S/G 0-3 h and S/G 3-6 h data (r = 0.927). However, a plot of these data revealed that one subject did not fit the normal correlation and it was judged, from this and from other indications, that this subject had not fully collected his 0-3 h sample. With this subject excluded, the correlation between S/G 0-3 h and S/G 3-

6 h was slightly improved (r = 0.948).

The decrease in S/G ratios between the 0-3 and 3-6 h collections suggests that the 1 g acetaminophen dose is sufficient to cause a temporary decrease in (hepatic) sulfonation capacity. However, since the average percentages of the 1 g acetaminophen dose recovered in the two post-dose urine collections were estimated to be very similar (see note above), the data suggest that, in the 3-6 h collection, increased glucuronidation compensates for decreased sulfonation.

Inter-relationships between different acetaminophen metabolite parameters within an individual collection Having excluded the 0-3 h data for the one subject who appeared not to have collected his 0-3 h urine properly, the correlation coefficients between the observed S/G ratios and the amount of S excreted were found to be 0.510 (0-3 h) and 0.835 (3-6 h) respectively. Likewise, the correlation coefficients between S/G and the amount of G excreted were found to be -0.738 (0-3 h) and -0.803 (3-6 h) respectively. Relationships between the measured urinary parameters and age, body mass and order of sample analysis

Notes to table: The pre-dose, 0-3 h and 3-6 h samples were prepared and analysed as separate lots. Within each lot, the samples were analysed in order of increasing subject number. Age was determined approximately as [(year of study) - (year of birth)].

Calculations and considerations regarding the potential significance of p-cresol sulfonation in regard to amount of acetaminophen sulfate produced In the 0-3 h collection, the average number of moles of acetaminophen sulfate (S) recovered was found to be 0.00062 and, excluding the one subject where there was strong evidence of incomplete sample collection, the range of values was found to be 0.00031 - 0.00087 moles S. Thus, for this collection, and with the one subject excluded, the difference between the two extremes was 0.00056 moles S. More importantly, and still working with the one exclusion, the 25 subjects showing the highest S/G values were found to have excreted on average 0.00073 moles of S whilst the 25 subjects showing the lowest S/G values were found to have excreted on average 0.00054 moles of S. The difference between these values is 0.00019 moles of S.

In the 3-6 h collection, the average number of moles of S recovered was found to be 0.00051 and the range of values was found to be 0.00025 - 0.00079 moles of S. Thus, for this collection, the difference between the two extremes was 0.00054 moles of S. Furthermore, the 25 subjects showing the highest S/G values were found to have excreted on average 0.00065 moles of S whilst the 25 subjects showing the lowest S/G values were found to have excreted on average 0.00039 moles of S. The difference between these values is 0.00026 moles of S. In our original ¹ H NMR spectra of the pre-dose urine samples, the maximum value for (integral PCS methyl peak)/(integral creatinine methylene peak) was found to be 0.182 and the ratio of these two peaks was subsequently found to be virtually unchanged when full relaxation was allowed between NMR acquisition pulses. Thus, the maximum observed pre-dose PCS/creatinine mole ratio was ca. 0.12 (since 0.182 x 2/3 = 0.121).

From the Geigy Scientific Tables (Volume 1 , eighth edition, 1981 , edited by C. Lentner) it is known that men excrete on average ca. 16 mmoles of creatinine daily (this value being determined from sampling 8 men in the age range of 20 - 45 years). Then assuming, for the purposes of this calculation, that for any one individual, the rates of creatinine and PCS excretion do not vary throughout the day, the maximal rate of PCS urinary excretion observed in our study would be 0.12 x 0.016 = 0.0019 moles/day or 0.00024 moles per 3 h period. [This agrees reasonably well but not perfectly with the maximum value for 'p-cresol' urinary excretion given in the Geigy tables (same volume and edition), where the extreme range is quoted as 0.59 - 1.08 mmol/day and where there is a note to the effect that p-cresol is excreted mainly as the sulfate and glucuronide]. In the absence of other information, the maximal rate of pre-dose p-cresol sulfonation is assumed to be the same as the maximal rate of pre-dose PCS urinary excretion (i.e. 0.0019 moles/day or 0.00024 moles per 3 h period) and, clearly, this is potentially significant in relation to the difference in average acetaminophen sulfate excretion between the high and low S/G groups.

Consider now the pre-dose (PCS methyl)/(creatinine methylene) integral ratio cut-off of 0.06 (see Figure 3). With this integral ratio, the PCS/creatinine mole ratio would be ca. 0.04 and the assumed pre-dose rate of p-cresol sulfonation would be ca. 0.04 x 0.016 = 0.00064 moles/day or 0.00008 moles per 3 hour period. The potential competitive significance of this relative to the difference in average acetaminophen sulfate excretion between the high and low S/G groups probably depends on a cumulative effect of ongoing p-cresol sulfonation on hepatic sulfonation capacity. Thus, for instance, we would estimate that, over the 9 hours preceding acetaminophen administration, 0.00024 moles of p-cresol sulfate would be produced by a subject at the 0.06 integral ratio cut-off.

PLS (Projection to Latent Structure) analysis

After repeated model building attempts, using both segmented and full resolution pre- dose spectral data, the only putative PLS-derived finding, which could have arisen by chance, was a weak negative correlation (correlation coefficient -0.321) between pre- dose creatinine (total area normalised segmented pre-dose data) and the total excretion of acetaminophen-related metabolites over the first post-dose collection period, with the latter quantity estimated as described in the supplementary methods.

In view of our subsequent findings, the relative lack of success derived from the PLS- based approach, which is used routinely in metabonomics, may potentially be explained by the non-linearity of the pre-to-post dose relationship shown in Figure 3. Thus, it is only when the pre-dose PCS/Creatinine integral ratio is high that the S/G ratio may be predicted and the overall predictive power of a PLS model of the data would be expected to be low. Another potentially confounding factor for the PLS analyses would be the partial correlation between p-cresol sulfate and phenylacetylglutamine.

PCA of segmented pre-dose ¹ H NMR spectral data (all subjects) As described in the supplementary methods, PCA was carried out on the original (2003) pre-dose ¹ H NMR spectra (all subjects included) after dividing each spectrum into consecutive 0.04 ppm-wide segments, integrating across each segment and normalising the integrals to give a constant value for a selected spectral region. In one case, PCA was performed after the integrals had been normalised to give a constant value for the δ 4.07 - 4.05 region, which principally contains the creatinine methylene singlet. PCA was also performed after the integrals had been normalised to give a constant value for the δ 3.07 - 3.03 region, which principally contains the creatinine methyl singlet. In examining the results of these analyses, the 99 subjects were assigned to three colour-coded classes according to the magnitude of their individual S/G (0-3 h) values. Thus, the 25 subjects with the greatest S/G (0-3 h) values were assigned to class 1 (blue) and the 25 subjects with the lowest S/G (0-3 h) values were assigned to class 3 (red). The remaining 49 subjects were assigned to class 2 (green). As would be expected, very similar results were produced by the two normalisation methods. In both cases, the optimal Class 1 - Class 3 separation was found on PC2, which represented between 5 and 6 % of the variance in each case and was dominated by the spectral segments relating to phenylacetylglutamine (PAG) with a lesser contribution from p-cresol sulfate (PCS). With a Bonferroni correction of 30, the P value for the 95% level of confidence becomes 0.0017 and the Mann Whitney test showed significant Class 1 - Class 3 separation on PC2 with P values of 0.0013 and 0.0011 being obtained for the two normalisation methods (creatinine methyl and creatinine methylene) respectively. Selected scores and loadings plots obtained from the PCA of the δ 3.07 - 3.03 normalised data are shown in Figure 7.

Subjects identified for exclusion

As noted elsewhere, the usual approach was for the univariate statistical analyses to be performed firstly with all subjects included and then with a number of subjects excluded. In total, twenty one subjects were identified for exclusion because of actual or suspected non-compliance with the study protocol. The particular reasons for making these exclusions were as follows:

One subject was excluded because of an apparently incomplete 0-3 h sample collection, which was contrary to the study protocol and surmised from the post-dose NMR data. Fifteen subjects were excluded for potential recent use of acetaminophen, which was contrary to the study protocol and surmised from UPLC-MS analysis of the pre-dose samples. One of these subjects would also have been excluded because the NMR analysis of his pre-dose revealed the presence of ketone bodies, which suggested fasting or adherence to an abnormal diet such as the Atkin's diet. This was also contrary to the study protocol, which indicated that the subjects should eat normally. Five subjects were excluded because of a small delay in changing between the two post-dose collections (noted during study), which was contrary to the study protocol. One of these subjects would also have been excluded for apparent recent consumption of alcohol, which was contrary to the study protocol, and was detected from the NMR analysis of his pre-dose and 0-3 h urine samples. Numerical data checks (all subjects included)

1) Pre-dose urine

Table of correlation coefficients

Notes to table:

a. Manual analysis of the 2003 NMR spectra was performed twice (see 'initial' and 'final' above). The data reported in the paper are those derived from the final (optimised) determination. b. The Matlab-derived data are only approximations to the true ratios because local baseline corrections were not applied prior to performing the integrations. Instead, an overall baseline correction was applied to each spectrum before it was loaded into Matlab. c. In the combined UPLC-MS and NMR analysis performed in 2007, UPLC-MS was used to obtain a measure of the relative amounts of PCS and hippurate in each sample whilst NMR was used to obtain a measure of the relative amounts of hippurate and creatinine in each sample. The appropriate NMR and UPLC-MS- derived measurements were then multiplied to obtain a measure of the relative amounts of PCS and creatinine in each sample.

2) Post-dose urine

Tables of correlation coefficients

Notes to these tables:

a. 'acetyls' denotes use of the resolution-enhanced acetyls peaks in the quantitation. b. 'aromatics' denotes that the quantitation was based on the use of the relevant aromatic peaks without resolution enhancement. c. '(1)' and '(2)' denote the first and second determinations of S/G respectively where both determinations were based on the 2003 spectra and integration of the relevant resolution-enhanced acetyls peaks. d. 7.32 ppm cluster/TSP' and 7.15 ppm cluster/TSP' denote the relevant integral ratios. The excellent 2003 vs. 2007 correlation coefficients for these parameters indicate reliable sample preparation.

Methods

The study volunteers were nominally healthy men aged between 18 and 64 years old. Urine samples were prepared for NMR analysis by mixing 440 μl of urine with 220 μl of phosphate buffer (pH ca. 7.4 to which sodium azide had been added as an antibacterial preservative) and the mixture centrifuged to remove suspended particulates. 550 μl of 'clear' buffered urine was transferred to a sample vial and 55 μl of a TSP/D ₂ O solution added to give a final TSP concentration of 1 mM. TSP (sodium 3-trimethylsilyl-[2,2,3,3- ² H ₄ ]-1 -propionate) is a chemical shift reference compound (δ 0) used in the NMR experiment and the D ₂ O provided a field/frequency lock for the NMR spectrometer. The ¹ H NMR spectra of the prepared urine samples were acquired at 303K on a Bruker Avance 600 NMR spectrometer equipped with a flow probe and operated by means of the 'Xwinnmr' software (all from Bruker Biospin, Rheinstetten, Germany). The 'noesypresat' pulse sequence was used to suppress the water signal and the spectral acquisitions automated using the 'lconnmr' software and a 'BEST' sample changer (both Bruker Biospin). Using Xwinnmr, 1 Hz line-broadening was applied to the pre-dose ¹ H NMR spectra by means of an exponential multiplication of the free induction decay signal and these time-domain data were Fourier-transformed to frequency-domain spectra with a single zero-filling and with a digital filter used to reduce the size of the residual water signal. The resulting spectra were manually phased to give an even baseline around the NMR signals and the baseline of each spectrum was manually adjusted to zero intensity using a straight-line baseline correction algorithm. The chemical shift scale was set by assigning the value of δ 0 to the signal from the added reference compound (TSP). The post-dose ¹ H NMR spectra were processed similarly, using 0.3 Hz line broadening, and subsequently with resolution enhancement, and a measure approximating to the mole ratio of acetaminophen sulfate (S) to acetaminophen glucuronide (G) determined by integration of the respective resolution-enhanced N-acetyl peaks. For each subject and collection period, relative measures of the amounts of S and G excreted were also determined, by reference to the added TSP and the mass of urine collected. Except where stated, these excretion values were not adjusted to excretion per unit body mass. Subsequently, the pre-dose spectra were loaded into Matlab using the 'MetaSpectra' routine provided by Dr O. Cloarec and normalised to a constant integral for the δ 4.07 to 4.05 region, which encompasses the creatinine methylene singlet. Average group spectra were then calculated in Matlab and various spectral plots were made and examined. After integrating each pre-dose spectrum over consecutive 0.04 ppm spectral segments and then normalising these integrals to a constant value for the integral for the δ 4.07 - 4.05 region, Principal Components Analysis of the integrals for the δ 9.1 - 6.9 region was performed in Pirouete 3.11 (from Infometrix, U.S.A.) using mean-centred variable scaling. The PCA was also repeated after normalisation to a constant integral for the δ 3.07 - 3.03 region, which encompasses the creatinine methyl singlet. Local baseline correction and integration of selected peaks in the pre-dose spectra was performed in Xwinnmr. Statistical significance of abstracted data was assessed using the Mann-Whitney U test. SUPPLEMENTARY METHODS

Outline of study protocol The clinical phase of the study (study PRC-03-01) was performed over a number of different days in March and April 2003 at the Pfizer Research Centre at Canterbury, Kent, U.K. Ninety-nine non-smoker men, aged between 18 and 64 years old on the day of their participation in the study, were recruited for an ethically-approved 1-day clinical trial in which each participant was identified by a subject number ranging from 1 to 99. The study protocol did not specify a standard diet but placed certain restrictions on the diet and on alcohol consumption. Volunteers were only eligible if not taking drugs, herbal medicines or dietary supplements. On the day of their participation in the study, the weight and height of each subject were recorded. The ethnicity of each subject was also recorded. Each subject provided a 'snap-shot' pre-dose urine sample and then took two 500 mg tablets of acetaminophen (paracetamol BP) by mouth with 250 ml of water. After dosing, each subject was required to collect all of the urine that he produced over two consecutive time periods, namely 0-3 hours and 3-6 hours from dosing. Each subject was requested to empty his bladder as completely as possible at the end of each post- dose time period and the mass of urine produced by each subject over each period was measured. Representative samples were stored frozen pending analysis at Imperial College London, London, U.K. with duplicate samples being sent for freezer storage at Pfizer, Sandwich, Kent, UK.

Preparation of urine samples for ¹ H NMR (Nuclear Magnetic Resonance) spectroscopic analysis

The urine samples were prepared for NMR analysis by mixing 440 μl of urine with 220 μl of phosphate buffer [an 81 :19 (v/v) mixture of 0.2 M Na ₂ HPO ₄ and 0.2 M NaH ₂ PO ₄ , pH ca. 7.4, to which 0.33 % (w/v) sodium azide had been added in order to hinder bacterial proliferation]. The urine-buffer mixture was left to stand for at least 10 minutes at room temperature and then ultra-centrifuged at 13,000 rpm for a further 10 minutes to remove suspended particulates. 550 μl of 'clear' buffered urine was transferred to an appropriate sample vial and 55 μl of a TSP/D ₂ O solution added to give a final TSP concentration of 1 mM. TSP (sodium 3-trimethylsilyl-[2,2,3,3- ² H ₄ ]-1-propionate) is a chemical shift reference compound (δ 0) used in the NMR experiment and the D ₂ O provided a field/frequency lock for the NMR spectrometer. The vials were capped and the samples frozen pending analysis. 'Blank' samples were prepared at the same time and using the same solutions. The method of preparing the blanks was identical to that used for the urine samples except that the stabilised phosphate buffer was used instead of urine.

Some workers acidify urine prior to ¹ H NMR spectroscopic analysis. However, organic sulfates such as acetaminophen sulfate (S) and p-cresol sulfate (PCS) are expected to be liable to hydrolyse to form acetaminophen and p-cresol respectively and, in comparison to the rate at neutral pH, the rate of such hydrolysis is expected to be much increased at low pH. Thus, we recommend use of the sample preparation procedure described above.

¹ H NMR spectroscopy and processing The ¹ H NMR spectra of the prepared urine samples and blanks were acquired at 600 MHz at a nominal 303K on a Bruker AVANCE 600 NMR spectrometer equipped with a flow probe and operated by means of the 'Xwinnmr' software (all from Bruker Biospin, Rheinstetten, Germany). The noesyprid ('noesypresat') pulse sequence was used to suppress the water signal during a relaxation delay of 3 s and during the pulse sequence mixing time of 0.1 s. Each spectrum was acquired using 8 dummy scans, 128 real scans, 32768 time domain points, a 7200 Hz spectral width and an acquisition time of 2.3 s per scan. The spectral acquisitions were automated using the 'iconnmr' software and a 'BEST sample changer (both Bruker Biospin). The overall run time per sample was ca. 20 minutes.

1 Hz line-broadening (Ib 1 Hz) was applied to the ¹ H NMR spectra of the pre-dose samples by means of an exponential multiplication of the free induction decay signal and these time-domain data were Fourier-transformed to frequency-domain spectra with a single zero-filling and with a digital filter (bcjnod qfil; bcfw 0.3 ppm) used to reduce the size of the residual water signal. The resulting spectra were manually phased to give an even baseline around the NMR signals and the baseline of each spectrum was manually adjusted to zero intensity using a straight-line baseline correction algorithm. The chemical shift scale was then set by assigning the value of δ 0 to the signal from the added reference compound (TSP). All these operations were performed on a Silicon Graphics computer using the 'xwinnmr' software (Bruker Biospin). The ¹ H NMR spectra of the post-dose samples were processed similarly to the above method (but with Ib 0.3 Hz instead of Ib 1 Hz and without using a digital filter) and subsequently with resolution enhancement (gaussian multiplication, Ib -1 Hz, gb 0.5, si 128K) in order to better resolve the N-acetyls peaks from acetaminophen and its metabolites.

Quantitation of acetaminophen-related compounds excreted post-dose

Conveniently, for the purposes of their urinary quantitation, acetaminophen and its principal metabolites each contain a common structural feature, an N-acetyl group that produces a single ¹ H NMR peak in the vicinity of δ 2.1 - 2.2. Thus, with the right analytical conditions, these N-acetyl signals provide a basis for measuring the total number of moles of acetaminophen-related compounds excreted. Furthermore, because there is some inter-compound variation in the exact chemical shift of the N-acetyl peak, the various signals also provide a basis for measuring the relative amounts of the different compounds, as previously described by Bales et al [Clin. Chem. 30: 1631-1636, (1984)]. However, in the present work we found that not all of the different N-acetyl signals could be readily resolved despite the use of resolution enhancement. Thus, in our present study, we simply quantified the acetaminophen-related compounds as acetaminophen sulfate (S), acetaminophen glucuronide (G) and 'other' with our final emphasis resting on determining the S/G ratio and the amounts of S and G excreted.

Although these data are not reported, the total amount of acetaminophen-related compounds excreted in the urine by each individual during each post-dose collection was first estimated, by reference to the signal for the added TSP and the mass of urine collected, as a*c, where a is [(the integral for the δ 2.210 - 2.135 region)/(TSP integral)], obtained from the non-resolution-enhanced spectrum, and c is the mass of urine collected. The use of this formula, which provides a relative measure of the total excretion of acetaminophen-related compounds, relies on the constancy of the NMR sample preparation [which was confirmed by repeated preparation and analysis of all the post-dose samples (see later and supplementary data)]. Additionally, the use, within the above formula, of urinary mass (c) in place of urinary volume relies on the different urine samples having very similar densities and this assumption was supported by density measurements on a number of representative 0-3 h samples, which indicated that urinary density varied within ± 2 % of the average value. The accuracy of the estimate obtained from the given formula also relies on the different -N-acetyls having similar T1 values and on there being no significant contributions from other compounds. Assuming these conditions are met, the measure provided by this formula should be proportional to the number of moles of acetaminophen-related compounds excreted.

The second stage in the analysis was to use the resolution-enhanced versions of the post-dose spectra to determine the proportions of the total δ 2.210 - 2.135 integral derived from S, G and 'other'. Thus, for each resolution-enhanced spectrum, after integrating the whole of the δ 2.210 - 2.135 region (which automatically assigned the total integral a value of 1) the individual fractional contributions from the S and G peaks [at ca. δ 2.182 (S) and at ca. δ 2.166 (G) respectively] were determined by further integration and recorded to two decimal places. The fractional 'other' contribution was subsequently calculated by subtracting the sum of the S and G fractions from 1. The S/G ratio was calculated from the ratio of the S and G fractions. [Our further NMR analysis, under conditions of full relaxation between acquisition pulses, indicates that these S/G ratios will be close approximations to the relevant mole ratios].

For each subject and post-dose collection period, measures of the individual amounts of S and G excreted were also determined. Thus, for example, the amount of S excreted was calculated as a ^* b*c where: a is [(the integral of the δ 2.210 - 2.135 region)/(TSP integral)], obtained from the non-resolution-enhanced spectrum; b is the fraction of the δ 2.210 - 2.135 integral that arises from S, determined from the resolution-enhanced spectrum; c is the mass of urine collected; and the use of this formula relies, as before, on the constancy of the sample preparation and on the density of the post-dose urine samples being nearly constant.

As a check on the preceding analysis, the S/G ratio determinations were also performed using direct integration of the relevant resolution-enhanced peaks and the integrations recorded without rounding. The results obtained were compared with the original data (see supplementary data). As a further check on the above determinations, further measures of S/G, amount of S excreted and amount of G excreted were obtained from the non-resolution-enhanced versions of the spectra, using the integrals of selected aromatic signals [the multiplets centred at ca. δ 7.32 (from S) and at ca. δ 7.15 (from G)], and the data obtained compared to the data derived using resolution enhancement (see supplementary data). Thus, for example, a measure of amount S excreted was determined as [(integral of δ 7.32 multiplet)/(TSP integral)]*[mass of urine collected]. Again, the use of such a formula depends on the constancy of the sample preparation method and on the near constancy of the urinary density.

The accuracy of the analyses described above depends on the absence of significant spectral overlaps from other compounds and the potential for such overlaps was expected to be lower when the determinations were based on the resolution-enhanced N-acetyls peaks rather than when the determinations were based on the aromatic signals. Thus, it is the N-acetyls-based data that are reported. However, reasonably good correlation was achieved between the two sets of data (see supplementary data), which strongly supports the validity of the reported data since any overlapping compounds would be unlikely to affect the measured N-acetyls and aromatic signals to the same extent. Additional confidence was generated by visual inspection of the shapes of the measured signals, by the manual placement of integration limits and by the use of a high field spectrometer (operating at 600 MHz for ¹ H), which increases signal dispersion compared to work at lower field strengths. No correction was made for any pre-dose signals occurring in the measured regions although the subsequent statistical analysis of the N-acetyls-derived data was performed both with and without fifteen subjects who were found by UPLC-MS (see below) to have acetaminophen metabolites in their pre-dose urine samples. However, it is noteworthy that, of these fifteen subjects, only three had sufficient levels of acetaminophen metabolites in their pre-dose samples that they were independently identified as such from the ¹ H NMR analysis, which is considerably less sensitive than the UPLC-MS analysis.

In considering potential peak overlaps, it should be recognised that, because of possible diurnal and acetaminophen-induced changes to the endogenous metabolites profile, the pre-dose spectra are not necessarily fully representative of the post-dose 'background' on which the spectra of acetaminophen and its metabolites are superimposed.

However, the pre-dose spectra do provide some indication of the potential for distortion of the measured quantities (S excretion, G excretion and S/G ratio) by pre-existing urinary components. Thus, after normalisation to constant creatinine, the N-acetyls region of each post-dose spectrum was compared with the same region in the corresponding pre-dose spectrum. Some similar comparisons were also performed in relation to the aromatic multiplets from S and G located at ca. δ 7.32 and at ca. δ 7.15 respectively. In principle, because the average S/G ratio was less than 1 , the measurements of S would be expected to be more vulnerable to overlaps than the measurements of G, with the lower S/G ratios in the 3-6 h collection increasing that vulnerability. However, the comparison of individual pre and post-dose spectra revealed that any pre-dose overlaps tended to be relatively insignificant. Furthermore, because of the pre-to-post-dose association indicated by this study, we looked, in particular, to see if any of the peaks from p-cresol sulfate (PCS) and phenylacetylglutamine (PAG) could have had an impact on our S and G determination and found that they could not. We also considered the possibility that, by competing with p-cresol for sulfonation, the dosed acetaminophen might have caused an increase in the urinary excretion of either p-cresol or p-cresol glucuronide. However, there is no possibility that the peaks arising from these two compounds could have distorted the N-acetyls-based analysis of the acetaminophen-related compounds excreted post-dose. Thus, we found by analysis of authentic p-cresol (Aldrich C85751) that its methyl group resonates at ca. δ 2.26 whilst the literature [Molecular and Biochemical Parasitology 146: 1-9 (2006)] indicates that the methyl group of p-cresol glucuronide resonates at ca. δ 2.31 [which is entirely consistent with our expectation, by analogy with the situation for acetaminophen and its metabolites, that its chemical shift would be intermediate between the corresponding peaks for p-cresol sulfate (at ca. 6 2.35) and p-cresol (at ca. δ 2.26)].

The validity of the analytical approach used is further supported by the earlier work of Bales et al [Clin. Chem. 30: 1631-1636 (1984)] who had also used ¹ H NMR spectroscopy to study the urinary excretion of acetaminophen and its metabolites and who had recorded, for reference purposes, the ¹ H NMR spectra of nine known or potential acetaminophen metabolites. They too had used the N-acetyls signals as the basis for the urinary excretion measurements and they found that their results agreed broadly with those reported by other workers using other methods (on other samples).

As reported in the supplementary data, we found that the S/G ratio and the amounts of S and G excreted did not correlate with subject body mass. Consequently, in order to avoid producing post-dose parameters that were correlated with body mass (and, thereby, to avoid the possibility of building a model or models that effectively only predicted body mass), our analysis was mainly based on using these post-dose parameters without correction to unit body mass. In any case, it would also have been questionable whether a body mass correction was appropriate for the S/G ratio. However, we also found that correcting the various parameters (S/G ratio, amount S excreted and amount G excreted) to unit body mass made very little difference to the graphical relationships between these parameters and the pre-dose PCS/creatinine ratio.

Quantitation ofp-cresol sulfate (PCS) excreted pre-dose

It was not possible to directly measure pre-dose PCS excretion because of the 'snap shot' nature of the pre-dose sample collection. Consequently, the pre-dose samples did not represent excretion over a measured period of time. Furthermore, it was not regarded as worthwhile to measure only a urinary PCS concentration because such concentrations would be expected to be markedly affected by inter-subject differences in fluid consumption. Consequently, PCS was quantified relative to creatinine using the ¹ H NMR spectra of the pre-dose urine samples, which had all been acquired and processed in an identical manner. [Creatinine is generally accepted as an internal reference compound for urinary metabolite quantitation [Clin. Chem. Acta 264: 227-232 (1997)] but the amount of creatinine excreted in the urine is known to be affected by gender and ethnicity [J. Appl. Physiol. 97: 941-947 (2004)]. However, in the present study, all the subjects were male and all but one was reported as 'white' (see supplementary data). Thus, the approach taken was deemed appropriate. Further information on the factors affecting urinary creatinine excretion, which is stated to depend primarily on muscle mass, is provided in the Geigy Scientific Tables (page 63, volume 1 , eighth edition, 1981). These factors include subject age and the meat content of the diet. In principle, for two otherwise-identical individuals of different overall size, we would expect, by simple scaling, to see different amounts of PCS and creatinine excreted but identical PCS/creatinine ratios. Thus, to a first approximation, PCS/creatinine ratios would be expected to be independent of body mass and this was supported by our own findings (see supplementary data). Another advantage of measuring PCS relative to creatinine, rather than relative to the total area of peaks in the NMR spectrum, was that this provides a parameter that is open to future investigation using conventional analytical methodologies]. For this quantitation, the PCS methyl signal at ca. δ 2.348 was chosen in preference the PCS aromatic signals centred at ca. δ 7.210 and δ 7.285 because of its greater peak height (and hence greater signal to noise ratio) and because the aromatic signals appeared to be more prone to overlap by peaks from other compounds. [However, using creatinine-normalised spectra, we were reassured to find an excellent linear relationship (correlation coefficient 0.985) between the spectral intensity at δ 7.219 (corresponding to a point on one of the aromatic peaks of PCS that appeared to be less prone to overlap) and the spectral intensity at δ 2.348. This provides strong support for using the δ 2.348 signal to quantify PCS]. For the quantitation, the creatinine methylene at ca. δ 4.06 was chosen in preference to the creatinine methyl at ca. δ 3.05 because, at a near neutral pH, it is known that the latter signal can be overlapped by the methyl signal of creatine, if creatine is present.

Initially, the integral ratio of the chosen peaks, at ca. δ 2.348 and at ca. δ 4.06 respectively, was obtained using Matlab, after applying an overall straight line baseline correction to each spectrum in Xwinnmr. Subsequently, using Xwinnmr, we set out to obtain a more accurate measurement of the PCS/creatinine integral ratio for each spectrum with local baseline corrections applied to the peaks of interest before manual integration. The integral ratios (I. R.) obtained from that procedure provide an estimate of the relative amounts of PCS and creatinine present, with the accuracy of this estimate depending on the absence of significant peak overlaps and on correct positioning of the local baseline corrections. The integral ratios obtained are not intended to be mole ratios but are relative measures that will be affected by the different number of protons giving rise to the two signals (PCS CH ₃ vs. creatinine CH ₂ ), by any relaxation differences between the two signals (since the NMR experiment was not designed to allow full relaxation between the multiple acquisition pulses) and by the slight suppression of the creatinine CH ₂ signal by the digital filter that was applied to reduce the magnitude of the residual water peaks. Further partial suppression of the creatinine CH ₂ signal could potentially arise through proton exchange with water with such exchange requiring that the creatinine methylene takes part in keto-enol tautomerism with its adjacent carbonyl group. The integral ratios obtained might potentially also be affected by the 1 Hz line broadening that was applied to increase signal to noise. However, all of the experimentally-controllable factors affecting integral ratios were constant and unchanging for our experiment and for the purposes of our analyses can be ignored. Likewise, the non-controllable factors are expected to be either constant or very nearly constant.

The percentage error associated with these integral ratio measurements is expected to be greater when the level of PCS is low. In this situation, overlaps from other compounds are potentially more significant and it becomes more difficult to determine the appropriate baseline correction and the limits of the integration. However, in the present work, accurate quantitation of low level PCS was not so important and it was more important to be confident of the quantitation when the PCS level was high. As a precaution, we looked specifically to see if the measured pre-dose PCS/creatinine integral ratios (I. R.) could have been affected by the possible presence of particular compounds (acetaminophen, acetaminophen sulfate, acetaminophen glucuronide, p- cresol and phenylacetylglutamine) and found that they could not. Furthermore, we do not envisage there having been any significant distortion of the measured pre-dose PCS/creatinine integral ratios by p-cresol glucuronide. As a further precaution, we also examined the constancy of the ratio of (integral creatinine CH ₃ )/(integral creatinine CH ₂ ) 5 obtained in the ¹ H NMR spectra of the pre-dose samples. Ignoring two spectra where there was obvious overlap of the creatinine CH ₃ signal by some other compound, the measured ratio was found to vary between ca. 1.6 and 1.9 with most values falling between 1.6 and 1.8, with no sign of any run order-related change. o Identification of relevant compounds

1) Acetaminophen

The NMR signals from this metabolite were identified by comparison of ¹ H NMR spectra of post-dose urine before and after addition of authentic acetaminophen5 (Sigma-Aldrich A5000).

2) Acetaminophen sulfate ( ¹ S')

The NMR signals from this metabolite were identified by comparison of ¹ H NMR spectra of post-dose urine before and after addition of authentic acetaminophen0 sulfate (Sigma-Aldrich UC448).

3) Acetaminophen glucuronide ( ¹ G')

The NMR signals from this metabolite were identified by comparison of ¹ H NMR spectra of post-dose urine before and after addition of authentic acetaminophen5 glucuronide (Sigma-Aldrich A4438).

4) Phenylacetylglutamine ('PAG')

The NMR signals from this metabolite were identified by comparison of ¹ H NMR spectra of pre-dose urine before and after addition of authentic0 phenylacetylglutamine (LGC certified reference material 169.01).

5) p-Cresol sulfate ('PCS')

The ¹ H NMR signals centred at ca. δ 7.210 and at ca. δ 7.285 showed the typical pattern for a para-disubstituted aromatic compound and STOCSY analysis [Cloarec5 et al. Statistical total correlation spectroscopy: an exploratory approach for latent biomarker identification from metabolic ¹ H NMR data sets. Anal. Chem. 77: 1282- 1289 (2005)] was used to identify ¹ H NMR signals originating from the same molecule. This indicated that the molecule might be p-cresol or p-cresol sulfate and that it was very unlikely to be p-cresol glucuronide. Comparison of ¹ H NMR spectra of pre-dose urine recorded before and after addition of p-cresol (Sigma-Aldrich C85751) proved that the unknown molecule was not p-cresol. Comparison of ¹ H NMR spectra of pre-dose urine recorded before and after addition of sulfatase

(Sigma-Aldrich S9626) was consistent with the molecule being p-cresol sulfate although the sulfatase used was stated to also have some glucuronidase activity. The ¹ H NMR signals from this metabolite were finally identified by comparison of ¹ H NMR spectra of pre-dose urine before and after addition of authentic p-cresol sulfate (synthesised, purified and characterised in-house). Yet further confirmation was subsequently obtained using a second lot of PCS that had been synthesised and characterised in-house without purification.

6) Creatinine

The NMR signals from this metabolite were identified by comparison of ¹ H NMR spectra of pre-dose urine before and after addition of authentic creatinine (Sigma- Aldrich C4255).

Synthesis, purification and characterisation of p-cresol sulfate (PCS) p-Cresol sulfate (PCS) was synthesised by the slow addition of chlorosulfonic acid (Fluka 26388, purity≥ 98.0%) to an ice-cooled solution of p-cresol (Sigma-Aldrich C85751 , purity 99%) in pyridine (Sigma-Aldrich 270407, purity≥ 99.9%). After removing volatiles with a rotary evaporator, column chromatography (silica, acetonitrile/methanol) was used to isolate a purified PCS fraction from the remainder of the reaction mixture. After dissolution in phosphate buffer, the identity of the isolated PCS was established by ¹ H NMR spectroscopy and by UPLC-MS [ultra performance liquid chromatography with detection of eluted compounds by mass spectroscopy; -ve ion mode; m/z 187 (molecular ion) and m/z 107 (corresponding to loss of SO ₃ )] with further ¹ H NMR analysis confirming that the material identified as PCS was not an impurity of the p-cresol used in its synthesis. A second lot of PCS was synthesised similarly but with the amount of added chlorosulfonic acid adjusted to give nearly complete conversion of p-cresol to PCS. This second lot of PCS was characterised by ¹ H NMR spectroscopy and by UPLC-MS without purification.

Further analysis of post-dose urine samples In 2007, ca. four years after the original NMR analysis, the post-dose urine samples (which had been stored during the interim period at -40 degrees C) were prepared for a repeat NMR analysis. The mode of preparation, using phosphate buffer (without sodium azide) and a solution of TSP in D ₂ O, was essentially the same as that used previously but on this occasion the samples were prepared in standard 5 mm NMR tubes rather than in vials and the final sample volume was 550 microlitres rather than 605 microlitres. ¹ H NMR spectra were acquired using Xwinnmr and the acquisitions automated using lconnmr and a standard 'BACS' 60 position sample changer (all Bruker Biospin). The acquisition parameters were the same as used previously except that the 'zgpr' pulse program was used instead of the 'noesyprid' program and, to save time, 64 scans were acquired per sample instead of the previous 128. After the normal processing, an S/G ratio was determined for each spectrum by integration of the relevant aromatic signal clusters at ca. δ 7.32 (S) and at ca. δ 7.15 (G) after manual application of local baseline corrections. The same signals were also used in obtaining measures of the concentrations of S and G in each individual urine sample. Thus, for each sample, values were determined for [(integral δ 7.32 cluster)/(TSP integral)] and [(integral δ 7.15 cluster)/(TSP integral)] and the results obtained were then compared with the equivalent data derived from the original 2003 spectra (see supplementary data). These comparisons provided a check on the reliability of the original sample preparation and also, thereby, on the reliability of the S and G excretion measurements derived from the original 2003 spectra.

Further checks were also performed to investigate how certain spectral parameters changed when the original noesyprid NMR experiment was replaced by the simpler zgpr experiment with full relaxation between acquisition pulses (recycle time > 20 s). Additionally, selected samples, having particularly distinctive spectral characteristics, were run manually to confirm that the NMR automation runs had maintained the correct sample order. In 2008, five years after the original NMR analysis, the freezer-stored 3-6 h samples were prepared for analysis on a UPLC-MS system which was also equipped with a diode array detector. The peaks arising from acetaminophen sulfate and acetaminophen glucuronide were confirmed from their characteristic mass spectra and were found to be well resolved from one another. For each sample, single wavelength (254 nm) data were extracted from the diode array results and used to calculate an acetaminophen sulfate/acetaminophen glucuronide ratio (S/G) for each sample. These ratios, which were independently determined, were found to correlate extremely well (correlation coefficient 0.991) with the S/G ratios derived from the original NMR analysis (2003; resolution-enhanced acetyls) of the 3-6 h urine samples. No outliers were found. This independent analysis fully supports the original NMR-based approach to measuring the S/G ratio in the post-dose samples.

Further analysis ofpre-dose urine samples

In 2007, ca. four years after the original NMR analysis, the pre-dose urine samples (which had been stored at -40 degrees C) were prepared for a repeat NMR analysis. The mode of preparation, using phosphate buffer (without sodium azide) and a solution of TSP in D ₂ O, was essentially the same as that used previously but on this occasion the samples were prepared in standard 5 mm NMR tubes rather than in vials and the final sample volume was 550 microlitres rather than 605 microlitres. ¹ H NMR spectra were acquired at 600 MHz using Xwinnmr and the acquisitions automated using lconnmr and a standard 'BACS' 60 position sample changer (all Bruker Biospin). The spectra were processed in Xwinnmr without the use of a digital filter and for each sample the PCS/creatinine ratio was determined by manual integration of the peaks at ca. δ 2.348 and ca. δ 4.06 after local baseline correction. The measured ratios were then compared with the PCS/creatinine ratios obtained from the original (2003) spectra (see supplementary data). To support the subsequent UPLC-MS analysis (see below) a hippurate/creatinine integral ratio was similarly obtained from each spectrum using the hippurate peak cluster at ca. δ 7.56 and the creatinine CH ₂ peak at ca. δ 4.06. Additionally, selected samples, having particularly distinctive spectral characteristics, were analysed manually to check that the NMR automation runs had maintained the correct sample order.

The four years old pre-dose urine samples were also analysed by UPLC-MS (-ve ion mode) with appropriate standards (of S ₁ G, PCS, creatinine and hippurate) used to establish retention times and to optimise the chromatography. The original aims of the UPLC-MS analysis were 1) to get an independent measure of the PCS/creatinine ratios, and 2) to identify subjects who had taken acetaminophen just prior to the start of the study by measuring the ratios S/creatinine and G/creatinine. However, creatinine was found to be poorly retained and barely detectable. Thus, the UPLC-MS method was redesigned so as to use hippurate as the internal reference point for quantitation and the 'x'/hippurate values (where x is S, G or PCS) obtained from UPLC-MS were multiplied by the relevant hippurate/creatinine conversion factors obtained from the associated NMR analysis. The PCS/creatinine ratios derived from a combination of UPLC-MS and NMR analysis were then compared with those derived solely from NMR (see supplementary data) in order to check the validity of the NMR-based quantitation of PCS/creatinine. The S/creatinine and G/creatinine values obtained were plotted against subject number and nine subjects with high values for both parameters were immediately identified as likely to have taken acetaminophen just prior to the start of the study, thereby being non- compliant with the study protocol. [Reassuringly, the three pre-dose samples that showed the highest values for S/creatinine and G/creatinine by UPLC-MS had likewise been identified from the original NMR analysis as containing the highest levels of these metabolites]. After excluding the nine subjects with the highest pre-dose values for S/creatinine and G/creatinine, the data for the remaining ninety subjects were re- examined and six further subjects were identified for exclusion on the basis of having pre-dose G/creatinine values greater than the mean + two standard deviations. In our view, the total proportion of subjects (15/99) identified for exclusion on this basis was probably somewhat higher than it needed to be. However, our preference for the second stage of the univariate statistical analysis (see later) was to exclude too many subjects rather than too few.

In 2008, ¹ H NMR analysis of selected samples indicated that the PCS methyl signal would not be overlapped by the signal that would arise from pyruvate. Further NMR analysis, at 400 MHz, showed that the integral ratio between the PCS methyl and the creatinine methylene was virtually unchanged when the recycle time between acquisition pulses was changed from 5.3 s to 21.3 s (zgpr pulse program). This finding supports the reliability of the abstracted PCS/creatinine ratios. Matlab procedures relating to Figure 2

The δ 10 to -1 region of each processed and baseline-corrected pre-dose spectrum (original data) was loaded into Matlab, using the in-house 'MetaSpectra' routine provided by Dr O. Cloarec (ex Imperial College London; now of Royal Holloway, University of London), wherein the data point interval was adjusted to 0.0002 ppm. The individual spectra were then normalised to a constant integral for the δ 4.05 to 4.07 region, which encompasses the creatinine methylene singlet, and assigned to one of three colour- coded groups according to the magnitude of their S/G 0-3 h values. Group average spectra were calculated and colour-coded plots of selected regions of the individual and group average spectra were prepared. The mode of plot preparation was to add, sequentially, the plots for the different groups and it should be noted that this could result in some obscuration of previously plotted spectra, as is clear from a comparison of Figures 2a and 2b.

Principal Components Analysis (PCA)

PCA is a pattern recognition method that is used to look for inherent similarities between objects or samples for which multivariate data have been obtained. A major strength of this method is that it is 'unsupervised' i.e. it is not directed by knowledge of some external factor or factors. In the present work, PCA was carried out on the original (2003) pre-dose ¹ H NMR spectra (all subjects included) after segmentation into consecutive 0.04 ppm-wide segments and integration across each segment. During the segmentation process, which was performed using the AMIX software (Bruker Biospin), the integrals obtained were normalised to a constant value for the δ 4.07 - 4.05 region, which principally contains the creatinine methylene singlet. PCA was then performed (using the Pirouette 3.11 software from Infometrix, USA) on the data for the δ 9.1 - 6.9 spectral region using mean-centred variable scaling and with the maximum number of extracted factors limited to 10. In examining the results of the PCA, the 99 subjects were assigned to three classes according to their individual S/G (0-3 h) values. Thus, the 25 subjects with the greatest S/G (0-3 h) values were assigned to class 1 and the 25 subjects with the lowest S/G (0-3 h) values were assigned to class 3. The remaining 49 subjects were assigned to class 2. The MATLAB function 'ranksum' from Statistics Toolbox™ was then applied to the scores for each PC to determine if there was any significant separation of the class 1 and class 3 subjects. This test is equivalent to a Mann Whitney U test. The appropriate Bonferroni correction was taken as 30 (see next section). Selected scores plots were also plotted for visual examination with the following colour-coding of the subjects: class 1 - blue; class 2 - green; class 3 - red.

[The above analysis was repeated after normalisation of the pre-dose spectral data to a constant value for the δ 3.07 - 3.03 region instead of to a constant value for the δ 4.07 - 4.05 region. The δ 3.07 - 3.03 region principally contains the creatinine methyl singlet but may also include a contribution from creatine].

Magnitude of Bonferroni correction The Bonferroni correction was used to control potential false positives arising from the multiple hypothesis testing that typically occurs in metabonomic analyses. However, the exact magnitude of the Bonferroni correction(s) that should be employed required careful consideration with different corrections being applied depending on the nature of the analysis performed.

Bonferroni correction for computational pattern recognition analyses

The rank of the pre-dose data matrix (≤ 99 when all subjects were included) sets an upper limit on the number of factors that could potentially be extracted using computational multivariate pattern recognition methods such as PCA (Principal Components Analysis) and PLS (Projection to Latent Structure). However, in the PCA performed in this study, the appropriate Bonferroni correction was taken as 30, on the basis that our typical approach to the PCA of a data set would involve three separate analyses (auto-scaling, mean-centred scaling, pareto scaling) with a maximum of 10 factors permitted to be extracted in each PCA. Bonferroni correction for our other analyses

These other analyses were based on sequentially comparing inter-group differences in the levels of each NMR-detectable pre-dose sample component. For these analyses the relevant factor was the number of NMR-quantifiable urinary components and this number is likely to change to some extent depending on the nature of the NMR equipment and experiment used. Whilst a figure of 200+ ¹ H NMR-visible metabolites per urine sample has been indicated by one source [Anal. Chem. 78: 4430 -4442 (2006)], supporting evidence was not provided and, from our own experience, our view is that this is likely to be a considerable overestimate when using, as in the present study, conventional (non- cryo) NMR probes and typical current spectrometers operating at up to 600 MHz. Furthermore, even if such a number of metabolites were visible, the number of components that would be readily quantifiable without using spectral deconvolution methods and reference spectra would be considerably less because of baseline intensity variation, noise and peak overlaps. Instead, we consider that a much more realistic estimate of the number of metabolites that are readily quantifiable by ¹ H NMR spectroscopy is provided by the April 2006 Chenomx document titled 'Targeted Profiling of Common Metabolites in Urine' by C. Vitols and H. Fu which lists 80 metabolites stated as being commonly seen in the NMR spectra of human urine (www.chenomx.com). Furthermore, even if the number of readily NMR-quantifiable urinary metabolites was as high as 200, which we doubt, use of a Bonferroni correction of 100 has been deemed acceptable in such circumstances [Metabolomics 2: 171-196 (2006)] because of the assumptions associated with the Bonferroni correction. In fact, if we follow a similar argument and have, as we suspect, less than 100 components in human urine that are readily quantifiable by ¹ H NMR spectroscopy, then a Bonferroni correction of 50 might be considered adequate. However, in this work, we have applied a correction of 100 in order to err on the side of caution.

Numerical single variable discovery procedures

Methods such as PCA (Principal Components Analysis) and PLS (Projection to Latent Structure) are aimed at simplifying the analysis of multivariate data where different variables may provide similar information. PCA is an unsupervised method aimed at finding inherent patterns in the data and is not directed by knowledge of external factors. PLS is a supervised method that is focussed on finding patterns in the data that correlate with an external variable. However, these methods are potentially less suited to the discovery of single variables that may correlate with an external variable. In order to determine if any individual components of the pre-dose spectra correlated with the variation in the S/G ratio post-dose, we applied variable discovery procedures described in international patent application WO2004038602A1 [Integrated spectral data processing, data mining and modeling system for use in diverse screening and biomarker discovery applications. Baker, D.]. Thus, the creatinine-normalised pre-dose spectra for the subjects at the two extremes of the S/G 0-3h distribution were compared (using 25 subjects from each end of that distribution) and all the integrated pre-dose spectral bands were then ordered according to their ability to discriminate these two S/G classes. In this case, after applying the relevant Bonferroni correction (100) to the most significant spectral bands, only three bands were found to be significant at the 95% level of confidence. These bands (at δ 2.347, 7.213 and 7.278) correspond to the spectral features of PCS. The procedure was repeated for the S/G 3-6 h distribution and the same bands were again the only ones found to be significant.

Univariate statistical analysis and exclusion criteria

For the key parts of the univariate statistical analyses, the approach taken was to perform the analysis a) with all 99 subjects included and b) after excluding all those subjects who did not comply with, or who did not appear to comply with, the study protocol. It was not considered practical to screen the subjects for prior use of drugs other than acetaminophen and the specific reasons adopted for subject exclusion were a) incorrect sample collection b) presence of ethanol in pre-dose sample indicating recent alcohol consumption c) presence of acetaminophen metabolites in pre-dose sample indicating recent use of acetaminophen d) presence of ketone bodies in pre-dose sample indicating that the subject was not eating normally. This led to the exclusion of 21 subjects with some subjects being excluded for more than one reason, which indicates the difficulty of performing studies in human subjects. Statistical significance was assessed using the Mann-Whitney U test with these analyses being performed in SPSS 14 (SPSS Inc.), R 2.5.0 and R 2.5.1 [R Development Core Team (2007). R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. ISBN 3-900051-07-0, URL http://www.R-project.org]. This analysis was provided by the Statistical Advisory Service of Imperial College London. However, the preceding statements do not apply to the statistical analysis of the PCA scores, which has been described in a previous section.

References (1) Evans, W. E.; Relling, M. V. Pharmacogenomics: translating functional genomics into rational therapeutics. Science (1999) 286, 487-491.

(2) Evans, W. E.; Johnson, J. A. Pharmacogenomics: the inherited basis for interindividual differences in drug response. Annu. Rev. Genomics Hum. Genet.

(2001) 2, 9-39.

(3) Spear, B. B.; Heath-Chiozzi, M.; Huff, J. Clinical application of pharmacogenetics.

TRENDS MoI. Med. (2001) 7, 201-204.

(4) Pagliarulo, V.; Datar, R. H.; Cote, R. J. Role of genetic and expression profiling in pharmacogenomics: the changing face of patient management. Curr. Issues MoI.

Biol. (2002) 4, 101-110.

(5) Evans, W. E.; Relling, M. V. Moving towards individualized medicine with pharmacogenomics. Nature (2004) 429, 464-468.

(6) Marsh, S.; McLeod, H. L. Pharmacogenomics: from bedside to clinical practice.

Hum. MoI. Genet. (2006) 15, R89-R93.

(7) Nebert, D. W.; Jorge-Nebert, L.; Vesell, E. S. Pharmacogenomics and "individualized drug therapy":high expectations and disappointing achievements.

Am. J. Pharmacogenomics (2003) 3, 361-370.

(8) Nebert, D. W.; Zhang, G.; Vesell, E. S. (2008) From human genetics and genomics to pharmacogenetics and pharmacogenomics: past lessons, future directions. Drug Metab. Rev. (2008) 40, 187-224.

(9) Clayton, T. A.; Lindon, J. C; Cloarec, O.; Antti, H.; Charuel, C; Hanton, G.;

Provost, J.-P.; Le Net, J.-L; Baker, D.; Walley, R. J.; Everett, J. R.; Nicholson, J. K. Pharmaco-metabonomic phenotyping and personalized drug treatment. Nature (2006) 440, 1073-1077.

(10) Lindon, J. C; Holmes, E.; Bollard, M. E.; Stanley, E. G.; Nicholson; J. K.

Metabonomic technologies and their applications in physiological monitoring, drug safety assessment and disease diagnosis. Biomarkers (2004) 9, 1 -31.

(1 1) Nebert, D. W.; Vesell, E. S. Can personalized drug therapy be achieved? A closer look at pharmaco-metabonomics. TRENDS Pharmacol. Sci. (2006) 27, 580-586.

(12) Li, H.; Ni, Y.; Su, M.; Qiu, Y.; Zhou, M.; Qiu, M.; Zhao, A.; Zhao, L; Jia, W.

Pharmacometabonomic phenotyping reveals different responses to xenobiotic intervention in rats. J. Proteome Res. (2007) 6, 1364-1370.

(13) Ala-Korpela, M. Potential role of body fluid ¹ H NMR metabonomics as a prognostic and diagnostic tool. Exp. Rev. MoI. Diag. (2007) 7, 761 -773.

(14) Newson, R. B.; Shaheen, S. O.; Chinn, S.; Burney, P. G. J. Paracetamol sales and atopic disease in children and adults: an ecological analysis. Eur. Respir. J.

(2000), 16, 817-823.

(15) Cummings, A. J.; King, M. L; Martin, B. K. A kinetic study of drug elimination: the excretion of paracetamol and its metabolites in man. Br. J. Pharmac. Chemother. (1967), 29, 150-157.

(16) Clements, J. A.; Heading, R. C; Nimmo, W. S.; Prescott, L. F. Kinetics of acetaminophen absorption and gastric emptying in man. Clin. Pharmacol. Ther.

(1978) 24, 420-431.

(17) Prescott, L. F. (1980) Kinetics and metabolism of paracetamol and phenacetin.

Br. J. CHn. Pharmacol. (1980) 10, 291S-298S.

(18) Forrest, J. A. H.; Clements, J. A.; Prescott, L. F. Clinical pharmacokinetics of paracetamol. CHn. Pharmacokinet. (1982) 7, 93-107.

(19) Lau, G. S. N.; Critchley, J. A. J. H. The estimation of paracetamol and its major metabolites in both plasma and urine by a single high-performance liquid chromatography assay. J. Pharm. Biomed. Anal. (1994) 12, 1563-1572.

(20) Critchley, J. A. J. H.; Nimmo, G. R.; Gregson, C. A.; Woolhouse, N. M.; Prescott,

L. F. Inter-subject and ethnic differences in paracetamol metabolism. Br. J. CHn.

Pharmacol. (1986) 22, 649-657.

(21) Burchell, B.; Coughtrie, M. W. H. Genetic and environmental factors associated with variation of human xenobiotic glucuronidation and sulfation. Environ. Health Perspect. (1997) 105, 739-747.

(22) Trygg, J.; Holmes, E.; Lundstedt, T. Chemometrics in Metabonomics. J.

Proteome Res. (2007) 6, 469-479. (23) Smith, E. A.; Macfarlane, G. T. Formation of phenolic and indolic compounds by anaerobic bacteria in the human large intestine. Microb. Ecol. (1997) 33, 180- 188.

(24) Broadhurst, D. I.; KeII, D. B. Statistical strategies for avoiding false discoveries in metabolomics and related experiments. Metabolomics (2006) 2, 171-196.

(25) Klaassen, C. D.; Boles, J. W. The importance of 3'-phosphoadenosine 5'- phosphosulfate (PAPS) in the regulation of sulfation. FASEB J. (1997) 11 , 404- 418.

(26) Gamage, N.; Barnett, A.; Hempel, N.; Duggleby, R. G.; Windmill, K. F.; Martin, J.

L.; McManus, M. E. Human sulfotransferases and their role in chemical metabolism. Toxicol. Sci. (2006) 90, 5-22.

(27) Morinaga, Y.; Fuke, C; Arao, T.; Miyazaki, T. Quantitative analysis of cresol and its metabolites in biological materials and distribution in rats after oral administration. Legal Medicine (2004) 6, 32-40.

(28) Schepers, E.; Meert, N.; Glorieux, G.; Goeman, J.; Van der Eycken, J.;

Vanholder, R. P-cresylsulphate, the main in vivo metabolite of p-cresol, activates leucocyte free readical production. Nephrol. Dial. Transplant (2007) 22, 592-596. (29) Martinez, A. W.; Recht, N. S.; Hostetter, T. H.; Meyer, T. W. Removal of p-cresol sulphate by hemodialysis. J. Am. Soc. Nephrol. (2005) 16, 3430-3436.

(30) de Loor, H.; Bammens, B.; Evenepoel, P.; De Preter, V.; Verbeke, K. Gas- chromatographic-mass spectrometric analysis for measurement of p-cresol and its conjugated metabolites in uremic and normal serum. CHn. Chem. (2005) 51, 1535-1538.

(31) Ramakrishna, B. S.; Gee, D.; Weiss, A.; Pannall, P.; Roberts-Thomson, I. C;

Roediger, W. E. W. Estimation of phenolic conjugation by colonic mucosa. J.

CHn. Pathol. (1989) 42, 620-623.

(32) Hopkins, M. J.; Sharp, R.; Macfarlane, G. T. Variation in human intestinal microbiota with age. Dig. Liver Dis. (2002) 34, Suppl 2:S12-S18.

(33) Galinsky, R. E.; Corcoran, G. B. Influence of advanced age on the formation and elimination of acetaminophen metabolites by male rats. Pharmacology (1986)

32, 313-320.

(34) Rowland, I. The influence of the gut microflora on food toxicity. Proc. Nutr. Soc.

(1981) 40, 67-74.

(35) Rowland, I. R. Reduction by the gut microflora of animals and man. Biochem.

Pharmacol. (1986) 35, 27-32.

(36) Rowland, I. R. Interactions of the gut microflora and the host in toxicology.

Toxicol. Pathol. (1988) 16, 147-153. (37) Mikov, M. The metabolism of drugs by the gut flora. Eur. J. Drug Metab. Pharmacokinet. (1994) 19, 201-207.

(38) Nicholson, J. K.; Wilson, I. D. Understanding 'global' systems biology: metabonomics and the continuum of metabolism. Nature Rev. Drug Discov. (2003) 2, 668-676.

(39) Nicholson, J. K.; Holmes, E.; Wilson, I. D. Gut microorganisms, mammalian metabolism and personalised health care. Nat. Rev. Microbiol. (2005) A.O.P.

(40) Ley, R. E.; Turnbaugh, P. J.; Klein, S.; Gordon, J. I. Human gut microbes associated with obesity. Nature (2006) 444, 1022-1023.

(41) Jia, W.; Li, H.; Zhao, L.; Nicholson, J. K. Gut microbiota: a potential new territory for drug targeting. Nature Rev. Drug Discov. (2008) 7, 123-129.

(42) Whitcomb, D. C; Block, G. D. Association of acetaminophen hepatotoxicity with fasting and ethanol use. JAMA (1994) 272, 1845-1850.

(43) Kawakami, K.; Kojima, K.; Makino, I.; Kato, I.; Onoue, M. Fasting enhances p- cresol production in the rat intestinal tract. Exp. Anim. (2007) 56, 301-307.

(44) Tanaka, E.; Yamazaki, K.; Misawa, S. Update: the clinical importance of acetaminophen hepatotoxicity in non-alcoholic and alcoholic subjects. J. CHn. Pharm. Ther. (2000) 25, 325-332.

(45) Thompson, D. C; Perera, K.; Fisher, R.; Brendel, K. Cresol isomers: comparison of toxic potency in rat liver slices. Toxicol. Appl. Pharmacol. (1994) 125, 51-58.

(46) Thompson, D. C; Perera, K.; London, R. Quinone methide formation from para isomers of methylphenol (cresol), ethylphenol, and isopropylphenol: relationship to toxicity. Chem. Res. Toxicol. (1995) 8, 55-60.

(47) Strott, C. A. Sulfonation and molecular action. Endocr. Rev. (2002) 23, 703-732. (48) Coughtrie, M. W. H. Sulfation through the looking glass - recent advances in sulfotransferase research for the curious. Pharmacogenomics J. (2002) 2, 297- 308.

(49) Kauffman, F. C. Sulfonation in Pharmacology and Toxicology. Drug Metab. Rev.

(2004) 36, 823-843.

(50) Van Der Marel, C; Anderson, B. J.; Tibboel, D.; Holford, N. H. G. Modelling paracetamol urine metabolites. Paediatric Anaesthesia (2002) 12, 827.

(51) Hempel, N.; Gamage, N.; Martin, J. L.; McManus, M. E. Human cytosolic sulfotransferase SULT1A1. Int. J. Biochem. Cell Biol. (2007) 39, 685-689.

(52) Ling, W. H.; Hanninen, O. Shifting from a conventional diet to an uncooked vegan diet reversibly alters faecal hydrolytic activities in humans. J. Nutr. (1992)

122, 924-930. (53) Sivsammye, G.; Sims, H. V. Presumptive identification of Clostridium difficile by detection of p-cresol in prepared peptone yeast glucose broth supplemented with p-hydroxyphenylacetic acid. J. Clin. Microbiol. (1990) 28, 1851-1853.

(54) Bone, E.; Tamm, A.; Hill, M. The production of urinary phenols by gut bacteria and their possible role in the causation of large bowel cancer. Am. J. CHn. Nutr.

(1976) 29, 1448-1454.

(55) Johnson, G. A.; Barsuhn, K. J.; McCaII, J. M. Sulfation of minoxidil by liver sulfotransferase. Biochem. Pharmacol. (1982) 31, 2949-2954.

(56) Kim, S. Y.; Laxmi, Y. R.; Suzuki, N.; Ogura, K.; Watabe, T.; Duffel, M. W.;

Shibutani, S. Formation of Tamoxifen-DNA adducts via O-sulfonation, not O- acetylation, of α-hydroxytamoxifen in rat and human livers. Drug Metab. Dispos. (2005) 33, 1673-1678.

(57) Thomas, N. L.; Coughtrie, M. W. H. Sulfation of apomorphine by human sulfotransferases: evidence of a major role for the polymorphic phenol sulfotransferase, SULT1A1. Xenobiotica (2003) 33, 1139-1148.

(58) Adams, R. F.; Murray, K. E.; Earl, J. W. High levels of faecal p-cresol in a group of hyperactive children. Lancet (1985) 326, 1313.

(59) Cao, L; Kirk, M. C; Coward, L. U.; Jackson, P.; Whitaker, J. N. p-Cresol sulphate is the dominant component of urinary Myelin Basic Protein Like Material. Arch. Biochem. Biophys. (2000) 377, 9-21.

(60) Jackson, P. L.; Cao, L.; Blalock, J. E.; Whitaker, J. N. The requirement of ammonium or other cations linked with p-cresol sulfate for cross-reactivity with a peptide of myelin basic protein. Arch. Biochem. Biophys. (2003) 418, 119-124.

(61) Steventon, G. B.; Heafield, M. T.; Waring, R. H.; Williams, A. C; Sturman, S.;

Green, M. Metabolism of low-dose paracetamol in patients with chronic neurological disease. Xenobiotica (1990) 20, 117-122.

(62) Bradley, H.; Waring, R. H.; Emery, P.; Arthur, V. Metabolism of low-dose paracetamol in patients with rheumatoid arthritis. Xenobiotica (1991) 21 , 689- 693.

(63) Alberti, A.; Pirrone, P.; Elia, M.; Waring, R. H.; Romano, C. Sulfation deficit in

"low-functioning" autistic children: a pilot study. Biol. Psychiatry (1999) 46, 420-

424.

(64) Horvath, K.; Perman, J. A. Autism and gastrointestinal symptoms. Cυrr.

Gastroenterol. Rep. (2002) 4, 251-258.

(65) Song, Y.; Liu, C; Finegold, S. M. Real-time PCR quantitation of Clostridia in feces of autistic children. Appl. Environ. Microbiol. (2004) 70, 6459-6465. (66) Parracho, H. M. R. T.; Bingham, M. O.; Gibson, G. R.; McCartney, A. L Differences between the gut microflora of children with autistic spectrum disorders and that of healthy children. J. Med. Microbiol. (2005) 54, 987-991.

(67) Strang, R. R. The association of gastro-duodenal ulceration and Parkinson's disease. Med. J. Aust. (1965) 1 , 842-843.

(68) Goodhart, P. J.; DeWoIf, W. E.; Kruse, L. I. Mechanism-based inactivation of dopamine beta-hydroxylase by p-cresol and related alkylphenols. Biochem. (1987) 26, 2576-2583.

(69) Elliott, A. A.; Elliot, J. R. Voltage-dependent inhibition of RCK1 K ⁺ channels by phenol, p-cresol, and benzyl alcohol. MoI. Pharmacol. (1997) 51, 475-483.

(70) Vanholder, R.; De Smet, R.; Lesaffer, G. p-Cresol: a toxin revealing many neglected but relevant aspects of uraemic toxicity. Nephrol. Dial. Transplant. (1999) 14, 2813-2815.

(71) Gaikwad, N. W.; Bodell, W. J. Formation of DNA adducts by microsomal and peroxidase activation of p-cresol: role of quinone methide in DNA adduct formation. Chem. Biol. Inter. (2001) 138, 217-229.

(72) Calderόn-Guzman, D.; Hernandez-lslas, J. L.; Espitia Vazquez, I. R.; Barragan- Mejia, G.; Hernandez-Garcia, E.; Del Angel, D. S.; Juarez-Olguin, H. Effect of toluene and cresols on Na ⁺ , K ⁺ -ATPase, and serotonin in rat brain. Reg. Toxicol. Pharmacol. (2005) 41 , 1-5

(73) Yan, Z.; Zhong, H. M.; Maher, N.; Torres, R.; Leo, G. C; Caldwell, G. W.;

Huebert, N. Bioactivation of 4-methylphenol (p-cresol) via cytochrome P450- mediated aromatic oxidation in human liver microsomes. Drug Metab. Dispos. (2005) 33, 1867-1876.