Microviridae bacteriophages and anthranilic acid in the diagnosis, prognosis and treatment of food intake related disorders

Abstract

The present invention relates to methods for diagnosis and monitoring of food addiction, obesity and related disorders, as well as for the evaluation of treatments directed to said disorders, based on the determination and comparison of Microviridae bacteriophages, preferably Gokushovirus, levels. It also relates to kits and uses for said purposes. Furthermore, the invention relates to anthranilic acid for use in the treatment or prevention of food addiction, obesity and related disorders, and non-therapeutic use in body fat reduction.

Description

[0001] Microviridae bacteriophages and anthranilic acid in the diagnosis, prognosis and treatment of food intake related disorders

[0002] The present invention belongs to the field of clinical diagnosis and prognosis. Specifically, the present invention relates to methods for diagnosis, monitoring of food addiction, obesity and related disorders, as well as for the evaluation of treatments directed to said disorders, based on the determination and comparison of Microviridae bacteriophages, preferably Gokushovirus, levels. Furthermore, the invention relates to anthranilic acid for use in the treatment or prevention of food addiction, obesity and related disorders.

[0003] BACKGROUND ART

[0004] Obesity has reached global pandemic levels and has become the leading preventable cause of death worldwide. Worldwide obesity has almost tripled in the last 50 years, with more than 1.9 billion adults (39%) being overweight and 650 million (13%) having obesity in 2016. It is the major risk factor for noncommunicable diseases, including cardiovascular disease, numerous cancers, diabetes, depression, or Alzheimer’s disease, and is the biggest driver of health-care costs, thereby contributing to a decrease in both life expectancy and quality of life (Bluher, M. Obesity: global epidemiology and pathogenesis. Nat Rev Endocrinol 15, 288-298 (2019)).

[0005] Food addiction plays a significant role in the current obesity pandemic and has increased its prevalence in recent times. It is a complex, maladaptive eating behaviour which results from an imbalance between the intestinal and extraintestinal homeostatic mechanisms and brain hedonic mechanisms resulting in a loss of control over food intake (Gupta, A., Osadchiy, V. & Mayer, E. A. Brain-gut-microbiome interactions in obesity and food addiction. Nat Rev Gastroenterol Hepatol 17, 655- 672 (2020)).

[0006] Similar to a substance-based addiction, individuals afflicted with this condition exhibit addiction-like compulsive consumption of food, particularly high palatable foods rich in fat and / or sugar, which can trigger changes in the brain’s reward circuitry. This behavioral anomaly is closely linked with obesity and other eating disorders, posing significant challenges in terms of effective treatment and imposing substantial socioeconomic burdens globally. Individuals diagnosed with food addiction using the Yale Food Addiction Scale Version 2.0 (YFAS 2.0; Gearhardt, A. N., Corbin, W. R. & Brownell, K. D. Development of the Yale Food Addiction Scale Version 2.0. Psychol Addict Behav 30, 113-121 (2016)) meet the criteria for substance abuse defined in Anitescu, M., 2019. Curr Opin Anaesthesiol32(3): 427-437: 1) taking the substance in larger amounts than was intended; 2) inability to control its use; 3) taking the substance for a longer period of time than was intended, and 4) continued use despite adverse consequences.

[0007] There is a growing body of evidence that the gut microbiota is a significant player in the pathophysiology of obesity and food intake (Cryan, J. F. et al. The microbiota-gut- brain axis. Physiol Rev 99, 1877-2013 (2019)).

[0008] For instance, WO2021223692A1 discloses methods for treating metabolic diseases by way of modulating recipients' gastrointestinal tract microorganism profile such as by fecal microbiota transplantation (FMT) treatment, as well as methods for assessing a patient's risk of developing obesity and / or related metabolic diseases

[0009] Despite a large number of studies implicating the gut microbiota in obesity and behaviour, there is a limited knowledge of the relationship between the microbiome and food addiction. On the other hand, currently, no effective treatments exist to treat or prevent food addiction, or they have very limited success

[0010] In view of the state of the art, there is a need to provide methods for diagnosis and prognosis of food intake disorders, as food addiction / obesity are, as well as to provide alternative treatments for them.

[0011] DESCRIPTION OF THE INVENTION

[0012] The inventors have identified that levels of Microviridae bacteriophages and Gokushovirus (which are also bacteriophages belonging to Microviridae family) are associated with food addiction and obesity. Thus, the inventors have developed several methods based on Microviridae and Gokushovirus levels determination and comparison with reference values, which allows food addiction and / or obesity diagnosis and monitoring, as well as the evaluation of food addiction / obesity treatments.

[0013] In particular, the inventors have demonstrated that Microviridae, and particularly Gokushovirus, has positive associations with subjects diagnosed as having food addiction (based on Yale Food Addiction Scale, also named as YFAS, which is a tool to assess food addiction in individuals), as it is shown in Example 2.1. of the present description (Figure 2e-f, Figure 3a-f showing strong positive association between bacteriophages from the Microviridae family and the YFAS scores; Figure 2, showing positive association between Gokushovirus levels and YFAS score). It is also shown in this Example that patients with obesity have higher YFAS (Figure 2a). The inventors also demonstrated a consistent positive association of Microviridae and Gokushovirus levels with obesity status (Figure 4b, c).

[0014] Furthermore, as shown and explained in Examples section, particularly in the results presented in 2.2, Gokushovirus counts are associated with increased connectivity in cortical and subcortical areas, particularly from the orbitofrontal cortex, which is consistently linked to obesity and addiction. Therefore, results provided in the Example section, show that Gokushovirus are associated with food addiction, phenotypic traits related to addiction vulnerability and obesity in independent subjects cohorts.

[0015] On the other hand, the inventors have also demonstrated that anthranilic acid administration reduces food intake and body weight, and exerts a protective effect on the development of food addiction in different animal models (mouse and Drosophila melanogaster, respectively), as shown in Examples section, particularly in the results presented in 2.7 and 2.8 Thus, the present invention also relates to anthranilic acid therapeutic and non-therapeutic uses, which can be applied to subjects diagnosed with food addiction and / or obesity with the diagnostic method of the invention.

[0016] Thus, the invention solves problems of the state of the art, related to food addiction / obesity diagnosis and prognosis, representing an alternative approach. In addition, it also provides a compound to be used in the treatment and / or prevention of such food intake related disorders.

[0017] Food addiction is a very complex entity, as it encompasses components of an eating disorder, substance use disorder, obsessive-compulsive disorder, and impulsive personality traits (Vasiliu, O. Current Status of Evidence for a New Diagnosis: Food Addiction-A Literature Review. Front Psychiatry 12, 824936 (2022)). Similar to a substance-based addiction, food addiction represents an addiction-like response to food, particularly foods rich in fat and / or sugar. Individuals diagnosed with food addiction using the Yale Food Addiction Scale Version 2.0 (YFAS 2.0) meet the criteria for substance abuse defined in Anitescu, M., 2019. Curr Opin Anaesthesiol 32(3): 427-437 are characterized for showing the following behaviours: 1) taking the substance in larger amounts than was intended; 2) inability to control its use; 3) taking the substance for a longer period of time than was intended, and 4) continued use despite adverse consequences. The YFAS 2.0 food addiction criteria can be summarized in three hallmarks also used in rodent models to mimic this disorder: heightened motivation to acquire food, persistent food-seeking, and compulsive behavior (Martin-Garcia, E., Domingo-Rodriguez, L. & Maldonado, R. An Operant Conditioning Model Combined with a Chemogenetic Approach to Study the Neurobiology of Food Addiction in Mice. Bio Protoc 10, e3777 (2020)).

[0018] As used in the present invention, the term "food addiction" refers to a regular, persistent and habitual pattern of overeating characterized by craving and seeking high caloric foods, overeating in response to stimuli other than hunger, diminished control over food consumption, continued consumption despite negative consequences and diminished ability to cut down and abstain from consumption of an excess of food. Food addiction is a chronic relapsing disorder that typically follows a course of over-eating, tolerance, withdrawal, high caloric food seeking behavior and relapse (initiation of overeating after a period of abstinence). According to the YFAS 2.0, categorical scoring of food addiction diagnosis is based on the presence of at least two diagnostic criteria in the previous 12 months plus confirmation of a clinically significant impairment or distress (Gearhardt AN, Corbin WR. and Brownell KD., 2016. Psychol Addict Behav. 30(1):113). The term “food addiction”, also encompasses related conditions, such as Binge Eating Disorder (BED). “Binge Eating Disorder (BED)” is characterized by “periods of eating where the patient eats more than their normal intake, eats more rapidly, even when not hungry, feels a loss of control over eating and feels guilt over the episodes (Vaidya, V., Malik, A. (2008). Eating disorders related to obesity. Future Medicine, 5:1)

[0019] The term "obesity" in general refers to an abnormal or excessive fat accumulation that presents a risk to health, being “overweight” a level of obesity. A crude population measure of obesity in adults is the body mass index (BMI), a person's weight (in kilograms) divided by the square of his or her height (in meters). The term "obesity" is herein adopted to describe a condition characterized by, preferably, a BMI > 30 kg / m2while the term “overweight” is herein adopted to describe a condition characterized by, preferably, a BMI > 25 kg / m2but < 30 kg / m2. For adolescents, "obesity" refers to a condition characterized by two standard deviations body mass index for age and sex from the World Health Organization (WHO) growth reference for school-aged children and adolescents.

[0020] Food addiction and obesity are food intake disorders linked by common features and symptoms: (i) both obesity and food addiction involve dysregulation of the brain reward system. In both cases, there is an abnormal response to rewarding stimuli, particularly related to food. This dysfunction can lead to a heightened desire for certain foods, loss of control overeating, and compulsive overeating behaviours; (ii) an increased connectivity in cortical and subcortical areas, particularly from the orbitofrontal cortex, is consistently linked to obesity and addiction (Rolls, E. T., Feng, R., Cheng, W. & Feng, J., 2021. Soc Cogn Affect Neurosci nsab083; doi:10.1093 / SCAN / NSAB083). The downregulation of D2 receptor in addiction and in obesity is associated with decreased activity in prefrontal regions in advanced phases of these disorders. Thus, altered regulation by D2 receptor-mediated dopaminergic signalling of these frontal regions in addicted and obese subjects could underlie the enhanced motivation for food and the difficulty in self-regulating food intake. Indeed, in the Examples, it is demonstrated the association of Gokushovirus levels with increased connectivity in mentioned brain regions; (iii) subjects suffering from food addiction and / or obesity, often experience intense cravings for certain types of food, especially those high in sugar, fat, and salt. Additionally, they may exhibit withdrawal- like symptoms when attempting to reduce or eliminate these foods from their diet, such as irritability, restlessness, and anxiety; (iv) obesity and food addiction are associated with a range of health consequences, including an increased risk of cardiovascular disease, type 2 diabetes, certain types of cancer, and other obesity- related conditions.

[0021] Methods of the invention

[0022] Based on the levels of Microviridae bacteriophages, and the levels of Gokushovirus, the inventors have developed three applications, as are the diagnosis and monitoring of food addiction and / or obesity, which are food intake disorders, as well as the assessment of the response of a subject to treatments directed to said disorders.

[0023] It should be mentioned that, in theory, the methods of the invention that will be described below are applicable to any subject. The term “subject”, as it is used herein, refers to any animal, preferably a mammal, and includes, but is not limited to, domestic and farm animals, primates, and humans. In a preferred embodiment of the methods of the invention, the subject is a human being, of any sex, age, or race. In a preferred embodiment alone or in combination with other preferred embodiments, the subject is a male. In another preferred embodiment, alone or in combination with other preferred embodiments, the subject is a female.

[0024] Furthermore, the sample in which the levels of Microviridae and / or Gokushovirus are to be determined, can be any biological sample from, or isolated from, the subject. Thus, in the present invention, a "sample" means a small part or amount of something which is considered representative of the whole and taken or separated from said whole in order to be subjected to a study, an analysis, or an experimentation. In particular, in the present invention, the term "sample" encompasses samples of biological origin isolated from the subject, such as, urine, blood, saliva or digestive system. Preferably, the sample is isolated from the subject digestive system, including, but not limited to, faecal / stool sample. Thus, in a preferred embodiment of the methods of the invention, the isolated biological sample is a faecal / stool sample. Techniques for obtaining biological samples from an individual are widely known in the state of the art, and any of said techniques can be used in the practice of the present invention.

[0025] Diagnostic method of the invention

[0026] Having described the foregoing, an aspect of the present invention relates to a method for in vitro diagnosis of food addiction and / or obesity in a subject, hereinafter the “diagnostic method of the invention”, said method comprising the following steps: a) determining the levels of Microviridae bacteriophages, preferably Gokushovirus, in an isolated biological sample from the subject, and b) comparing the levels of Microviridae, preferably Gokushovirus, determined in step a) with control values, wherein elevated levels of Microviridae, preferably Gokushovirus, compared to control values, indicate that the subject suffers from food addiction and / or obesity.

[0027] The term "diagnose", as it is used herein, refers to the action of identifying a specific disorder, disease, nosological entity, syndrome, or any health-disease condition, by means of analyzing a series of clinical parameters or symptoms characteristic of said disorder and which distinguish it from other disorders with similar clinical conditions and / or from a subject not suffering from said disorder. In the present invention it relates to the identification of a disorder associated with food intake in a subject, wherein said disorder is food addiction and / or obesity, and the clinical parameter is the levels of Microviridae bacteriophages, preferably Gokushovirus.

[0028] In a first step [step a)], the diagnostic method of the invention comprises determining the levels of Microviridae and / or Gokushovirus in an isolated biological sample from a subject.

[0029] In the present invention, the term “levels” refers to the concentration, amount of Microviridae bacteriophages, preferably Gokushovirus, in a biological sample. As understood by a person skilled in the art, the determination of said levels of Microviridae bacteriophages, as well as Gokushovirus, can be carried out by means of conventional methodologies well known in the state of the art. Examples of methods for determining the levels of Microviridae and / or Gokushovirus include, but are not limited to, polymerase chain reaction (PCR), including qPCR, immunoblotting, immunoprecipitation, ELISA, hemagglutination assay, sequencing techniques (Santiago-Rodriguez TM. Identification and Quantification of DNA Viral Populations in Human Urine Using Next-Generation Sequencing Approaches. Methods Mol Biol. 2018;1838: 191-200), such as Whole-Genome Shotgun Sequencing, or CRISPR- based techniques (Huang T, Zhang R, Li J. CRISPR-Cas-based techniques for pathogen detection: Retrospect, recent advances, and future perspectives. J Adv Res. 2022 Oct 30: S2090-1232(22)00240-5)

[0030] The levels of Microviridae or Gokushovirus may be calculated or expressed as the abundance (either relative abundance or absolute abundance) of said virus in an isolated biological sample from the subject.

[0031] The “relative abundance”, when used in the context of the present invention, refers to relative amount of the viral species belonging to Microviridae, preferably Gokushovirus, more preferably Gokushovirus WZ-2015a, out of the amount of all viral species in the analyzed sample. The relative abundance can be expressed in a percentage form. For instance, the relative abundance of one particular viral species can be determined by comparing the quantity of DNA specific for this species (e.g., determined by quantitative polymerase chain reaction (qPCR) or sequencing techniques) in one given sample with the quantity of all viral DNA (e.g., determined by qPCR or sequencing techniques) in the same sample.

[0032] The “absolute abundance”, when used in the context of the present invention, refers to the amount of DNA derived from the species belonging to Microviridae, preferably Gokushovirus, more preferably Gokushovirus WZ-2015a (which can be determined by techniques known in the state of the art, such as qPCR or sequencing techniques), out of the amount of all DNA in the analyzed sample Alternatively, mentioned levels may be calculated or expressed as the total viral load of Microviridae, preferably Gokushovirus, in the sample. “Total viral load” refers to the amount of virus in a subject sample. This may be expressed as the number of viral particles per volume of the sample.

[0033] On the other hand, levels of Microviridae, preferably Gokushovirus may be calculated or expressed simply as counts / reads of DNA fragments belonging to Microviridae, preferably Gokushovirus (for instance, using sequencing techniques), from the isolated biological sample.

[0034] In a preferred embodiment of the present invention, the levels determined are expressed or calculated using an interquartile log-ratio (iqlr) transformation from the Microviridae reads / counts, preferably from the Gokushovirus reads / counts.

[0035] A way of carrying out the determination of the levels of Microviridae or Gokushovirus in any of the methods of the invention, is described in the Example section, in particular in Examples 1.3., Extraction of Faecal Genomic DNA and Whole-Genome Shotgun Sequencing, and 1.4. Microviridae and Gokushovirus WZ-2015a interquartile log ratios.

[0036] Thus, the methods of the invention comprise determining the levels of one or more types of Microviridae bacteriophages.

[0037] Moreover, since the levels of Microviridae and Gokushovirus bacteriophages can be calculated or expressed from DNA present in the sample, the sample can be treated to isolate nucleic acids prior to this step a). Techniques for isolating nucleic acids are routine laboratory practice being widely known to one skilled in the art.

[0038] Microviridae is a family of bacteriophages with a single-stranded DNA genome. Gokushoviruses constitute a subfamily within the Microviridae, being single-stranded, circular DNA bacteriophages found in metagenomic datasets from diverse ecosystems worldwide, including human gut microbiomes. Gokushoviruses belong to Superkingdom: Viruses; Clade: Monodnaviria; Kingdom: Sangervirae; Phylum: Phixviricota; Class: Malgrandaviricetes; Order: Petitvirales; Family: Microvirida;

[0039] Subfamily: Gokushovirinae; No rank: unclassified Gokushovirinae.

[0040] In a particular embodiment, the Gokushovirus comprises a nucleotide sequence having a sequence identity of, at least, 85%, 90%, 91 %, 92%, 93%, 94%, 95%, 96%, 97%, 98% or 99% with a nucleotide sequence selected from SEQ ID NO: 1 to SEQ ID NO: 87. In a more particular embodiment, the Gokushovirus comprises a nucleotide sequence having a sequence identity of 100% with a nucleotide sequence selected from SEQ ID NO: 1 to SEQ ID NO: 87.

[0041] Preferably, the Gokushovirus comprises a nucleotide sequence having a sequence identity of, at least, 85%, 90%, 91 %, 92%, 93%, 94%, 95%, 96%, 97%, 98% or 99% with SEQ ID NO: 1. Even more preferably, the Gokushovirus comprises a nucleotide sequence having a sequence identity of 100% with SEQ ID NO: 1.

[0042] SEQ ID NO: 1 relates to nucleotide sequence of the Gokushovirus WZ-2015a genome with GeneBank access number KT264835.1. The taxonomic identity of Gokushovirus WZ-2015a is NCBI:txid1758150. This virus potentially infects the species faecalibacterium prausnitzii.

[0043] SEQ ID NO: 1

[0044] ATGGCTAGAAAGATTCCTTCTGTGATGAATCATACTTTTTCTGAGGCTCCTGTA GTTAATGTGCGTCGTTCTGTTTTTGATCGTTCTTGTGGTCATAAGACGACCTTTC CTGCTGGTGTTTTAGTTCCTGTTTATGTGGATGAGATTTTGCCTGGAGATACTTT TTCGATGGATGTTTCTTTTTTAGCAAGGTTGACGACGCCTATCCATCCTATTATG GATAATATGAATATTACACTTCATGCTTTCTTTGTTCCGAATCGTTTAGTTTGGT CTGATTGGGAGAATTTTATTTGTAATACGAAGGATTATGAAACGGATACTGCTG GTGATCAGATTACTATTCCGACAATTTCTTTGAATTCTAAGGAAGGTGTTTATTT TAATGCAGGATCTTTATTTGATTATATGGGAATTCCTCCTTTGAATGGAACAGGA ACTACTGGAAGTGGTGTTAATTATTATGTAGATGTTAATGCTTTACCTGTTGCTG CTTATGGTTTGATTTGGTCTGAGTTTTATAGAGATGAGAATTTATTTGCTCCTTT GAATGTGAAACAAGTTTTATCTCAAGGAACTTTTTCTTGGCAGTTGGAAGATGG AACCGTAATGAATACTTATGGAGGTTTGCTTCCTCGTTGTAAACGTCATGATTAT TTTACATCTTCTTTGCCTTGGCCACAGAAAGGTCCAGCTTGTGTTTTACCATTG GCTAGTTCAGCTCCTGTTGTTGGGTTGCCTGGTCCTACAGGAGTTCATGCTGAT

[0045] ACTTCTCAGCTTGGTGGTACAATTAATCAGTTGAGAGAAGCTTTGCAGATTCAG

[0046] AGTCTTTTAGAGATTGATGCGAGAGGAGGAACTCGATATATTGAGTTGATATTA

[0047] TCACATTTTGGTGTAACTTCTCCGGATGCGAGATTACAGCGTCCTGAATATTTA

[0048] GGTGGAGCTAGAGTTCCTGTGAATATTTCTCCTATTCCTCAGACTTCTGCGACT

[0049] CAACAGGACACAAATCCTGGAGATACTTTTGTTCCTACGCCTCAAGGTAATTTA

[0050] GCTGGTATAGGGACTTCTTCGCATGCTGGAAGGCTTTTTTCTAAGTCTTTTGTT

[0051] GAGCATGGATATATTATGGTTTTAGCTTCTGTGGATGCTGATTTGAATTATCAGC

[0052] AGGGTTTAGAGAGGATGTGGTCTCGTAAGACTCGTTATGATTTTTATTGGCCAT

[0053] CTTTGGCACATCTTGGTGAGCAGCCTGTGTTGAATAAAGAGATTTATCTTCAAG

[0054] GTTATTCTTTGAGTTCTACTGGAGAGAAGGTTCCTCTTGGATCTGTTGGTCAGA

[0055] CGGATAATGATGTTTTTGGTTATCAAGAGCGTTATGCGGAGTATCGTTATAAGC

[0056] CTTCTTTGATTACAGGAGCTTTTAGATCAACACCTTACGGTTCTAATTATACTTC

[0057] TTCTGGAGGTAATGCTTTTGGAAATACTTCTTCTTTAGATTTATGGCATTTAGCT

[0058] CAGGCTTTTACTAATTTGCCTTCTTTGTCTAAATCTTTTATTGAGGTAAATATTCC

[0059] TTTGGATAGAGTTTTAGCAGTATCTACTTCTATAGCTCCTCCATTTTTATTTGATT

[0060] CATTTTTTAGAGTAAAGGCAGTGAGACCGATGCCGACTTATGCAATTCCTGGAT

[0061] TAGGAGCACGTTTATGACAATATTTCATCATGTATTACATTTCTTATTTTTAGCAC

[0062] TTTCTGGCGCCGCTACTTTAATTCCTTTGTTTGTGAGGAATAATATGAAGCTTAA

[0063] TCAGTTAGAATCTGATGTTGAGGCTTTAAAGGATGCTGTTGAGCATTTTGATTCT

[0064] GCTTTAGGTAAGTAAATGGATTGGAGTAGTATAGCTGCATCTGCTGCTGCAGGT

[0065] TTATTTCCTGCTATAGGTCAGTTGTTTGCTAATCATTCTAATAAGAAGCTTTATTA

[0066] TGATCAGAAAGCTTTTTTAGAGAAGATGTCTGGTTCTGCAGTTCAGCGTGCTGT

[0067] TGCGGATATGAAGCAGGCTGGTATTAATCCTATTTTAGCTTCAGGTAGTCATGC

[0068] TTCATCTCCGAGTCCTGGATTGCCTGCTATTGAGAATCCTCTTAAGGATGTTGT

[0069] TTCTAATGGTCTTTCTATAGCTTCTGCTAAGGCTAATATTGATTTGGTTAGGCAG

[0070] CAGCGGGATTTGTTGAAAGCTCAGGCTGAGAAGACTGATGTGGAGAAACGTCT

[0071] TGTTCAAGAGAAAATAGATAATCCTCTAAGGAGTGTTCCCTACTGGGGATTAGC

[0072] GTATTCCGCTTTTAATTATGCGAAAAATAATCCTCCTTCTTCTGATTTTTTGAATG

[0073] AGCCTCTTCTTTCTCGTTTTTGGTCTTATCTTGCGCGACGTTATAGTGAATGGTC

[0074] TGCTAAGAATGAGCATTATTATAGTCATCGTGTTCAATCTTATGGAAGGAGGAA

[0075] AAGATGAAAAAGGTAAAAGTATATTCTGCATTTGATAGGCCTGATCATGAAGGT

[0076] TATGATTTTTTTGATGAGATGTCTCCAGTGCAGCAGCATTTTCGTGATCAGTCTG

[0077] ATATTAATTTTATTGTTAATTATTGGTCACGTACTGGAAATTTTAGTTCAGTTAAT CCTATTCCTGCTAGATATGTTGATTGTACTGCTGTACAAGATTACCAGAATGCTT

[0078] TGGATACTTTGTGTAAAGCGGATGAGATTTTGGAAGCGATGACTCCTCAACAAA

[0079] GATTGAGATTTCATAATGATCCTATTGCATTTATTGAATATTTTTCAGATCCTGCT

[0080] AATATTGATGAGTTGAAGGCTTTAGATAGAGGTTTGCAGCTTCATCCTAAGGAG

[0081] CGTGCTCCCCATAGCGGAGCGACGCAGGATGATGCGGTAAGTTCTTCCTTAAA

[0082] CGCCGTAGAGCCTCATCAGGTTGCAGATAAACACTTGGCATAGATAACCGTAC

[0083] TTGTTGTTATCTATGCCAAGTGACACCAATTAGTAATTTTGAGGTTTTTATGGCT

[0084] AGAAGAAGAAGAATGTCGCGAAGAGTGAGTAAGAGAGTTTTTAAGAAACATTAT

[0085] AAATGGAATCGTAGAAATTTTATTACTAAAGTTCCTCGTGGAGGAATTAGATTTT

[0086] AATTTGTTTTTTGATAAAAAAAAGAGTTAGTGTTTTTTGCTTCCCTAACTCTTTTA

[0087] ATCACAGGATTTTTTTTATGTGCATCAATCCCAGTTTAATTTACTTTATCAGAGAT

[0088] TACGATTCTCGTGAAGTGCTTTTTAAAGAAGTTATGAATTTGAAGCATGAGGCTT

[0089] CTCTTAGTTCTCCTTTTGTTTTTCATCCTACTTTTGATACTCCGTATAAGAAGATT

[0090] CCTGGAGTTGGAGAGTATTCATTGCTTCCTCCTGGTAAGCATTTTAACAATTCT

[0091] CAGTTGGAGTATTTTAAGATTTCTTGTGGAAAGTGTAGGTGGTGTTTGATTCGT

[0092] CGCTCTGCGCATTGGGCTGTTCGTTGTCTTTATGAGACTTATATGCATGAAAAG

[0093] TCCTGTTTTTTGACTTTGACTTATGACGAGGTTCATTTGCCTTCTGATCGTCAAT

[0094] TAGTTATCAGACATGTTCAGCTATTTTTGAAAAGATTACGTAAATATCTTTCTCA

[0095] GAGAGCTTTGCCTCGTATACGCTTTTTTGGTTGTGGAGAGTATGGAGCTAAGAA

[0096] TTATCGTCCCCATTATCATTTGATGATCTTTGGTTATGATTTTGGAATTTCTTCTA

[0097] TTCCCTTTAATGTGCGTTCTGTTAAGAATTATCCTTCTTGGGTAGATAATCCTAT

[0098] TGTTAGTAAGCTTTGGCCTTATGGTTTTCATCTTATTGGAGAGGCTACGTATCA

[0099] GTCTGCACGTTATGTGGCTGCTTATATTTTTGGTAAGCTTAAGAATAAGGCTGA

[0100] GAAGAAGGATTTTGTTGTTAAAGAGTATGTGCATAGTTCTCGTCGTCCTGGTGT

[0101] TGGAATTCCGTGGTTAGAAAAAAATTGGCGTGATGTTTATCCCTTGGATTGTGT

[0102] GACCTTTCCGTCTGATAATCATACTGTTAAACAGCTTCCACCGCCTCGTGCTTT

[0103] TGATAAGTGGCTTTCTCTAAATCAGCCTGAGGTTTTTAAGAAAGTTATTGATAGA

[0104] AGGATTGAGTATTATGAGTCTTGTGAGGATTTTGTTTATGATTATGAGGATGTTT

[0105] TGAAGTCTACTAATAGAGATATTAATTTATCTGCTTTGCTTTCTTCTTGTTCTCGT

[0106] TCTTTGTGATTTTAATTTTTTTTTTGACATTTAACCCCCCCCCCCATTTATTTTTCT

[0107] >

[0108] TTTTTTAGAGGTTGTGGATGTTGTTAGTATATTCTTTATTTGATTCTGCTGTTGGA

[0109] GAGTTTGGTACTCCTGTGTGTTATGCTTCCGAGTCTTTAGCTAGGAGGTCTTTT

[0110] TCTTTGTTAGTGCTTGAGGATTCACCTCATTATGCTTCTTATAGGAATTATTCTT

[0111] CAGATTTTTCTGTCTTTTATTTAGGAACGTATTGTTCTGAGTCAGGAAGTTTTGA TTTATTACCTTCTCCTAAGGCAGTTTTTCGTTTAGATGAGTTGCTTAATGATAGG AGAGATTAA

[0112] Thus, in a preferred embodiment of any of the methods of the invention, alone or in combination with other preferred embodiments, the Gokushovirus is Gokushovirus WZ-2015a.

[0113] Once the Microviridae or Gokushovirus levels have been determined in step (a), the diagnostic method of the invention comprises comparing said levels with control values [step b)].

[0114] In the diagnostic method of the invention, the term "control values" refers to the Microviridae levels, or Gokushovirus levels, in a subject non-having food addiction and / or non-having obesity; or to the mean levels of Microviridae, or Gokushovirus, in a cohort of subjects non-having food addiction and / or non-having obesity. A subject non-having food addiction, whose levels may be used as control values in the context of the present invention, may be a subject who, according to the YFAS 2.0, does not show a categorical scoring of food addiction diagnosis based on the presence of at least two diagnostic criteria in the previous 12 months plus confirmation of a clinically significant impairment or distress (Gearhardt AN, Corbin WR. and Brownell KD., 2016. Psychol Addict Behav. 30(1):113). As described previously, a subject non-having obesity, whose levels may be used as control values, may be a subject having a BMI below 30 kg / m2, preferably BMI below 25 kg / m2.

[0115] On the other hand, methods for determining the Microviridae or Gokushovirus levels have been described in the preceding paragraphs.

[0116] Finally, by comparing the levels of Microviridae, or Gokushovirus levels, with the control value, it can be determined whether or not an individual has food addiction and / or obesity. Thus, elevated levels of Microviridae, preferablyGo us / iov / s, relative to the control value indicate that the subject has food addiction and / or obesity.

[0117] In the present invention, the term “elevated”, interchangeably used with “higher” or “increased”, referring to the term “levels”, compared to control values, is understood to mean that the levels are greater than the control values in at least 1.2 times, 1.25 times, 1.5 times, 2 times, 3 times, 4 times, 5 times, 6 times, 7 times, 8 times, 9 times, 10 times, 20 times, 30 times, 40 times, 50 times, 60 times, 70 times, 80 times, 90 times, 100 times or even more (with respect to the control value).

[0118] Alternative to the above-disclosed method, the present invention also relates to the use of the levels of Microviridae, preferably Gokushovirus, for the in vitro diagnosis of food addiction and / or obesity.

[0119] Therefore, in another aspect, the present invention relates to the use of the levels of Microviridae, preferably Gokushovirus, for the in vitro diagnosis of food addiction and / or obesity in a subject.

[0120] All the definitions and particular embodiments disclosed for the diagnosis method of the invention are applicable to the present aspect of the invention.

[0121] Monitoring method of the invention

[0122] Another method of the present invention, also based on the Microviridae or Gokushovirus levels, is a method for monitoring the evolution of food addiction and / or obesity. Thus, another aspect of the present invention relates to a method for in vitro monitoring the evolution of food addiction and / or obesity in a subject, hereinafter the “monitoring method of the invention”, comprising the following steps: a) determining in, at least, two different moments in time, the levels of Microviridae bacteriophages, preferably Gokushovirus, in an isolated biological sample from the subject, and b) comparing the levels of Microviridae, preferably Gokushovirus, determined in said moments with one another, wherein an increase in Microviridae levels, preferably Gokushovirus levels, indicates that the food addiction and / or obesity evolves negatively in the subject. Terms used to define the present aspect of the invention have been explained in preceding aspects, applying, as well as their preferred embodiments, to the present aspect of the invention.

[0123] In the present invention, the expression “monitoring the evolution of food addiction and / or obesity”, refers to predicting and / or evaluating the course of food addiction / obesity in a subject. Monitoring the evolution of food addiction / obesity can ease clinical decisions, such as customizing the treatment strategy to be followed for said subject, defining groups of patients according to its degree of progression, or improving the design and analysis of clinical studies and trials by stratifying patients. In the present invention, the evolution or course of the disease can be monitored by determining and comparing the levels of Microviridae and / or Gokushovirus. at different moments in time.

[0124] Step (a) of the monitoring method of the invention comprises determining the levels of Microviridae and / or Gokushovirus in an isolated biological sample from the subject in, at least, two different moments in time. The term “levels”, and examples of methods for determining the levels of Microviridae or Gokushovirus have been defined and explained in the diagnostic method of the invention, applying equally to the present aspect of the invention.

[0125] In the present invention, the expression "different moments in time" is understood to mean moments sufficiently separated in time, in other words, the minimum time lapse between two moments, which allows concluding how food addiction and / or obesity evolves over time. Preferably, in the present invention, the time lapse between the two moments considered sufficient to conclude how the disorder / s evolves is from, at least, 2 hours, to, at least, 1 or 2 years.

[0126] Once the levels of Microviridae, preferably Gokushovirus, have been determined, they are compared with each other to see whether the level has decreased or increased over time (step b) of the monitoring method of the invention). Thus, an increase in the levels of Microviridae, preferably Gokushovirus, over time, indicates that the subject’s food addiction / obesity is progressing negatively. As used herein, the phrase "an increase in the levels of Microviridae, or Gokushovirus, over time" means that the levels of Microviridae or Gokushovirus determined are elevated relative to the levels of the same determined at an earlier time. Similar to the diagnostic method of the invention, an increase in the levels is understood to mean that the levels are greater in at least 1.2 times, 1 .25 times, 1.5 times, 2 times, 3 times, 4 times, 5 times, 6 times, 7 times, 8 times, 9 times, 10 times, 20 times, 30 times, 40 times, 50 times, 60 times, 70 times, 80 times, 90 times, 100 times or even more the levels determined in the same determined at an earlier time.

[0127] The term “evolves negatively" refers to the fact that the subject exhibits a worsening of its health condition related to food addiction and / or obesity. For instance, but without limitation to, this worsening may be associated with increased connectivity in cortical and subcortical areas, consistently linked to obesity and addiction or phenotypic traits related to addiction vulnerability and obesity, such as impulsivity, appetitive associative learning and cognitive flexibility, as explained with more detail in the Example section. In the present monitoring method of the invention, an increase in the levels of Microviridae, preferably Gokushovirus, over time indicates that the food addiction and / or obesity evolves negatively in the subject.

[0128] Alternative to the above-disclosed method, the present invention also relates to the use of the levels of Microviridae, preferably Gokushovirus, for the in vitro monitoring of the evolution of food addiction and / or obesity in a subject.

[0129] Therefore, in another aspect, the present invention relates to the use of the levels of Microviridae, preferably Gokushovirus, for the in vitro monitoring of the evolution of food addiction and / or obesity in a subject.

[0130] All the definitions and particular embodiments disclosed for the monitoring method of the invention are applicable to the present aspect of the invention.

[0131] Treatment evaluation method of the invention Another application of the invention, also based on the same concept of determining and comparing levels of Microviridae or Gokushovirus, relates to a method for evaluating the response to food addiction / obesity treatments.

[0132] Thus, in another aspect, the present invention relates to method for evaluating in vitro the response of a subject to a treatment for food addiction and / or obesity, hereinafter the “treatment evaluation method of the invention”, comprising the following steps: a) determining, before and after treatment, the levels of Microviridae bacteriophages, preferably Gokushovirus, in an isolated biological sample from the subject, and b) comparing the levels of Microviridae, preferably Gokushovirus, determined before the treatment with those obtained after the treatment, wherein lower levels of Microviridae, preferably Gokushovirus, after treatment than before treatment indicate that the treatment is effective and / or that the subject is responding positively to treatment.

[0133] Terms used to define the present aspect of the invention have been explained in previous inventive aspects, applying, as well as their preferred embodiments, to the present aspect of the invention.

[0134] The expression “evaluating the response to a treatment" refers to analyzing the capacity of a treatment, particularly a treatment for food addiction and / or obesity, to cause a desired result or effect in a subject, and / or verifying whether the treatment causes a change in the food intake disorder, particularly food addiction / obesity in the subject.

[0135] The treatment evaluation method of the invention comprises a first step of determining the levels of Microviridae and / or Gokushovirus in a biological sample isolated from the subject before and after the treatment applied to the subject and, once these levels are determined, they are compared to each other in a second step. Thus, if the levels of Gokushovirus after treatment are lower than before treatment, then the treatment is effective and / or the subject responds positively to the treatment. The term “levels”, and examples of methods for determining the levels of Microviridae or Gokushovirus have been defined and explained in the diagnostic method of the invention, applying equally to the present aspect of the invention.

[0136] In the treatment evaluation method of the invention, a decrease in the levels of Microviridae, or Gokushovirus, being compared, indicates that treatment is effective and / or the subject responds positively to treatment.

[0137] In the present invention, a subject which “responds positively to treatment” is understood to mean that the treatment has had an effect of improving the subject’s food intake disorders, in particular, food addiction and / or obesity.

[0138] In the present aspect of the invention, a decrease in the levels of Microviridae, or Gokushovirus’, or lower levels of Microviridae, or Gokushovirus, means that the levels of Microviridae, or Gokushovirus, determined are less in at least 1.2 times, 1.25 times, 1.5 times, 2 times, 3 times, 4 times, 5 times, 6 times, 7 times, 8 times, 9 times, 10 times, 20 times, 30 times, 40 times, 50 times, 60 times, 70 times, 80 times, 90 times, 100 times or even more, after treatment than before treatment. I

[0139] Health professionals treat food intake disorders, and in particular, food addiction or obesity by taking measures to improve the causes or symptoms of the disorder in a patient. Treatment may consist of pharmacological or non-drug therapies (including the application of one or more of them). Drug-based therapies may include: selecting and administering one or more drugs to the patient, adjusting the dose of a drug, adjusting the dosing schedule of a drug, and adjusting the duration of drug therapy. Healthcare professionals may select drugs, adjust the dose, the dosing schedule of a drug, and the duration of therapy based on the nature of the drug, the nature of the patient's symptoms, the patient's response to previous treatment, and the patient's response to the drug.

[0140] A person skilled in the art can identify appropriate treatments for food addiction or obesity based on the medical literature. Examples of treatments for obesity include, without limitation to dietary changes, regular exercise, nutritional counselling, medication (in some cases), bariatric surgery (in extreme cases), psychological therapy, medical follow-up and a comprehensive approach combining several methods.

[0141] Examples of treatments for food addition include, without limitation to therapies such as cognitive behavioural therapy, interpersonal therapy and acceptance and commitment therapy; medication may also be used in some cases; group support; nutritional counselling; and mindfulness practices may be helpful. The choice of treatment depends on the severity of the disorder and the individual's needs, and often requires a combination of these strategies

[0142] In a preferred embodiment of the treatment evaluation method of the invention, alone or in combination with other preferred embodiments of the invention, the subject is suffering from food addiction and / or obesity.

[0143] Alternative to the above-disclosed method, the present invention also relates to the use of the levels of Microviridae, preferably Gokushovirus, for the in vitro evaluation of the response of a subject to a treatment for food addiction and / or obesity.

[0144] Therefore, in another aspect, the present invention relates to the use of the levels of Microviridae, preferably Gokushovirus, for the in vitro evaluation of the response of a subject to a food intake disorder treatment, wherein said disorder is food addiction and / or obesity.

[0145] All the definitions and particular embodiments disclosed for the treatment evaluation method of the invention are applicable the present aspect of the invention.

[0146] Kit of the invention

[0147] The uses and methods of the invention are put into practice based on determining the levels of Microviridae, preferably Gokushovirus, in isolated biological samples from a subject. The means used for this may be part of a kit.

[0148] Thus, another aspect of the present invention relates to a kit, hereinafter the “kit of the invention”, which comprises means for the in vitro determination of the levels of Microviridae bacteriophages, preferably Gokushovirus,. Examples of means which may be comprised within the kit of the invention include, without limiting to, primers and / or probes recognizing specifically Microviridae and / or Gokushovirus’, buffers, nucleotides, etc., and any other mean useful to identify Microviridae and / or Gokushovirus by PCR, qPCR, massive sequencing, Sanger sequencing, etc.

[0149] Furthermore, as understood by a person skilled in the art, the kit may comprise other components useful in putting the present invention into practice, such as buffering solutions, delivery vehicles, material supports, positive and / or negative control components, etc. In addition to the mentioned components, the kits may also include instructions for practicing the object of the invention. These instructions may be present in the mentioned kits in a variety of forms, one or more of which may be present in the kit. One way in which these instructions may be present is as information printed on a suitable medium or substrate, for example, a sheet or sheets of paper on which the information is printed, in the kit packaging, in a package insert, etc. Another medium would be a computer-readable medium, for example, a CD, a USB, etc., in which the information has been recorded. Another medium that may be present is a website address that can be used over the Internet to access information at a remote site. Any suitable means may be present in the kits.

[0150] The kit of the invention is useful in the diagnosis of food addiction and / or obesity, the monitoring of food addiction and / or obesity, and the in vitro evaluation of the response to treatments for food addiction and / or obesity, and can be used in any of the methods of the invention that have been described above.

[0151] Thus, in another aspect, the invention relates to the use of the kit of the invention in any of the methods of the invention.

[0152] In another aspect, the invention relates to the use of the kit of the invention for the in vitro food addiction and / or obesity diagnosis in a subject.

[0153] In another aspect, the invention relates to the use of the kit of the invention for the in vitro monitoring of the evolution of food addiction and / or obesity in a subject. In another aspect, the invention relates to the in vitro use of the kit of the invention to evaluate the response of a subject to a food addiction and / or obesity treatment.

[0154] The terms used to define the kit and the uses of the kit of the invention have been explained for preceding aspects of the invention, applying, as well as their preferred embodiments, to the different uses of the kit of the invention. Likewise, the terms "subject" has been defined in other aspects of the present invention and is applicable to the kit of the invention and uses thereof. Likewise, the preferred embodiments of said terms are also applicable to the kit of the invention and the uses thereof.

[0155] Therapeutic uses of the invention

[0156] In addition to the methods of the invention (diagnosis, monitoring and evaluation of the response to treatments), directed to food addiction and / or obesity, the present invention also relates to therapeutic and non-therapeutic use of compositions comprising anthranilic acid, whose administration the inventors have shown to reduce food consumption and body weight and demonstrated to have a protective effect on the development of food addiction in animal models.

[0157] Thus, other aspect of the invention relates to anthranilic acid for use in the treatment, and / or prevention, of food addiction and / or obesity in a subject.

[0158] Another aspect relates to anthranilic acid for use in the treatment, and / or prevention, of food addiction and / or obesity in a subject, wherein the subject has been diagnosed to suffer from food addiction and / or obesity with the diagnostic method of the invention.

[0159] Anthranilic acid (also known as 2-aminobenzoic acid; molecular formula: C7H7NO2; CAS number: 118-92-3; in the present invention also mentioned as AA) is an aminobenzoic acid that is benzoic acid having a single amino substituent located at position 2. It is a metabolite produced in L-tryptophan-kynurenine pathway in the central nervous system. Anthranilic acid chemical structure is the following:

[0160] In the present invention, anthranilic acid also relates to its salts and solvates.

[0161] Furthermore, anthranilic acid can be comprised in, or it can be administered, as an active compound in the form of a composition. Therefore, another aspect of the invention relates to a composition comprising anthranilic acid, hereinafter the “composition of the invention”, for use in the treatment, and / or prevention, of food addiction and / or obesity disorders in a subject

[0162] As used in the present aspects of the invention, the term "to treat" or "treatment" comprises: inhibiting the disease or disorder, i.e. , stopping its development; relieving the disease or disorder, i.e., causing regression of the disease or disorder; and / or stabilizing the disease or disorder in a subject. In the present invention, the disease or disorder is food addiction and / or obesity.

[0163] As used in the present aspects of the invention, the term “prevention” means the avoidance of occurrence of the disorder, disease, or pathological condition in a subject. Particularly, the subject may have predisposition for the pathological condition, but has not yet been diagnosed. In the present invention, the disease or disorder is food addition and / or obesity.

[0164] The terms “food addiction” and “obesity” have been defined and explained previously in other aspects of the present invention, applying equally to the present aspect of the invention.

[0165] For the purposes of therapeutic uses of the invention, obesity encompasses related or associated diseases / disorders, which refer to diseases / disorders caused by obesity or overweight, including, without limitation to, diabetes, metabolic syndrome, hypertension, hyperglycaemia, inflammation, type-2 diabetes, cardiovascular disease, hypercholesterolemia, hormonal disorders, infertility, etc. and those disease or disorders which obesity or overweight is a risk factor of suffering them.

[0166] For the purposes of therapeutic uses of the invention, food addiction encompasses related or associated diseases / disorders, which refer to diseases / disorders caused by food addiction, including, without limitation to, Binge Eating Disorder (BED).

[0167] The term “subject”, as it is used herein, refers to any animal, preferably a mammal, and includes, but is not limited to, domestic and farm animals, primates, and humans. In a preferred embodiment of the methods of the invention, the subject is a human being, of any sex, age, or race. In a more preferred embodiment, the subject suffers from food addiction and / or obesity. In an even more preferred embodiment, the subject suffering from food addiction and / or obesity has been diagnosed with the diagnostic method of the invention.

[0168] The composition of the invention may be formulated for pharmaceutical administration, i.e. , forming part of pharmaceutical products to be administered to the subject by any means of administration. Thus, in a preferred embodiment, the composition of the invention is a pharmaceutical composition (“pharmaceutical composition of the invention”).

[0169] In a preferred embodiment, the pharmaceutical composition further comprises a pharmaceutically acceptable carrier and / or excipient.

[0170] The term "excipient" refers to a substance that helps the absorption of any components or compounds of the composition of the invention, namely, the anthranilic acid, or stabilizes the components or compounds and / or assists the preparation of the pharmaceutical composition in the sense of giving it consistency or flavours to make it more pleasant. Thus, the excipients may have the function, by way of example but not limited thereto, of binding the components (for example, starches, sugars or cellulose), sweetening, colouring, protecting the active ingredient (for example, to insulate it from air and / or moisture), filling a pill, capsule or any other presentation or a disintegrating function to facilitate dissolution of the components, without excluding other excipients not listed in this paragraph. Therefore, the term "excipient" is defined as that material that included in the galenic forms, is added to the active ingredients or their associations to enable their preparation and stability, modify their organoleptic properties or determine the physical and chemical properties of the pharmaceutical composition and its bioavailability. The "pharmaceutically acceptable" excipient must allow the activity of components or compounds of the pharmaceutical composition, that is, be compatible with the effect exerted by anthranilic acid.

[0171] The "vehicle" or "carrier" is preferably an inert substance. Carrier functions are to facilitate the incorporation of other components or compounds, allow better dosage and administration and / or give consistency and form to the pharmaceutical composition. Therefore, the carrier is a substance used in the drug to dilute any of the components or compounds of the pharmaceutical composition of the present invention to a given volume or weight; or that even without diluting these components or compounds, it is able to allow better dosage and administration and / or give consistency and form to the drug. When the presentation is liquid, the pharmaceutically acceptable carrier is the diluent. The carrier can be natural or unnatural. Examples of pharmaceutically acceptable carriers include, without being limited thereto, water, salt solutions, alcohol, vegetable oils, polyethylene glycols, gelatine, lactose, starch, amylose, magnesium stearate, talc, surfactants, silicic acid, viscous paraffin, perfume oil, monoglycerides and diglycerides of fatty acids, fatty acid esters petroetrals, hydroxymethylcellulose, polyvinylpyrrolidone and the like.

[0172] Furthermore, the excipient and the carrier must be pharmacologically acceptable, i.e., the excipient and the carrier are permitted and evaluated so as not to cause damage to the subject to whom it is administered.

[0173] As understood by the person skilled in the art, anthranilic acid can be present in an effective amount, preferably in a therapeutically effective amount, in order to exert its effects, such as reducing body weight, protecting against the development and / or progression of food addiction, or reducing food intake / consumption in a subject. Examples of this effects and assays to assess them are shown in Examples section, in particular in the results presented in 2.7 and 2.8. The effective amount may vary depending on, for example, the age, body weight, general health, sex and diet of the subject, as well as on the mode and time of administration, the excretion rate or any potential co-treatment with other drugs.

[0174] In a preferred embodiment, alone or in combination with other preferred embodiments, the anthranilic acid concentration is between 0,01 to 10 mg / kg (including the end values of the range).

[0175] The compositions of the present invention can be formulated for administration to an animal, and more preferably to a mammal, including a human, in a variety of forms known in the prior art. Thus, they may be, without limitation, in aqueous or nonaqueous solutions, in emulsions or in suspensions. Examples of non-aqueous solutions are, for example, but without limitation, propylene glycol, polyethylene glycol, vegetable oils, such as olive oil, or injectable organic esters, such as ethyl oleate. Examples of aqueous solutions are, for example, but not limited to, water, alcoholic solutions in water, or saline media. Aqueous solutions may be buffered or unbuffered and may have additional active or inactive components. Additional components include salts to modulate ionic strength, preservatives including, but not limited to, antimicrobial agents, antioxidants, chelators, or the like, or nutrients, including glucose, dextrose, vitamins, and minerals. Alternatively, the compositions can be prepared for administration in solid form. The compositions may be combined with various vehicles or inert excipients, including but not limited to binders, such as microcrystalline cellulose, gum tragacanth, or gelatin; excipients, such as starch or lactose; dispersing agents, such as alginic acid or corn starch; lubricants, such as magnesium stearate, glidants such as colloidal silicon dioxide; sweetening agents, such as sucrose or saccharin; or flavouring agents, such as peppermint or methyl salicylate.

[0176] Additionally, the composition of the invention may comprise an adjuvant. "Adjuvant" means any substance which enhances the effectiveness of the pharmaceutical composition of the invention. Examples of adjuvants include, but are not limited to, adjuvants consisting of aluminium (alum) salts, such as aluminium hydroxide, aluminium phosphate or aluminium sulphate, oil-in-water or water-in-oil emulsion formulations such as Complete Freund's Adjuvant (ACF) as well as Incomplete Freund's Adjuvant (AIF), mineral gels, gels, block copolymers, Avridine™, SEAM62, adjuvants consisting of bacterial cell wall components such as adjuvants including liposaccharides (e.g. lipid A or Monophosphoryl Lipid A (MLA), trehalose dimycolate (TDM), and cell wall skeleton components (CWS), heat shock proteins or their derivatives, adjuvants derived from ADPribosylated bacterial toxins, including diphtheria toxin (DT), pertussis toxin (PT), cholera toxin (CT), the heat labile toxins of E.coli heat labile toxins (LT1 and LT2), Endotoxin A and Pseudomonas exotoxin, B. cereus exoenzyme B, B. sphaerieus toxin, C. botulinum toxins C2 and C3, C. limosum exoenzyme as well as the toxins of C. perfringens, C. spiriforma and C. difficile, S. aureus, EDIM and toxin mutants such as CRM-197, non-toxic mutant diphtheria toxin; saponins such as ISCOMs (immunostimulatory complexes), chemokines and cytokines such as interleukins (IL-I IL-2, IL-4, IL-5, IL-6, IL-7, IL-8, IL-12, etc), interferons (such as interferon gamma), macrophage colony-stimulating factor (M- CSF), tumour necrosis factor (TNF), defensins 102, RANTES, MIPI-alpha, and MEP- 2, muramyl peptides such as N-acetyl-muramyl-L-threonyl-D-isoglutamine (thr-MDP), N-acetyl normuramyl-L-alanyl-D-isoglutamine (nor-MDP), N-acetylmuramyl-L- alanyl- Disoglutaminyl- L-alanine-2-( 1 '-2'-dipalmitoyl-s- n-glycero-3 huydroxyphosphoryloxy)ethylamine (MTP-PE) etc; adjuvants derived from the CpG family of molecules, synthetic CpG dinucleotides and oligonucleotides comprising CpG motifs, lysosum exoenzyme from C. Limosum and synthetic adjuvants such as PCPP, cholera toxin, Salmonella toxin, alum and similar, aluminium hydroxide, N- acetyl-muramyl-L-threonyl-D-isoglutamine (thr-MDP), N-acetyl-nor-muramyl-L- alanyl-D-isoglutamine, MTP-PE and RIBI, which contains three components extracted from bacteria, monophosphoryl lipid A, trehalose dimycolate and cell wall skeleton (MPL+TDM+CWS) in a 12% / Tween 80 squalene emulsion.

[0177] Anthranilic acid, or the composition of the invention, preferably the pharmaceutical composition of the invention, may be administered by any appropriate administration route. To this end, said composition will be formulated in the suitable pharmaceutical form for the selected administration route.

[0178] Suitable routes of administration may, for example, include oral, depot, transdermal, rectal, transmucosal, or intestinal administration; parenteral delivery, including intramuscular, subcutaneous, intravenous, intramedullary injections, as well as intrathecal, direct intraventricular, intraperitoneal, intranasal, or intraocular injections.

[0179] Nevertheless, a preferred route of administration is the oral route.

[0180] Another aspect of the invention refers to the use of anthranilic acid in the manufacture of a medicament (or a pharmaceutical composition) for the treatment and / or or prevention, of food addiction and / or obesity in a subject. In a preferred embodiment, the subject has been diagnosed to suffer from food addiction and / or obesity with the diagnostic method of the invention.

[0181] Techniques and procedures for the production of medicaments are widely known in the state of the art.

[0182] The terms used to define the present aspect of the invention have been previously explained, and they and their preferred embodiments are applicable to the present aspect of the invention.

[0183] In another aspect, the invention relates to a method for the treatment, and / or prevention, of food addiction and / or obesity in a subject, comprising the administration of anthranilic acid.

[0184] In a preferred embodiment, the method further comprises a step, previous to the anthranilic acid administration, of identifying a subject suffering food addiction and / or obesity, preferably, by the diagnostic method of the invention.

[0185] The terms used to define the present aspect of the invention have been previously explained, and they and their preferred embodiments are applicable to the present aspect of the invention.

[0186] Non-therapeutic use of the invention

[0187] The present invention also encompasses non-therapeutic uses of anthranilic acid, or the composition of the invention (including its preferred embodiments, alone or in combination each other). Thus, in another aspect, the present invention relates to the non-therapeutic use of anthranilic acid, or the composition of the invention in body fat reduction of a subject.

[0188] Anthranilic acid, or the composition of the invention can also be used for non- therapeutic fat reduction in non-obese subjects, i.e. “normal-weight” subjects having a BMI < 25 kg / m2or "overweight" subjects as defined above and without any obesity- associated health implications. Such use is exclusively for aesthetic or cosmetic reasons (cosmetic use) and not based on a medical indication. Also, body fat reduction and / or maintenance of body weight in non-obese subjects might involve body weight reduction and / or maintenance. The cosmetic product will uniquely have a cosmetic effect in the subject who uses it related to body fat reduction. The term “cosmetic effect” is explained below. Analogously, the present invention relates to a non-therapeutic method for body fat reduction comprising administering anthranilic acid, or the composition of the invention, to the subject.

[0189] The term “cosmetic effect” as used herein refers to the desired advantageous impact of anthranilic acid, or the composition of the invention, with regard to appearance, which is associated with loss or maintenance of body fat, and preferably an enhancement of body shape and definition. Non-therapeutic, cosmetic treatment with anthranilic acid, or the composition of the invention, may also induce loss of weight and / or serve for weight control in order to prevent a (non-pathological) body fat accumulation and / or gain of weight.

[0190] In a preferred embodiment, alone or in combination with other preferred embodiments, the subject has been diagnosed to suffer from food addiction with the diagnostic method of the invention.

[0191] DESCRIPTION OF THE DRAWINGS

[0192] Fig. 1. Consort diagram for the discovery cohort (IRONMET-CGM).

[0193] Fig. 2. (a) Violin plots of YFAS according to the obesity status in the discovery cohort (IRONMET-CGM, n=88). Overall and between group significance were assessed with the Kruskal-Wallis and Wilcoxon tests (two-sided), respectively. The boxplots display the first and third quartiles as the lower and upper hinges (25th and 75th percentiles) respectively, the median is represented by the middle hinge and the mean by a red dot. Whiskers extend up to 1.5 times the inter-quartile range (IQR) above the upper quartile and below the lower quartile; b) Violin plots of behavioral hallmarks of addiction motivation, according to the obesity status in the discovery cohort (IRONMET-CGM, n=88). Overall and between group significance were assessed with the Kruskal-Wallis and Wilcoxon tests (two-sided), respectively. The boxplots display the first and third quartiles as the lower and upper hinges (25th and 75th percentiles) respectively, the median is represented by the middle hinge and the mean by a red dot. Whiskers extend up to 1.5 times the inter-quartile range (IQR) above the upper quartile and below the lower quartil; c) Violin plots of behavioral hallmarks of addiction c) persistence according to the obesity status in the discovery cohort (IRONMET- CGM, n=88). Overall and between group significance were assessed with the Kruskal-Wallis and Wilcoxon tests (two-sided), respectively. The boxplots display the first and third quartiles as the lower and upper hinges (25th and 75th percentiles) respectively, the median is represented by the middle hinge and the mean by a red dot. Whiskers extend up to 1.5 times the inter-quartile range (IQR) above the upper quartile and below the lower quartile; d) Violin plots of behavioral hallmarks of addiction compulsivity according to the obesity status in the discovery cohort (IRONMET-CGM, n=88). Overall and between group significance were assessed with the Kruskal-Wallis and Wilcoxon tests (two-sided), respectively. The boxplots display the first and third quartiles as the lower and upper hinges (25th and 75th percentiles) respectively, the median is represented by the middle hinge and the mean by a red dot. Whiskers extend up to 1.5 times the inter-quartile range (IQR) above the upper quartile and below the lower quartile; e) ANCOM-BC microbiome differential abundance (two-sided Z-test using the statistic W = log2Fold Change / standard error) plot at the family level for YFAS controlling for age, BMI, education years and sex (only when analyzing individuals of both sexes) in the discovery cohort (IRONMET- CGM). Fold change associated with a YFAS score and Iog10 p-values adjusted for multiple testing; f) ANCOM-BC microbiome differential abundance (two-sided Z-test using the statistic W = log2Fold Change / standard error) plot at the family level for YFAS controlling for age, BMI, education years and sex (only when analyzing individuals of both sexes) in the validation cohort 1 (IRONMET). Fold change associated with a YFAS score and Iog10 p-values adjusted for multiple testing; g-j) Scatter plots (two-sided partial Spearman’s rank correlation test adjusted for age, BMI, sex and education years) between the iqlr-transformed Microviridae levels and g) motivation in the discovery cohort, h) persistence, i) compulsivity, and j) the sensitivity to punishment and reward (SPSR). The ranked residuals are plotted; k-p) Scatter plots for the associations between the iqlr-transformed Gokushovirus WZ- 2015a in the discovery cohort and k) YFAS, I) motivation, m) persistence, n) compulsivity, o) SPSR, and p) negative urgency. Scatter plots show tendency line with 95% confidence interval, each dote represent and independent participant.

[0194] Fig. 3. a) fastANCOM microbiome differential abundance volcano plot at the family level for YFAS controlling for age, BMI, gender, and education years as covariates in the discovery cohort IRONMET-CGM. The W-statistic represents the number of times the null hypothesis is rejected by the analysis for a given family from all K models. Beta represents the effect size given by the coefficient of the log-linear regression model of each family against the YFAS score. Dotted lines show W-statistic quantile detection thresholds: >0.6 (red), >0.7 (light blue), >0.8 (green), >0.9 (dark blue); b) ZicoSeq microbiome differential abundance volcano plot at the family level for YFAS controlling for age, BMI, gender, and education years as covariates in b) the discovery cohort (IRONMET-CGM) Fold change associated with a unit change in the YFAS score and Iog10 p-values adjusted for multiple testing are plotted for each family. Significantly different families are coloured according to the phylum; c) ANCOM-BC differential abundance volcano plot at the family level for YFAS controlling for age, BMI, gender, and education years as covariates in women of the validation cohort (IRONMET). Fold change associated with a unit change in the YFAS score and Iog10 p-values adjusted for multiple testing are plotted for each family. Significantly different families are coloured according to the phylum; d) ANCOM-BC microbiome differential abundance volcano plot at the family level for YFAS controlling for age, BMI, gender, and education years as covariates in men of the discovery cohort (IRONMET-CGM). Fold change associated with a unit change in the YFAS score and Iog10 p-values adjusted for multiple testing are plotted for each family. Significantly different families are coloured according to the phylum; e) ANCOM-BC microbiome differential abundance volcano plot at the family level for YFAS controlling for age, BMI, gender, and education years as covariates in women with obesity of the validation cohort (IRONMET). Fold change associated with a unit change in the YFAS score and Iog10 p-values adjusted for multiple testing are plotted for each family. Significantly different families are coloured according to the phylum; f) ZicoSeq microbiome differential abundance volcano plot at the family level for YFAS controlling for age, BMI, gender, and education years as covariates in women and of the discovery cohort (IRONMET- CGM). Fold change associated with a unit change in the YFAS score and Iog10 p- values adjusted for multiple testing are plotted for each family. Significantly different families are coloured according to the phylum; g) ZicoSeq microbiome differential abundance volcano plot at the family level for YFAS controlling for age, BMI, gender, and education years as covariates in men of the discovery cohort (IRONMET-CGM). Fold change associated with a unit change in the YFAS score and Iog10 p-values adjusted for multiple testing are plotted for each family. Significantly different families are coloured according to the phylum.

[0195] Fig. 4. Description of the validation cohort2 (Aging Imageomics cohort), b) Violin plots of the iqlr-transformed Microviridae and c) Gokushovirus WZ-2015a according to the obesity status in the validation cohort 2. Overall monotonic trend significance was assessed with the Kruskal-Wallis and Mann-Kendall tests (two-sided), respectively. Red dots represent the mean, d) fastANCOM microbiome differential abundance plot at the family level for inhibitory control assessed with the SCWT-WC controlling for age, BMI, and education years as covariates in men. The W-statistic represents the number of times the null hypothesis is rejected by the analysis for a given family from all K models. Beta represents the effect size given by the coefficient of the log-linear regression model of each family against the SCWT-WC score. Dotted lines show W- statistic quantile detection thresholds: >0.6 (red), >0.7 (light blue), >0.8 (green), >0.9 (dark blue). Features with a conventional detection threshold >0.7 are considered significantly abundant. Violin plots for the e,g) SCWT-WC and f,h) the SCWT- Interference according to the iqlr-transformed Microviridae and Gokushovirus WZ- 2015a quartiles after controlling for age, BMI, sex, and education years in the e,f) whole cohort and g,h) men. P for trend derived from two-sided Mann-Kendall test. The boxplots display the first and third quartiles as the lower and upper hinges (25th and 75th percentiles) respectively, the median is represented by the middle hinge and the mean by a red dot. Whiskers extend up to 1.5 times the inter-quartile range (IQR) above the upper quartile and below the lower quartile. SHAP summary plot for the i) intrinsic ignition and j) metastability in the resting-state networks associated with the iqlr-transformed Gokushovirus WZ-2015a levels identified through the Boruta algorithm after controlling for age, BMI, sex (only when applicable), and education level in men. Each dot represents an individual sample. The X-axis represents the SHAP value: the impact of a specific resting-state network on the iqlr-transformed Gokushovirus WZ-2015a levels prediction of a specific individual. Features are sorted in decreasing order based on their overall importance for final prediction (average SHAP values shown in bold).

[0196] Fig. 5. ANCOM-BC differential abundance plot at the family level for the Stroop Color Word Test - Word Color (SCWT-WC) controlling for age, BMI, sex, and education years as covariates in men from the Aging Imageomics cohort (n=475). Fold change associated with a unit change in the SCWT-WC score and Iog10 p-values adjusted for multiple testing are plotted for each family. Significantly different families are colored according to the phylum, b) SHAP summary plot for the b) intrinsic ignition and in the resting-state networks associated with the iqlr-transformed Gokushovirus WZ-2015a levels identified through the Boruta algorithm after controlling for age, BMI, sex (only when applicable), and education level in the whole cohort and c) b) SHAP summary plot for the metaestability and in the resting-state networks associated with the iqlr-transformed Gokushovirus WZ-2015a levels identified through the Boruta algorithm after controlling for age, BMI, sex (only when applicable), and education level in men, respectively. Each dot represents an individual sample. The X-axis represents the SHAP value: the impact of a specific resting-state network on the iqlr- transformed Gokushovirus WZ-2015a levels prediction of a specific individual. Features are sorted in decreasing order based on their overall importance for final prediction (average SHAP values shown in bold).

[0197] Fig. 6. a) Volcano plot of microbial genes associated with YFAS calculated by ANCOM-BC (two-sided Z-test using the statistic W = log2Fold Change / standard error) adjusting for age, sex, BMI, years of education (IRONMET-CGM). Fold change and Iog10 p-values adjusted for multiple comparisons are plotted for each family; b) Dotplot of KEGG enriched pathways based on gene set enrichment analysis (GSEA) in IRONMET-CGM; c) GSEA plot showing the distribution of the gene set and the enrichment score of the tryptophan metabolism pathway, d) Network showing the subset of genes from the tryptophan metabolism that contribute the most to the enrichment score (before the peak score); e) Volcano plot of metabolites associated with the presence or absence (<10 counts) of Gokushovirus WZ-2015a indentified employing robust linear regression model (t-statistic based on M-estimation with Huber weighting solved using Iteratively Reweighted Least Squares) adjusting for age, sex, BMI, years of education. Fold change and Iog10 p-values adjusted for multiple comparisons are plotted; f) Dot-plot of KEGG enriched pathways from metabolites significantly associated with the presence / absence of Gokushovirus WZ- 2015a; g) Metabolite concept network depicting significant metabolites involved in the selected pathways; h) Schematic representation of the metabolism of tryptophan and tyrosine, precursors of serotonin and dopamine. Tryptophan catabolism occurs via three main pathways: the kynurenine pathway (light purple, >90%), the microbial- derived indole pathway (yellow, 5-10%), and the serotonin pathway (light green). Only a small amount of tryptophan (~2%) is converted to 5-hydroxytryptophan by tryptophan hydroxylase (TPH) and then to serotonin by aromatic L-amino acid decarboxylase (AADC) with pyridoxal-5’-phosphate (vitamin B6) as a cofactor. Anthranilic acid (AA) plays a key role in the tryptophan, as it is produced in the first step of microbial tryptophan synthesis from chorismic acid and the kynurenine pathway catabolism leading to 3-hydroxyanthranillic acid (3-HAA). Tyrosine is converted to L-DOPA by tyrosine hydroxylase (TH) and then to dopamine by AADC. Monoamine oxidase (MAO) degrades serotonin and dopamine into 5- hydroxyindoleacetic acid (5-HIAA) and homovanillic acid. Tetrahydrobiopterin (BH4), tightly regulated through the one- metaboligm through the folate and methionine (methylation) cycles (dark green), is vital for serotonin and dopamine syntehsis, acting as a cofactor of AADC. Metabolites associated with Gokushovirus WZ-2015a in human plasma, Gokushovirus WZ-2015a in the nucleus accumbens I dorsal striatum (FMT experiment) or AA supplementation experiment are highlithed in bold, underlined, or italics, respectively. Metabolites in a purple boxes were negatively associated with food addiction (negative association with Gokushovirus WZ-2015a or positive association with AA supplementation), while those in red boxes were positively associated with food addiction (positive association with Gokushovirus WZ- 2015a or negative association with AA supplementation). Kynurenic acid (KYA), Xanthurenic acid (XA), 5-Hydroxyindole acetic acid (5-HIAL), 5-Hydroxyindole acetaldehyde (5-HIAA), indole-3-acetaldehyde (lAAId), indole-3-acetonitrile (IAN), indole-3-acetaldoxime (lAOx), indole-3-pyruvic acid (IPYA), 3,4- Dihydroxyphenylacetic acid (DOPAC), S-Adenosyl methionine (SAM), S- Adenosylhomocysteine (SAH), 5-Methyltetrahydrofolate (5-MTHF), Phenylethanolamine N-methyltransferase (PNMT).

[0198] Fig. 7. Transplantation of microbiota and virus from human donors with presence of Gokushovirus WZ-2015a induces food addiction in mice, a) Experimental design for the FMT study. Microbiota from n=22 human donors with absence (<10 counts, n=8) or presence (>10 counts, n=14) of Gokushovirus WZ-2015a was delivered to n=22 recipient mice pre-treated with antibiotics for 14 days. An operant conditioning was then performed for 18 days, b) Body weight during the experimental procedure in the two groups, c) accuracy index measured by the percentage of correct lever presses (active / (active+inactive)), d) operant behavior assessed by the number of active lever presses, e) persistence of response measured by the non-reinforced active lever presses in the 10 minute pellet free period (PFP), f) cognitive flexibility assessed by the number of inactive lever presses after switching the levers in day 17 so that the active became inactive, g) impulsivity measured by the non-reinforced active lever presses during the 10s time out period after each pellet delivery. The X-Y graphs represent the mean with the SEM for the mice in the presence group and the absence group. Statistics analyses were performed using a two-way repeated-measures ANOVA (t: time, G: group; txG: time x group interaction) and two-sided LSD post-hoc tests. FR1 : fixed-ratio 1 , FR5: fixed-ratio 5, RL: reversal learning; h) Experimental design for the FVT study. Microbiota from n = 20 human donors with absence (<10 counts, n=7) or presence (>10 counts, n=13) of Gokushovirus WZ-2015a was delivered to n = 20 recipient mice by oral gavage, the vehicle (PBS) was administered to n=11 mice, i) Timeline of the test performed during the food addiction experimental sequence in mice, j-l) The 3 addiction-like criteria for the experimental groups: j) motivation (PV-P=0.0007, PA-P=0.0422), k) persistence to response (PV-P=0.006, PA-P=0.0557), and I) compulsivity (PA-P=0.0925). m-o) Tests for the 3 phenotypic traits related to vulnerability to food addiction: m) impulsivity, n) appetitive associative learning (PV-P=0.0024, PA-P=0.0083), and o) cognitive flexibility. Bar graphs represent mean with S.E.M. Significance was calculated using one-way ANOVA with Fisher’s LSD test for multiple comparisons when comparing three or more groups and two-sided unpaired T-test when comparing 2 groups (# p<0.1 , * p<0.05, ** p<0.01 , and *** p<0.001).

[0199] Fig. 8. a) Schematic representation of the dopaminergic and glutamatergic projections in the sagittal and parietal sections of a mice brain, b) Volcano plot of metabolites identified in the NAc of the recipient mice according to the presence or absence (<10 counts) of Gokushovirus WZ-2015a in the human donor’s microbiota using robust linear regression models (t-statistic based on M-estimation with Huber weighting solved using Iteratively Reweighted Least Squares) controlling for donors age, sex, BMI, and education years. Fold change and Iog10 p-values adjusted for multiple comparisons are plotted; c) Dot-plot of KEGG enriched pathways in recipient mice from metabolites significantly associated with the presence or absence (<10 counts) of Gokushovirus WZ-2015a from the human donor’s microbiota (one-sided hypergeometric test). P-values adjusted for multiple comparisons are plotted (q-value <0.1); d) Metabolite concept network depicting significant metabolites (from b) involved in the selected pathways, e) Volcano plot of differentially expressed mPFC genes in recipient mice according to the presence or absence of Gokushovirus WZ- 2015a in the human donor microbiota identified using robust linear regression models (t-statistic based on M-estimation with Huber weighting solved using Iteratively Reweighted Least Squares) controlling for donors’ age, sex, BMI, and education years. Fold change and log 10 p-values adjusted for multiple comparisons are plotted; f) KEGG pathway enrichment map of significantly over-represented pathways identified from mPFC downregulated in the presence of Gokushovirus WZ-2015 in the donor microbiota. Node’s size represents the number of genes associated with each pathway, g) Network of differentially expressed genes participating in the selected pathways from the enrichment map.

[0200] Fig. 9. Dot-plot of HMDB pathways enrichment in the NAc of recipient mice from metabolites significantly associated with the presence or absence (<10 counts) of Gokushovirus WZ-2015a from the human donors microbiota. Dots are coloured according to q value; b) Metabolite concept network depicting significant metabolites involved in the selected pathways; c) Volcano plot of metabolites identified in the DS of the recipient mice according to the presence or absence (<10 counts) of Gokushovirus WZ-2015a in the human donor’s microbiota using robust linear regression model controlling for donors age, sex, BMI, and education years, d) Dotplot of KEGG enriched pathways in the DS of recipient mice from metabolites significantly associated with the presence or absence (<10 counts) of Gokushovirus WZ-2015a from the human donors microbiota. Dots are coloured according to q value, e) Metabolite concept network depicting significant metabolites involved in the selected pathways.

[0201] Fig. 10. Associations of obesity, tryptophan-related metabolites and microbial genes with the Gokushovirus WZ-2015a. a) Description of the validation cohort 3 (Healthy Imageomics). b) Violin plots of the iqlr-transformed Microviridae levels according to the obesity status and c) Violin plots of the BMI according to the iqlr-transformed Gokushovirus WZ-2015a quintiles in the validation cohort 3 (Kruskal-Wallis and two- sided Dunn tests). The boxplots display the first and third quartiles as the lower and upper hinges (25th and 75th percentiles) respectively, the median is represented by the middle hinge and the mean by a red dot. Whiskers extend up to 1.5 times the inter-quartile range (IQR) above the upper quartile and below the lower quartile, d) Volcano plot of differential microbiome families associated with the presence of food addiction in all individuals (n=147), e) the presence of food addiction in women (n=101), and f) motivation in men (n=46), identified using ANCOM-BC (two-sided Z- test using the statistic W = log2Fold Change / standard error) controlling for age, BMI, and sex (only when analyzing individuals of both sexes). Fold changes and Iog10 p- values adjusted for multiple comparisons are plotted. g) Boxplots of the normalized variable importance measure for the tryptophan-related metabolites associated with the iqlr-transformed Gokushovirus WZ-2015a identified using the Boruta algorithm in the validation cohort 3 (n=829). h) SHAP summary plot for the tryptophan-related metabolites associated with the iqlr-transformed Gokushovirus WZ-2015a identified using the Boruta algorithm in the validation cohort 3 (n=829). i) Log2 fold change (FC) of tryptophan-related molecular functions (KEGG orthologs, KO) associated with the iqlr-transformed Gokushovirus WZ-2015a levels controlling for age, BMI, sex, and education level identified using fastANCOM (two-sided Wald test). Bars are colored according to the p-value adjusted for multiple testing, j) Schematic representation of the tryptophan metabolism. Metabolites and microbial molecular functions negatively associated with the iqlr-transformed Gokushovirus WZ-2015a levels are highlighted in red. Fig. 11. a) Dot-plot of KEGG enriched pathways in the mPFC of recipient mice from genes significantly downregulated with the presence or absence (<10 counts) of Gokushovirus WZ-2015a from the human donors microbiota. Dots are coloured according to q value; b) Dot-plot of GO biological processes enrichment analysis from genes significantly downregulated with the presence or absence (<10 counts) of Gokushovirus WZ-2015a from the human donors microbiota in the mPFC of recipient mice. Dots are coloured according to q value; c) Gene concept network depicting significant genes involved in the selected metabolic processes.

[0202] Fig. 12. Associations of AA supplementation and food addiction in mice, a) Timeline of the food addiction experimental sequence in mice, n=23 mice were divided in two groups: control (n=11) and supplemented with 50mg / Kg with AA (n=12). b) Operant conditioning maintained by chocolate-flavored pellets according to the dietary groups. The X-Y graph represent the mean with the SD. Statistics analyses were performed using a two-way repeated measures ANOVA (s: Session, G: group; txG: time x group interaction) and two-sided LSD post-hoc tests. FR1 : fixed-ratio 1 , FR5: fixed-ratio 5. c-e) The 3 addiction-like criteria for the experimental groups: c) motivation (P=0.031), d) persistence to response (P=0.13), and e) compulsivity. f-h) Tests for the 3 phenotypic traits related to vulnerability to food addiction: f) impulsivity (P=0.06), g) appetitive associative learning, and h) cognitive flexibility. Bar graph represent median with interquartile range. i,j) Bar graphs representing relative abundance of AA in the i) nucleus acumens and the j) dorsal striatum of mice with (n=11) and without dietary (n=8) supplementation of AA. The X-Y graph represent the mean with the SEM. Statistics analyses were performed using two sided Unpaired T-test (***p<0.001 , **p<0.01 , *p<0.05, #p<0.1). k) Volcano plot of Nac differential metabolites between the AA and control groups identified employing robust linear regression models (t- statistic based on M-estimation with Huber weighting solved using Iteratively Reweighted Least Squares). Fold change and Iog10 p-values adjusted for multiple comparisons are plotted. I) Dot plot of significant (q-values <0.1) over-represented KEGG pathways identified from significantly differential metabolites between the AA and control groups, m) Metabolite concept network depicting significant metabolites involved in the selected pathways, n) Schematic representation of the glutamate and tyrosine metabolism and its connection to the TAG cycle. Metabolites highlighted in bold, italics were associated with food addiction in human plasma and in the AA supplementation experiment in mice, respectively. Metabolites in a purple box were negatively associated with food addiction (positive association with AA supplementation), those in red boxes were positively associated with food addiction (negative association with AA supplementation). Alpha-ketoglutarate (a- ketoglutarate), Tetrahydrofolate (THF), 5,10-methylenetetrahydrofolate (5,10- diMTHF), Succinate (SSA), gamma-aminobutyric acid (GABA), succinic semialdehyde (SSA). o) Gene concept network showing significant genes participating in selected significant pathways downregulated after AA supplementation.

[0203] Fig. 13. Dot plot of KEGG module-based pathway over-representation analysis based on the molecular function significantly associated with the iqlr-transformed Gokushovirus WZ-2015a levels identified using fastANCOM (padj<0.05).

[0204] Fig. 14. Associations of AA supplementation in feeding and adiction like behaviors in Drosophila and a second FMT from human donors with presence of Gokushovirus. Boxplots for a) 48-h food intake measured by EX-Q (PV-AA<0.0001), b) dye uptake after 40h of starvation (PV-AA<0.0001), and c) body weight in flies fed SD and SD supplemented with 200 mg / L of AA (PV-AA<0.0001). d) EX-Q when Dop2R was downregulated (PControl-RNAi<0.0009) or e) overexpressed in neurons (PControl- UAS<0.0282). f) Schematic representation of the two-choice CAFFE assay. PI to 15% ethanol food compared with non-ethanol food for g) control and AA dietary supplemented flies (PD3_C-D3_AA=0,0162), h) flies with mushroom body downregulation of Dop2R with and without dietary supplementation of AA. PI was calculated from consumption values over three consecutive days, i) Cumulative plot of the PI over time and PI at 40 min (PimpTNT-impTNT_AA=0.0072). A minimum of three biological replicates were performed per experimental condition (panels a-e,i). Boxplots (a-e) represent median with 25-75th percentile and whiskers 5-95 percentile, X-Y graphs (g-e,i) and bar graphs (i) represent mean with S.E.M. Significance was calculated using two sided unpaired t-test (a-c,e), Welch’s test (d), one-way Brown- Forsythe ANOVA with Dunnett’s correction for multiple comparisons (g-h), or Mann- Whitney II test (i) (*p<0.05, **p<0.01 , and ***p<0.001 , ****p<0.0001). j) Experimental design for the FMT study, k) Scatter plots (two-sided partial Spearman’s rank correlation test) between the human donor iqlr transformed Gokushovirus WZ-2015a levels and the recipient mice weight. Plots show the tendency line with the 95% Cl. I.m) Violin plots for the recipient mice weight according to the presence or absence of Gokushovirus WZ-2015a in I) men (absence n=3; presence n=7) and m) women donors (absence n=8; presence n=6) (two-sided Mann-Whitney test). The boxplots display median with 25th-75th percentiles. Whiskers extend up to 1.5 times the interquartile range, n) Volcano plot of differentially expressed genes in the mPFC of the recipient mice associated with the presence / absence of Gokushovirus in the human donor identified using robust linear regression models (t-statistic based on M- estimation with Huber weighting solved using Iteratively Reweighted Least Squares). Fold change and Iog10 p-values adjusted for multiple comparisons are plotted, o) KEGG-based pathway over-representation analysis of the recipient’s mice differentially expressed mPFC genes (one-sided hypergeometric test). P-values adjusted for multiple comparisons are plotted (q-value <0.1). p) Volcano plot of differential recipient mice fecal microbiota associated with the human donor iqlr- transformed Gokushovirus abundances identified using ANCOM-BC (two-sided Z-test using the statistic W = log2Fold Change / standard error). Fold changes and Iog10 p- values adjusted for multiple comparisons are plotted.

[0205] Fig. 15. a) Dot plot of significant (q-values <0.1) over-represented HMDB pathways identified from significantly differential metabolites between the AA and control mice in the NAc; b) Metabolite concept network depicting significant metabolites involved in the selected pathways.

[0206] Fig. 16. a) Volcano plot of metabolites associated with YFAS identified employing robust linear regression model in the IRONMET-CGM cohort adjusting for age, sex, BMI, years of education, b) Scatter plots of the partial Spearman’s rank correlations (adjusted for age, BMI, sex and education years) between 3-ethilmalic acid and YFAS residuals. Scatter plots show tendency line with 95% confidence interval, each dote represent and independent participant; c) Dot-plot of KEGG enriched pathways from metabolites significantly associated with YFAS in IRONMET-CGM cohort; d) Metabolite concept network depicting significant metabolites involved in the selected pathways; e) Dot plot of significantly over-represented HMDB pathways identified from significantly differential metabolites associated with YFAS identified in the IRONMET-CGM; f) Metabolite concept network depicting significant metabolites involved in the selected pathways.

[0207] Fig. 17. Cumulative plot of the PI over time. The middle line represents the mean and the shading the S.E.M and Cumulative PI at 40 min of flies choosing between a 5% sucrose or 5% sucrose food + 15% ethanol. Flies expressing impTNT or TNT in the dopaminergic neurons by means of the ple-Gal4 promoter with and without dietary supplementation of AA (200mg / L). PI ranging from -1 to +1 , positive values indicate a preference for ethanol and negative values indicate preference for sucrose (PimpTNT-TNT<0.0001). The graph represents data from a minimum of 3 independent experiments. Bar graphs represent mean with S.E.M. Significance was calculated using Mann-Whitney test (****p<0.0001).

[0208] Fig. 18. a) Gene-gene interaction network constructed using differentially expressed mPFC genes (abs(logFC>1), pFDR<0.2) via the Search Tool for the Retrieval of Interacting Proteins / Genes (STRING) database. The network nodes are genes and the edges represent the predicted functional interactions. The thickness indicates the degree of confidence prediction of the interaction; b) Dot plot of GO BP, CC and MF over representation analysis of differentially expressed mPFC genes according to the presence or absence (<10 counts) of Gokushovirus WZ-2015a in the human donor’s microbiota; c) Gene-concept network depicting the connections of those genes involved in the GO biological process, cellular compartment and molecula function over-representation results, d) Gene-concept network depicting the connections of those genes involved in the KEGG-based over-representation results.

[0209] Fig. 19. Gut microbiome and YFAS in IRONMET-CGM. a-i) ANCOM-BC microbiome differential abundance (two-sided Z-test using the statistic W = log2FoldChange / standard error) volcano plot at the family level for a) YFAS motivation, b) YFAS persistence, c) YFAS compulsivity for the discovery cohort IRONMET-CGM; d) YFAS motivation, e) YFAS persistence, f) YFAS compulsivity for women of the discovery cohort IRONMET-CGM; g) YFAS motivation, h) YFAS persistence, i) YFAS compulsivity for men of the discovery cohort IRONMET-CGM. Fold change associated with a unit change in the YFAS score and Iog10 p-values adjusted for multiple testing are plotted for each family. Heat map of two-sided Spearman’s correlations between iqlr-transformed abundances of Gokushovirus WZ- 2015a and YFAS, addiction-like criteria (motivation, persistence, compulsivity) and sensitivity to reward (SPSR) and punishment (Nil (LIPPS)) for j) women and k) men. All analyses were performed controlling for age, BMI, sex, and education years as covariates.

[0210] Examples

[0211] 1. Methods and materials

[0212] 1.1. Clinical cohorts

[0213] Discovery Cohort (IRONMET-CGM, n=88)

[0214] The Glucose, Brain and Microbiota study (IRONMET+CGM, ClinicalTrials.gov Identifier: NCT03889132) is a case-control study conducted at the Endocrinology Department of Dr. Josep Trueta University Hospital. The recruitment of subjects started in March 2019 and finished in January 2022. The cohort includes a subset of middle-aged patients (47.6 years [41.5-54.9]) with obesity (body mass index (BMI) >30 kg / m2) and age-matched and sex-matched subjects without obesity (BMI 18.5- <30 kg / m2) that have measurements of the Yale Food Addiction Score (YFAS). Exclusion criteria included: type 2 diabetes mellitus, chronic inflammatory systemic diseases, acute or chronic infections in the previous month; use of antibiotic, antifungal, antiviral or treatment with proton pump inhibitors; severe disorders of eating behavior or major psychiatric antecedents; neurological diseases, history of trauma or injured brain, language disorders and excessive alcohol intake (>40 g OH / day in women or 80 g OH / day in men). The Institutional review board - Ethics Committee and the Committee for Clinical Research (CEIC) of Dr. Josep Trueta University Hospital (Girona, Spain) approved the study protocol and informed written consent was obtained from all participants.

[0215] Validation Cohort 1 (IRONMET, n=29) This is a cross-sectional case-control study conducted at the Endocrinology Department of Dr. Josep Trueta University Hospital. The recruitment of subjects started in January 2016 and finished in October 2017. Consecutive middle-aged subjects, 27.2-66.6 years, were included. Patients with obesity (body mass index (BMI) >30 kg / m2) and age-matched and sex-matched subjects without obesity (BMI 18.5-<30 kg / m2) were eligible. Only a subset of patients that underwent food addiction measurement were included. Exclusion criteria included: type 2 diabetes mellitus, chronic inflammatory systemic diseases, acute or chronic infections in the previous month; use of antibiotic, antifungal, antiviral or treatment with proton pump inhibitors; severe disorders of eating behavior or major psychiatric antecedents; neurological diseases, history of trauma or injured brain, language disorders and excessive alcohol intake (>40 g OH / day in women or 80 g OH / day in men). The Institutional review board - Ethics Committee and the Committee for Clinical Research (CEIC) of Dr. Josep Trueta University Hospital (Girona, Spain) approved the study protocol and informed written consent was obtained from all participants.

[0216] Validation Cohort 2 (Aging Imageomics, n=942)

[0217] The Aging Imageomics Study is an observational study including participants residing in the province of Girona (Northeast Catalonia, Spain) recruited for two independent cohort studies: the Maturity and Satisfactory Ageing in Girona study (MESGI50) and the Improving intermediate Risk management study (MARK). Detailed description of the cohorts can be found elsewhere (Puig, J. et al. The aging imageomics study: rationale, design and baseline characteristics of the study population. Meeh Ageing Dev 189, 111257 (2020)). Briefly, the MESGI50 cohort included a representative population of the province of Girona aged > 50 years old, while the MARK cohort included a random sample of patients aged 35-74 years with intermediate cardiovascular risk recruited in public primary care centers. Eligibility criteria included: age > 50 years, dwelling in the community, no history of infection during the last 15 days, no contraindications for MRI, and consent to be informed of potential incidental findings. The Aging Imageomics Study protocol was approved by the ethics committee of the Dr. Josep Trueta University Hospital.

[0218] Validation Cohort 3 (Health Imageomics, n=835) The Health Imageomics Study is a multicentric observational study also including participants residing in the province of Girona (Northeast Catalonia, Spain). It includes a representative population of the province of Girona aged >16 and < 52 years old recruited in public primary care centers. Exclusion criteria included: age > 52 years, not dwelling in the community ( / .e. institutionalized), terminal disease, intellectual disability, diagnosis of dementia, difficulties to understand Catalan or Spanish. The Health Imageomics protocol was approved by the ethics committee of the Dr. Josep Trueta University Hospital. A subset of participants (n=147) underwent measurement of food addiction.

[0219] 1.2. Neuropsychological Assessment in Humans

[0220] Yale Food Addiction Scale (YFAS)

[0221] YFAS 2.0 is a validated, auto-administered test that consists of 36 items based on the Diagnostic and Statistical Manual of Mental Disorders (DSM-V) criteria for substance dependence, but in this case related to food addiction, including cognitive distortion, craving, abstinence, and tolerance (Granero, R. et al. Validation of the Spanish Version of the Yale Food Addiction Scale 2.0 (YFAS 2.0) and Clinical Correlates in a Sample of Eating Disorder, Gambling Disorder, and Healthy Control Participants. Front Psychiatry 9, 208 (2018)). A higher Y-FAS total score indicates more food addiction. Moreover, the test calculated three variables related to addiction: compulsivity (criteria 11 of the DSM-V) related to inhibitory control, motivation (criteria 9-10 of the DSM-V) related to the necessity of consummation, and persistence in response (criteria 6-7 of the DSM-V), difficulty to stop pursuing rewards.

[0222] Sensitivity to Punishment and Sensitivity to Reward (SPSR)

[0223] The Sensitivity to Punishment and Sensitivity to Reward (SPSR) Questionnaire is an auto-administered test based on Gray’s anxiety and impulsivity dimensions (Torrubia, R., Avila, C., Molto, J. & Caseras, X. The Sensitivity to Punishment and Sensitivity to Reward Questionnaire (SPSRQ) as a measure of Gray’s anxiety and impulsivity dimensions. Pers Individ Dif31 , 837-862 (2001)). It is a 48-item yes-or-no survey that includes two scales: the sensitivity to punishment (24 items) and the sensitivity to reward (24 items). It is related to the behavior inhibition system (BIS) dimension, which is associated with anxiety, and the behavior approach system (BAS) dimension, which is associated with motivation to act in response to rewards. We obtained a sum score of the two scales “Total SPSR” in order to encompass a score that refers to the motivation system according to Corr et al. 68, which it take into account the both systems; punishment and reward as only one dimension “approach and avoidance system”.

[0224] Stroop Color-Word Test (SCWT) (Golden’s version)

[0225] The SCI / VT was administered to assess cognitive flexibility, selective attention, inhibition and information processing speed. This version consists of three different parts: 1) 100 words (color names) are printed in black ink and the subject is asked to read them as fast as possible; 2) 100 “XXX” are printed in color ink (green, blue and red) and the subject is asked to name as fast as possible the ink color; and 3) 100 color names (from the first page) printed in color ink (from the second page), the color name and the ink color do not match and the subject is asked to name the ink color (and not to read the color name). The subject is given 45 seconds for each task and the last item completed is noted, obtaining three scores: one for each part of the test (word “W’; color “C”; and word-COLOR “WC”). The interference (“I”) index was also obtained from the subtraction WC-WC’ (WC’=WxC / W+C). Standard administration procedures were followed as indicated in the test manual (Golden, C. A Manual for the Clinical and Experimental Use of the Stroop Color and Word Test. Faculty Books and Book Chapters (1978)).

[0226] UPPS-P Impulsive Behavior Scale

[0227] The UPPS-P Impulsive Behavior Scale is a 59-item questionnaire (using a 4 point likert-type scale) designed to measure impulsivity across dimensions of the Five Factor Model of personality: Positive Urgency, Negative Urgency, Lack of Premeditation, Lack of Perseverance, and Sensation Seeking (Whiteside, S. P. & Lynam, D. R. The Five Factor Model and impulsivity: using a structural model of personality to understand impulsivity. Pers Individ Dif 30, 669-689 (2001)). Positive and Negative Urgency measure the tendency to act impulsively while facing positive and negative emotional situations, respectively. Lack of Premeditation measures the tendency to act rashly before engaging in that act. Lack of Perseverance measures the difficulty to stay focused on hard or boring tasks. Sensation Seeking measures the tendency to experience novelty. Positive Urgency, Negative Urgency, and Sensation Seeking, are “reversed-scored” items (1 : strongly disagree; 4: strongly agree), with higher ratings indicating less impulsivity. Lack of Premeditation and Perseverance are “standard-scored” items (1 : strongly agree; 4: strongly disagree), with higher ratings indicating greater impulsivity. We used a validated short 20-item Spanish version of the LIPPS-P (Candido, A., Orduha, E., Perales, J. C., Verdejo- Garcia, A. & Billieux, J. Validation of a short Spanish version of the LIPPS-P impulsive behaviour scale. Trastor Adict 14, 73-78 (2012)). In this study we only focused on Positive and Negative Urgency.

[0228] 1.3. Extraction of Faecal Genomic DNA and Whole-Genome Shotgun Sequencing

[0229] Discovery cohort (IRONMET-CGM) and Validation cohort 3 (IRON MET)

[0230] Total DNA was extracted from frozen human stools using the QIAamp DNA mini stool kit (QIAGEN, Courtaboeuf, France). Quantification of DNA was performed with a Qubit 3.0 fluorometer (Thermo Fisher Scientific, Carlsbad, CA, USA), and 1 ng of each sample (0.2 ng / pl) was used for shotgun library preparation for high-throughput sequencing, using the Nextera DNA Flex Library Prep kit (Illumina, Inc., San Diego, CA, USA) according to the manufacturers’ protocol. Sequencing was carried out on a NextSeq 500 sequencing system (Illumina) with 2 X 150-bp paired-end chemistry, at the facilities of the Sequencing and Bioinformatic Service of the FISABIO (Valencia, Spain).

[0231] Validation cohorts 2 and 3 (Aging and Health Imageomics) and mouse FMT 2 study DNA was extracted from stool samples using the PowerSoil DNA extraction kit (MO BIO Laboratories) following the manufacturer's protocol. Between 400 and 500 ng of total DNA were used for library preparation for Illumina sequencing employing Illumina DNA Prep kit (Illumina). All libraries were assessed using a TapeStation High Sensitivity DNA kit (Agilent Technologies) and quantified by Qubit (Invitrogen). Validated libraries were pooled in equimolar quantities and sequenced as a paired- end 150-cycle run on an Illumina NextSeq2000. Raw reads were filtered for QV>30 using an in-house python script. For the taxonomic and functional diversity analysis of the microbiota present in the samples of all cohorts, FASTQ output files were first pre-processed using fastp (Chen, S., Zhou, Y., Chen, Y. & Gu, J. fastp: an ultra-fast all-in-one FASTQ preprocessor. Bioinformatics 34, i884— i890 (2018)), a FASTQ data pre- processing tool for quality control, trimming of adapters, and quality filtering. Clean reads were mapped against the Homo sapiens genome database (GRCh38.p13) using Bowtie2 (Langmead, B. & Salzberg, S. L. Fast gapped-read alignment with Bowtie 2. Nat Methods 9, 357-9 (2012)) to remove reads from human origin. Unmapped reads were run using the SqueezeMeta v1.3.1 (Tamames, J. & Puente-Sanchez, F. SqueezeMeta, a highly portable, fully automatic metagenomic analysis pipeline. Front Microbiol 9, 3349 (2019)) using the co-assembly mode to pool all samples in a single assembly. Contigs assembly was carried out with megahit (Li, D., Liu, C. M., Luo, R., Sadakane, K. & Lam, T. W. MEGAHIT: An ultra-fast single-node solution for large and complex metagenomics assembly via succinct de Bruijn graph. Bioinformatics 31 , 1674-1676 (2015)) the mapping of reads in contigs was performed with Bowtie2. Prodigal (Hyatt, D. et al. Prodigal: Prokaryotic gene recognition and translation initiation site identification. BMC Bioinformatics 11 , 119 (2010)) was used for ORFs prediction, Diamond (Buchfink, B., Reuter, K. & Drost, H. G. Sensitive protein alignments at tree- of-life scale using DIAMOND. Nat Methods 18, 366-368 (2021)) for ORF search and alignment against the GenBank nr database for taxonomic assignment, and the KEGG database for functional annotation. For the mouse FMT2 study, fastq files were decompressed, filtered and 3 ends-trimmed by quality, using prinseq-lite-0.20.4 (Schmieder, R. & Edwards, R. Quality control and preprocessing of metagenomic datasets. Bioinformatics 27, 863-864 (2011)) and overlapping pairs were joined using FLASH-1.2.11 (Magoc, T. & Salzberg, S. L. FLASH: fast length adjustment of short reads to improve genome assemblies. Bioinformatics 27, 2957-2963 (2011)). Fastq files were then converted into fast files, and mouse host reads were removed by mapping the reads against GRCm38.p6 reference mouse genome (Sept 2017) using Bowtie2 (Langmead, B. & Salzberg, S. L. Fast gapped-read alignment with Bowtie 2. Nat Methods 9, 357-9 (2012)) with end-to-end and very sensitive options. Taxonomic annotation was implemented with Kaiju v1.6.2 (Menzel, P., Ng, K. L. & Krogh, A. Fast and sensitive taxonomic classification for metagenomics with Kaiju. Nat Commun 7, 11257 (2016)) on mouse-free reads. Open reading frames (ORFs) from Gokushoviruses were predicted using Genmark S (Mills, R., Rozanov, M., Lomsadze, A., Tatusova, T. & Borodovsky, M. Improving gene annotation of complete viral genomes. Nucleic Acids Res 31 , 7041-7055 (2003)) and the resulting gene and protein sequences were compared against database nr of Genbank by BLAST program. Genomic comparison of all Gokushovirus genomes was performed with the VIRIDIC program with default parameters (Moraru, C., Varsani, A. & Kropinski, A. M. VIRIDIC — A Novel Tool to Calculate the Intergenomic Similarities of Prokaryote-Infecting Viruses. Viruses 12, (2020)) in order to estimate the genetic relatedness. Viral network interaction showing relatedness of Gokushovirus with other viruses in databases was performed with the Vcontac2 program as described in Martinez-Garcia, M., Martinez-Hernandez, F. & Martinez, J. M. Single-Virus Genomics: Studying Uncultured Viruses, One at a Time (2020). Identification of Gokushovirus hosts was performed with CRISPR spacer-protospacer search using a massive CRISPR array database as described in detail (Shmakov, S. A., Wolf, Y. I., Savitskaya, E., Severinov, K. V. & Koonin, E. V. Mapping CRISPR spaceromes reveals vast host-specific viromes of prokaryotes. Communications Biology 2020 3:1 3, 1-9 (2020)).

[0232] 1.4. Microviridae and Gokushovirus WZ-2015a interquartile log ratios

[0233] Taxa and bacterial functions were filtered so that only those with >10 reads in at least 10% of the samples were selected. Then, to take into account the compositional structure of the metagenomics data and rule out possible spurious associations, we applied an interquartile log-ratio ( / q / r) transformation to the filtered raw counts using the “ALDEx2” R package 85. Initially, a uniform prior of 0.5 was applied to the dataset and then a centered log-ratio (c / r) was calculated for each sample ( / ): where n^ are the raw counts for the / th taxon in the jth sample and g(N) is the geometric mean of N, g ... nD. Afterwards, the taxa that were between the first and third quartile of the variance of the dr values across all samples were retained. Then, the geometric mean abundance of only the retained features was calculated and used as the denominator for log-ratio calculations. In this case, the estimation of the 0 counts for the geometric mean calculation was carried out using a probability distribution. Thus, the table of reads counts for the retained features was converted to a distribution of posterior probabilities using a Monte Carlo sampling (K=128 instances) from a Dirichlet distribution for each sample: [pi, p2, ... ] £)iric / iZet([n1, n2, ... ] + 0.5). An uninformative prior of 0.5 was used to model the frequency of features with zero counts.

[0234] This multivariate distribution ensures that none of the inferred proportions is ever exactly zero even if the associated count is zero, and that the probability is conserved. The marginal distributions of i are wide when the associated read count is small and narrow if the number of counts are large. Therefore, it accounts for the fact that one read out of 100 total reads has much lower precision than 1 ,000 reads out of 100,000 total reads, despite having the same proportion. Each Monte Carlo instance kj was then transformed using a log-ratio. Finally, we calculated the median of the iqlr- transformed instances and obtained the corresponding values for the Microviridae or Gokushovirus HZZ-2015a.

[0235] 1.5. Resting-state functional Magnetic Resonance Imaging {Ageing Imageomics)

[0236] Image acquisition and pre-processing

[0237] Image acquisition and pre-processing for the fMRI study have been described in detail elsewhere (Escrichs, A. et al. Whole-Brain Dynamics in Aging: Disruptions in Functional Connectivity and the Role of the Rich Club. Cereb Cortex 31 , 2466-2481 (2021)). All images were acquired on a 1.5T Ingenia (Philips Healthcare, Best, The Netherlands) with eight channel head coils. As a part of a larger study protocol a multislice fluid attenuation inversion recovery (T2-FLAIR) with TR / TE / TI =6500 / 120 / 2200 ms, flip angle 90°, in-plane resolution 0.78x0.78mm, slice thickness 5mm without gap and 20 axial slices was used to exclude pre-existing brain lesions. Resting state functional MRI (rs-fMRI) was performed acquiring a gradient echo-planar imaging (EPI) sequence [repetition time (TR) = 2500 ms; echo time (TE) = 40 ms; flip angle = 83°; field of view (FOV) = 230 x 230 mm; and voxel size = 3.5 x 3.5 x 5 mm without gap] with 122 continuous functional volumes acquired axially for 5 min. Subjects were asked to relax, remain as motionless as possible, remain awake, and keep their eyes closed to minimize stimuli, including visuals. EPI images were automatically oriented using Conn and pre-processed as described in Escrichs, A. et al. using the Data Processing Assistant for Resting-State fMRI (DPARSF) toolbox based on Statistical Parametric Mapping (SPM12). The time series were used to parcellate the wholebrain functional network and the eight large-scale resting-state networks.

[0238] Intrinsic-Ignition Framework

[0239] We applied the Intrinsic-Ignition Framework to obtain the effect of naturally occurring activation events that reflect the capability of a given brain area to propagate activity to other brain areas. In brief, we transformed the BOLD time series to phase space by filtering the signals within the narrowband (0.04-0.07 Hz) and computed the Hilbert transform to obtain the phases of the signal between each pair of brain areas at each time point. The binary events were defined by transforming the time series into z- scores, zi(t), and fixing a threshold, 0. Then, a phase lock matrix Pjk(t), was computed which describes the state of phase synchronization between brain areas j and k at time f as: where <Pj(t) and <Pj(t) correspond to the phases of the BOLD time series for brain areas / and k at timet. Then, the integration was defined by measuring the length of the largest connected component in the phase lock matrix Pjk(t) and the value of integration was computed as the length of the connected component considered as an adjacent graph ( / .e., the largest subcomponent). Finally, for each brain area, we averaged across the network the integration evoked at each time t within the set time window and also obtained the standard deviation. The mean integration (ignition) reflects spatial diversity and the broadness of communication across the whole brain network. Higher intrinsic-ignition values reflect higher functional connectivity. The standard deviation (metastability) represents the variability over time across the whole-brain network. Higher metastability reflects more flexible switching across time (i.e. more complex brain dynamics).

[0240] 1.6. Metabolomics analyses

[0241] Global untargeted plasma metabolic profiling (IRONMET-CGM) Samples were randomized before any procedures to eliminate potential technical variations from sample preparation and instrument drift. Each individual sample was vortexed then centrifuged at 15,000 g for 5 min. Supernatant was taken and split into four aliquots for different labeling methods, backup and preparation of pooled sample. To prepare the pooled sample, which was used as the reference, 30 pL were taken from each individual sample, and then combined and mixed thoroughly. For protein precipitation all samples were spun down. Then, 90 pL of LC-MS grade methanol were added to perform protein precipitation. The methanol extract was completely dried after incubation at -20 °C for 0.5 hour, then temporally stored in -80 °C freezer until labeling.

[0242] Chemical Isotope Labeling: Data was collected using Dansyl-labeling Kit or DmPA- labeling Kit. The amine- / phenol- labelling protocol strictly followed the SOP provided in the kit. Briefly, buffer reagent (Reagent A) and12C2-labeling (for the individual samples and the pooled sample) or13C2-labeling (for the pooled sample) reagent (Reagent B) was added into samples. The mixtures were incubated at 40 °C for 45 minutes. After that, quenching reagent (Reagent C) was added to quench the excessive labelling reagent. The mixtures were incubated at 40 °C for another 10 minutes. Finally, pH adjusting reagent (Reagent D) was added. The carboxyl- labelling - labelling protocol strictly followed the SOP provided in the kit. Briefly, catalysing reagent (Reagent A) and12C2-labeling (for the individual samples and the pooled sample) or13C2-labeling (for the pooled sample) reagent (Reagent B) was added into samples. The mixtures were incubated at 80 °C for 60 minutes. After that, quenching reagent (Reagent C) was added to quench the excessive labelling reagent. The mixtures were incubated at 80 °C for another 30 minutes and the chemical isotope labelling procedure was complete. The12C2-labeled individual sample was then mixed with13C2-labeled reference sample in equal volume. The mixture was ready to be analysed by LC-MS. Prior to LC-MS analysis of the entire sample set, quality control (QC) sample was prepared by equal volume mix of a12C-labeled and a13C-labeled pooled sample.

[0243] LC-MS Analysis Condition: The LC-MS analysis was strictly followed the SOP (i.e., Rapid LC-MS Analysis for HP-CIL Metabolomics Platform). QC samples were injected every 20 sample runs to monitor instrument performance. The LC analysis was performed on a Thermo Scientific Vanquish LC linked to Bruker Impact II QTOF Mass Spectrometer (plasma samples) or Agilent 1290 LC linked to Agilent 6546 QTOF Mass Spectrometer (brain mice samples) while chromatographic separation with an Agilent eclipse plus reversed-phase C18 column (150 x 2.1 mm, 1.8 pm particle size). The mobile phase was composed of 0.1 % formic acid in water (v / v) (A) and 0.1% formic acid in acetonitrile (v / v) (B). The column temperature was maintained at 40 °C, and gradient elution was performed with a flow rate of 400 pL / min starting at 25% B increasing to 99% B over 10 min, maintained until 15 min, and finally returning to 25% B at 15.1 min until 18 min. Mass range = m / z 220-1000 and acquisition rate = 1 Hz.

[0244] Data Processing: Analysis was performed using IsoMS Pro 1.2.16 (NovaMT Inc.) and NovaMT Metabolite Database v3.0. LC-MS data from 2-channel analysis were exported to .csv file with Bruker DataAnalysis 4.4. The exported data were uploaded to IsoMS Pro 1 .2.16. After Data Quality Check, data Processing was performed. Only metabolites identified with high-confidence (>90%) using a two-tier approach were used. In tier 1 peak pairs were searched against a labelled metabolite library (CIL library) based on accurate mass and retention time, while in tier 2 a linked identity library (LI library) was used for identification based on accurate mass and predicted retention time matches.

[0245] Targeted analysis of tryptophan metabolism by UPCL-MS / MS (Health Imageomics)

[0246] Serum samples (100 pL) were placed in a centrifuge tube and mixed with 10 pL of internal standard (0.2 mg / L nicotinic acid-d4, 1 mg / L serotonin-d4, 0.2 mg / L kynurenic acid-d5, 100 mg / L Tryptophan-d5, 0.2 mg / L xanthurenic acid-d4, 0.2 mg / L 3- Hydroxyanthranilic acid-d3, 0.5 mg / L 5-Hydroxyindole-3-acetic Acid-d2, 10 mg / L Kynurenine-d4, 1 mg / L 3-lndolepropionic-d2, 1 mg / L lndole-2, 4,5,6, 7-d5-3-acetic-a,a- d2, and 1 mg / L 2-Picolinic acid-d4) and 500 pL of 10 mM ammonium formate in methanol. Samples were vortexed and centrifuged for 5 minutes at 15000 rpm and 4 °C. Supernatants were purified with Phree phospholipid removal plates (Phenomenex, California, United States), evaporated and resuspended in 100 pL of 0.1 % formic acid in methanokwater (2:8) and transferred to glass vials for analysis. The chromatographic separation was performed with a gradient. Mobile phase A was 0,1 % formic acid and B was 0,1% formic acid in methanol. The column temperature was set at 50 °C and the injection volume was 1 pL.

[0247] The assignment of metabolites was performed by direct comparison with commercial standards for all the analytes. The method was validated with standard addition to a pool of serum samples. The same pool of samples was analysed at different times throughout the study and was used as quality control. Repeatability was assessed as intraday (the same day) and interday (five different days) deviation. In order to normalize the signalling from different samples across all the analyses, several internal standards were added to the extractions for mass spectrometry analyses, allowing the normalization and the possibility to calculate the absolute quantification of each metabolite identified.

[0248] 1.7. Mice experimental procedures

[0249] Animal maintenance

[0250] All the behavioral experiments were conducted in the animal facility at Universitat Pompeu Fabra-Barcelona Biomedical Research Park (UPF-PRBB; Barcelona, Spain). All animal procedures were performed in accordance with the guidelines of the European Communities Council Directive 2010 / 63 / EU regulating animal research and were approved by the local ethical committee (Comite Etic d'Experimentacio Animal-Parc de Recerca Biomedica de Barcelona, CEEA-PRBB). Upon arrival to the animal facilities, all animals were let to adapt during 5 days to housing conditions (12 hours reversed light / dark cycle, 08:00 AM lights off). Mice were housed individually in controlled laboratory conditions with temperature maintained at 21±1 °C and humidity at 55 ± 10%; with food and water available ad libitum. All the experiments were performed during the dark phase of a reverse light / dark cycle (light on at 8:00 pm, light off at 8:00 am). All the experiments were performed under blind and randomized conditions.

[0251] 1.8. Fecal Microbiota Transplantation (FMT) experiment 1

[0252] Experimental design Twenty-two wild-type C57BL / 6J male mice were used, animals were divided into 2 groups: a) FMT from donors without Gokushovirus WZ-2015a (<10 counts, n=14); and b) FMT from donors with Gokushovirus WZ-2015A (>10 counts, n=8). Mice were given an ad libitum cocktail of antibiotics during 14 days in drinking water to deplete gut microbiota. Antibiotic cocktail consisted of ampicillin (1 g / L), metronidazole (1 g / L), vancomycin (400 mg / L), ciprofloxacin HCI (250 mg / L) and imipenem (250 mg / L). Afterwards, mice were subjected to a 72 hours wash out and then, the FMT groups were colonized via daily oral gavage of donor microbiota (200 pL) for 3 days. Booster inoculations were given twice weekly throughout the study to reinforce donor microbiota phenotype. Ten days after the first oral gavage animals were subjected to an operant behavior procedure (see below) for the next 18 days. Food and water were available ad libitum during all the experiment. Body weight gain of mice was controlled during all the experiment.

[0253] Operant behavior procedure

[0254] Mouse operant chambers (Model ENV-307A-CT, Med Associates, Georgia, VT, USA) were used for operant responding. At the start of each food self-administration session, a house ceiling light turned on during the first 3 seconds of the session to indicate the start of the session. All sessions lasted 60 min and regular-flavored pellets were used. The food self-administration session consisted of two pellet periods of 25 min and a 10 min pellet-free period in between both pellet periods (25 / 10 / 25). In the two pellet periods, animals received a pellet after an active response paired with a stimulus light (cue light). After performing an active response on the active lever, a time-out period of 10 sec was set where the cue light was off and no reward (pellet) was provided. No pellets were provided in the inactive lever. Responses on active lever, inactive lever and during the time-out period were recorded. The start of the pellet-free period was signaled by the illumination of the entire operant chamber. During this period no pellet was delivered. In the operant conditioning sessions, mice were under fixed ratio 1 (FR1) of reinforcement during 7 days (one active lever-press resulted in a delivery of one pellet). Following FR1 phase, animals were subjected to an increase FR up to 5 (5 lever-presses in order to obtain one reward) for 8 days. After each session mice were returned to their home cages. Persistence to response

[0255] Non-reinforced active responses during the pellet-free period (10 min) were measured as a persistence of food-seeking behavior.

[0256] Cognitive flexibility

[0257] After 8 days of FR5, animals were exposed to 2 sessions of reversal learning (RL). In these 2 sessions, active and inactive levers were switched. Thus, the active lever during FR1 and FR5 phases became inactive and vice versa. Higher number of leverpresses in the inverted active lever (inactive during FR phases) indicates higher scores of cognitive flexibility.

[0258] Impulsivity

[0259] Non-reinforced active responses during the time-out periods (10s) after each pellet delivery were measured as impulsivity-like behavior indicating inability to stop a response once it is initiated.

[0260] Statistical analysis

[0261] All statistical analysis was performed with GraphPad Prism 9.3.1. ANOVA with repeated measures was used to analyze the results. Comparisons between groups at each time point were analyzed by Student t-test. Within-subject factors were time (Days) and between-subject factor was the FMT group (2 levels: absence vs presence of Gokushovirus WZ-2015a) The criterion for significance (alpha) was set at 0.05.

[0262] 1.9. Fecal Microbiota Transplantation (FMT) experiment 2

[0263] Experimental design

[0264] Twenty-two wild-type C57BL / 6J male mice were given ad libitum cocktail of antibiotics during 14 days in drinking water to deplete gut microbiota. Antibiotic cocktail consisted of ampicillin (1 g / L), metronidazole (1 g / L), vancomycin (400 mg / L), ciprofloxacin HCI (250 mg / L) and imipenem (250 mg / L). After 14 days of antibiotic intake, animals were subjected to a 72 hours wash out and then colonized via daily oral gavage of donor microbiota (200 pL) for 3 days. Animals were orally gavaged with: a) fecal microbiota from patients with absence of Gokushovirus WZ-2015a (<5 Count, n = 11); or b) fecal microbiota from patients with presence of Gokushovirus WZ-2015a (>5 Count, n = 11). Booster inoculations were given twice weekly throughout the study to reinforce donor microbiota phenotype for 4 weeks.

[0265] 1.10. Fecal Viral Transplantation (FVT) experiment

[0266] Faecal virus extraction

[0267] Faecal viral transplants were prepared using faecal samples collected from a total of 11 donors. Thawed faecal material was diluted with sterile saline (previously filtered with a 0.200 pm PES filter) to 0.15-0.25g / ml and homogenized, centrifuged at 6000 x g for 30 min at 4 °C. Supernatant was filtered through a 0.70 pm filter and subsequently centrifugated again at 6000 g for 30 min at 4°C. Supernatant was then filtered through a 0.45 pm PES filter to remove bacteria (Millex Syringe-driven Filter Unit 33mm, REF: SLHVM33RS). The bacteria-depleted faecal filtrate was then concentrated using an Amicon Ultra-15 centrifugal filter (REF: UFC910024, Millipore Sigma), which has a pore size small enough to capture all known VLPs (~3 nm; Bonilla et al., 2016), Centrifuge at 5000 x g for 10-15 minutes at 4°C until a volume of 3.4 ml of concentrated sample is obtained. 1 ml of 50% Glycerol (previously filtered with a 0.200 pm PES filter) were added to the 3.4 ml of sample and stored in aliquots and frozen at -80 °C for posterior administration to mice. The purity and virus-like particle concentration was assessed by SYBR gold staining and epifluorescence microscopy, where bacterial cells are easily distinguished by size and fluorescence intensity (Backhed, F., Manchester, J. K., Semenkovich, C. F. & Gordon, J. I. Mechanisms underlying the resistance to diet-induced obesity in germ-free mice. Proc Natl Acad Sci U S A 104, 979-984 (2007).).

[0268] Experimental design FVT

[0269] Thirty-one wild-type C57BL / 6J male mice were used, animals were divided into 3 groups: a) Vehicle (n=11); b) FVT from donors with absence of Gokushovirus l / l / Z- 2015a (<10 counts, n=7); and c) FMT from donors with Gokushovirus WZ-2015A (>10 counts, n=13). FVT groups were colonized via daily oral gavage of donor microbiota (200 pL) for 3 days. Booster inoculations were given twice weekly throughout the study to reinforce donor microbiota phenotype. Control animals were subjected to the same protocol but instead of receiving donor microbiota they received oral gavage of 200 pL of PBS. Animals were subjected to an operant behavior procedure (Figure 7 h,i).

[0270] Operant behavior apparatus

[0271] Operant responding maintained by chocolate-flavored pellets was performed in mouse operant chambers (Model ENV-307A-CT, Med Associates, Georgia, VT, USA). The chambers were equipped with two retractable levers, one randomly assigned as the active lever and the other as the inactive for the entire experimental protocol. Pressing on the active lever resulted in a food pellet delivery paired with a stimulus-light (cue-light) located above the active lever, whereas pressing on the inactive lever had no consequences. A food dispenser equidistant between the two levers allows the delivery of food pellets when pertinent. The operant chambers were made of aluminium and acrylic and were housed inside soundproof boxes equipped with fans to provide ventilation and white noise. The chambers' floor consists of a metal sheet with holes. The floor was changed in the shock test sessions by a grid floor made of metal bars able to conduct electrical current, which was also used as a contextual cue for the aversive cue reactivity test the day after the shock test allowing mice to discriminate between different contexts.

[0272] Food pellets

[0273] During the operant conditioning sessions, animals received after pressing the active lever a 20 mg chocolate-flavored pellet consisting in a highly palatable isocaloric food (TestDiet, Richmond, IN, USA). These pellets had a similar caloric value (3.44 kcal / g: 20.6% protein, 12.7% fat, 66.7% carbohydrate) to the standard maintenance diet provided to mice in their home cage (3.52 kcal / g: 17.5% protein, 7.5% fat, 75% carbohydrate) with some slight differences in their composition: chocolate flavor (2% pure unsweetened cocoa) and enhanced sucrose content (8.3% standard diet food vs 50.1 % highly palatable pellets). These pellets were presented only during the operant behavior sessions, and animals were maintained on standard chow for their daily food intake. Operant training

[0274] A total of 23 mice were trained for 32 sessions to obtain chocolate-flavored pellets. In the operant conditioning sessions, mice were under an FR1 schedule of reinforcement for 7 days (1 lever-press resulted in 1 pellet delivery) followed by 25 days of FR5 (5 lever-presses resulted in 1 pellet delivery) (Figure 7a). The beginning of each session was signaled by turning on a house light placed on the chamber's ceiling during the first 3 s. Daily operant training sessions maintained by chocolate-flavored pellets lasted 1 h and were composed of 2 pellet periods (25 min each) separated by a pellet- free period (10 min). During the pellet periods, pellets were delivered contingently after an active response paired with a stimulus light (cue light). A time-out period of 10 s was established after each pellet delivery, where the cue light was off, and no reinforcer was provided after responding on the active lever. Responses on the active and inactive lever performed during the time-out periods were recorded. In contrast, the pellet-free period was signaled by the illumination of the entire operant box, and no pellet was delivered after responding on any lever. Mice were returned to their home cages after each session. All the mice subjected to operant behavior were male, two months old and weighted 30 ± 3 g at the beginning of the experiment. The male sex was chosen considering the previous literature that has validated the operant food addiction model only in male, but not in female mice (Domingo-Rodriguez, L. et al. A specific prelimbic-nucleus accumbens pathway controls resilience versus vulnerability to food addiction. Nature Communications 2020 11:1 11 , 1-16 (2020)).

[0275] As previously described (Martin-Garcia, E. et al. New operant model of reinstatement of food-seeking behavior in mice. Psychopharmacology (Berl) 215, 49-70 (2011)), the operant response was acquired when all the following conditions were achieved: (1) mice maintained a stable response with less than 20% deviation from the mean of the total number of reinforcers earned in 3 consecutive sessions (80% of stability); (2) at least 70% responding on the active lever; and (3) a minimum of 10 reinforcers per session.

[0276] The food addiction criteria were evaluated at the end of the protocol. The food addiction criteria gathered the main hallmarks of addiction based on DSM-IV (Deroche-Gamonet, V., Belin, D. & Piazza, P. V. Evidence for addiction-like behavior in the rat. Science 305, 1014-7 (2004)), DSM-5 and now included in the food addiction diagnosis through the YFAS 2.0. ( Gearhardt, A. N., Corbin, W. R. & Brownell, K. D. Development of the Yale Food Addiction Scale Version 2.0. Psychology of Addictive Behaviors 30, 113-121 (2016)).

[0277] Persistence of response (FR5 sessions 25-27)

[0278] Persistent desire or unsuccessful efforts to cut down displayed by continuous foodseeking behavior even if the food reward is signaled as not available. It is measured by the number of non-reinforced active responses during the pellet-free period (10 min) on the 3 consecutive days before the progressive ratio (PR).

[0279] Motivation (after FR5 session 27)

[0280] Considerable effort and time spent in obtaining the reward measured by the PR schedule of reinforcement. The response required to earn one single pellet escalated according to the following series: 1 , 5, 12, 21 , 33, 51 , 75, 90, 120, 155, 180, 225, 260, 300, 350, 410, 465, 540, 630, 730, 850, 1000, 1200, 1500, 1800, 2100, 2400, 2700, 3000, 3400, 3800, 4200, 4600, 5000, and 5500. The maximal number of responses that the animal is willing to perform to obtain one pellet is referred to as the breaking point. The maximum duration of the PR session was 5 h or until mice did not respond on any lever during 1 h.

[0281] Compulsivity (after FR5 session 31)

[0282] Continued use despite negative consequences evaluated as the resistance to punishment when chocolate-flavored pellets intake is coupled with an aversive stimulus. Mice were placed in an operant box without the metal sheet with holes and consequently with the grid floor exposed (contextual cue). During this session, mice underwent an FR5 schedule in which they received an electric foot-shock (0.18 mA, 2 seconds) after 4 responses and received another electric foot-shock (0.18 mA, 2 seconds) and a pellet paired with the cue light after the 5th response. The schedule was reinitiated after a time-out period (10 seconds after pellet delivery) and after the fourth response if mice did not perform the fifth response within 60 seconds. The total number of shocks performed in 50 minutes was used to evaluate compulsivity-like behavior, previously described as resistance to punishment. Four additional phenotypic traits were also evaluated as factors of vulnerability to addiction in each period:

[0283] Impulsivity (FR5 sessions 25-27)

[0284] The inability to stop a response once it is initiated was measured as the number of non-reinforced active responses during the time-out period (10 s) after each pellet delivery. This impulsivity-like behavior was delimited to the 3 consecutive days before the progressive ratio.

[0285] Cognitive inflexibility (after FR5 session 16)

[0286] Cognitive inflexibility is defined as the incapacity to shift responding to stimuli that have previously predicted the availability of reward. It is measured by the ability to modify the operant behavior when the active and the inactive levers were reversed in a single training session without previous learning. The errors are the number of active-reversed responses (previous inactive lever in a typical training session) performed in 1 h.

[0287] Appetitive cue-reactivity (after FR5 session 20)

[0288] The cue-induced food-seeking test, which consisted of a 90-min session, assessed the conditioning to an appetitive stimulus (cue-light). In the first 60 min, all active and inactive lever-presses were recorded but produced no consequences. In the next 30 min, the cue light associated with pellet delivery during a typical operant training session was illuminated with no contingent pellet reinforcement. To signal the change in the schedule, the cue light was presented twice non-contingently and for 4 s.

[0289] Aversive cue-reactivity (after FR 5 session 31 and shock)

[0290] To study the conditioning to an aversive stimulus (grid floor), non-reinforced active responses during the following session after the shock test were measured. Mice were placed in the operant box for 1 h with the same grid floor used during the shock test. However, during this session, pressing the active lever had no consequences: no shock, no chocolate-flavored pellets, and no cue light.

[0291] Statistical analysis of behavioral data IBM SPSS 19 (SPSS Inc., Chicago, USA) was used to analyze all the data. Normality was determined by Kolmogorov-Smirnov’s test. Parametric tests were performed if normality criteria were met. Repeated measures ANOVA was used to test the evolution over time. One-way ANOVA or two-way ANOVA were used when required for comparisons between groups followed by subsequent post hoc analysis (DMS test) when required. Non-parametric tests were performed if normality criteria were not met. Kruskal Wallis or Friedman test was applied followed by Mann-Whitney’s U test when required. Chi-square analyses were performed to compare the percentage of addicted and nonaddicted mice, considering the observed frequencies with those obtained in the control WT group. Results were expressed as individual values with the median and the interquartile range. A probability of 0.05 or less was considered statistically significant.

[0292] 1.11. AA supplementation experiment

[0293] Mice C57BI / 6J wild-type (WT) mice (n=23) were purchased from Charles River (France). Mice were supplemented in drinkable water with Anthranilic acid 0.3 mg / ml (n=12) or water (n=11). Animals were subjected to an operant behavior procedure as described in the FVT experiment.

[0294] 1.12. Sample Preparation for gene expression in mouse mPFC and metabolomics of Nucleus accumbens and Dorsal Striatum

[0295] The mice brains were quickly removed and the medial nucleus accumbens, dorsal striatum, and prefrontal cortex (mPFC) was dissected according to the atlas of stereotaxic coordinates of mouse brain (Paxinos, G. & Franklin, K. B. J. The Mouse Brain in Stereotaxic Coordinates. (Academic Press, San Diego, 1997)), and directly frozen in dry ice and stored at -80 °C.

[0296] 1.13. RNA-sequencing of mice brains

[0297] RNA Quality Control, library preparation, and sequencing of mice brains (FMT experiment 2) Quality control of the RNA was performed using the RNA 6000 Nano chip (Agilent) on an Agilent Bioalyzer 2100 obtaining RIN values between 8.7 - 9.8. Libraries were prepared from 500 ng of total RNA using the TruSeq stranded mRNA library preparation kit (Illumina, #20020594) with TruSeq RNA Single Indexes (Illumina, #20020492 and #20020493) according to the manufacturer’s instructions reducing the RNA fragmentation time to 4.5 min. Prepared libraries were analysed on a DNA 1000 chip on the Bioanalyzer and quantified using the KAPA Library Quantification Kit (Roche, #07960204001) on an ABI 7900HT qPCR instrument (Applied Biosystems). Sequencing was performed with 2x50 bp paired-end reads on a HiSeq 2500 (Illumina) using HiSeq v4 sequencing chemistry. Raw sequencing reads in the fastq files were mapped with STAR version 2.5.3a to the Gencode release 17 based on the GRCm38.p6 reference genome and the corresponding GTF file. The table of counts was obtained with the FeatureCounts function in the package subread, version 1.5.1.

[0298] RNA Quality Control, library preparation, and sequencing of mice brains (FMT experiment 1 and AA experiment)

[0299] Total RNA concentration was calculated by Quant-IT RiboGreen (Invitrogen, #R11490). To assess the integrity of the total RNA, samples are run on the TapeStation RNA screentape (Agilent, #5067-5576). Only high-quality RNA preparations, with RIN greater than 7.0, were used for RNA library construction. A library was independently prepared with 1 ug of total RNA for each sample by Illumina TruSeq Stranded mRNA Sample Prep Kit (Illumina, Inc., San Diego, CA, USA, #RS- 122-2101). The first step in the workflow involves purifying the poly-A containing mRNA molecules using poly-T-attached magnetic beads. Following purification, the mRNA is fragmented into small pieces using divalent cations under elevated temperature. The cleaved RNA fragments are copied into first strand cDNA using SuperScript II reverse transcriptase (Invitrogen, #18064014) and random primers. This is followed by second strand cDNA synthesis using DNA Polymerase I, RNase H and dUTP. These cDNA fragments then go through an end repair process, the addition of a single ‘A’ base, and then ligation of the adapters. The products are then purified and enriched with PCR to create the final cDNA library. The libraries were quantified using KAPA Library Quantificatoin kits for Illumina Sequecing platforms according to the qPCR Quantification Protocol Guide (KAPA BIOSYSTEMS, #KK4854) and qualified using the TapeStation D1000 ScreenTape (Agilent Technologies, # 5067-5582). Indexed libraries were then submitted to an Illumina NovaSeq (Illumina, Inc., San Diego, CA, USA), and the paired-end (2x100 bp) sequencing was performed by the Macrogen Incorporated.

[0300] 1.14. Global untargeted metabolic profiling of mice nucleous accumbens and dorsal striatum

[0301] First, samples were randomized. To extract metabolites from tissue, samples were transferred in plastic tubes with 6 ceramic beads and weighed. 500 pL LC-MS grade MeOH / water (4:1 v / v) was then added and homogenized at 5 m / s for 15 seconds. Homogenates were then incubated at -20°C for 10 minutes and centrifuged at 15,000 g for 10 minutes. For each sample, all supernatant was transferred to a new vial and dried. Sample extracts were re-suspended in 30 pL of LC-MS grade water. For each sample, the total concentrations were determined by NovaMT Sample Normalization kit and adjusted to a concentration of 8 mM with water, vortexed and then centrifuged at 15,000 g for 1 min. For preparing of pooled sample, equal amount was taken from each individual sample to prepare the pooled sample, which was used as the reference. Chemical isotope labelling and LC-MS Analysis Condition was performed as described above for humans. Metabolite identification was performed using IsoMS Pro 1.2.20 (NovaMT Inc.) and NovaMT Metabolite Database v3.0. After Data Quality Check, Data Processing was performed. Peak pairs without data present in at least 80.0% of samples in any group were filtered out. Data were normalized by Ratio of Total Useful Signal.

[0302] 1.15. Drosophila methods

[0303] Fly stocks and maintenance

[0304] Flies were raised and maintained on standard diet (SD) (1 L: yeast 27,5 g, yellow cornmeal 52 g, sugar 110 g, agar 5 g, propionic acid 5 ml, 0,1 % methylparaben in ethanol 5 ml). Conditional RNAi lines targeting Dop2R the Drosophila orthologue of (Drd3) (stock 11470, referred as Dop2R -RNAi)-, genetic background controls (60100 and 60000 referred as Control 1 and Control, respectively) and UAS-Dcr2 (60009) were obtained from the Vienna Drosophila RNAi Centre (VDRC, Vienna, Austria). The following genetic strains were obtained from the Bloomington stocks center (Indiana University): w; ple-Gal4 (BL8848), Dop2R overexpression line (UAS-Dop2 ?), stock BL86134, a line to expresses the light chain of tetanus toxin (UAS-TNT, BL28838) to inhibit synaptic transmission and it corresponding control (UAS-impTNT) that expresses a mutated form of the tetanus toxin light chain gene (BL28840). UAS- Dop2R was subsequently isomerized for 6 generations into the vdrc60100 (Control 1) genetic background. The Drosophila wild-type (l / l / f) strain used in this study was originally obtained from Bestgene. The final stock was created by exchanging the w allele of the strain that the company regularly uses to inject P-element-based transgenes by a w+allele136. UAS-TNT and UAS-impTNT were subsequently isomerized for 6 generations into l / l / f strain genetic background. Pan neuronal RNAi expression was induced by w; UAS-Dcr-2 E / av-GAL4 promoter line at 28°C. All flies were maintained in a 12:12 hours light dark cycle. All behavioral experiments were performed at 25 °C.

[0305] Dietary supplementation with anthranilic acid

[0306] Standard diet (1 L: yeast 27,5 g, yellow cornmeal 52 g, sugar 110 g, agar 5 g) was generated avoiding the addition of preservatives propionic acid and the methylparaben, to prevent possible interactions with the metabolites. The food was let to cool down util 45 °C and was supplemented with 200 mg / L of Anthranilic acid (AA) (A89855, Sigma). Flies with dietary supplementation AA were reared at 25 °C.

[0307] Excreta quantification (EX-Q)

[0308] Excreta quantification was adapted from previously described (Wu, Q. et al. Excreta Quantification (EX-Q) for Longitudinal Measurements of Food Intake in Drosophila. iScience 23, 100776 (2020)). SD containing 1 % w / v Blue 1 dye (861146, Sigma) was poured into plastic caps (ZE 4,5 mm). 15 three-to-five-day old mated females were placed into 50 mL centrifuge Falcon Tubes (ZE 27 mm) with several air holes to enable flies to breathe. As a unique food source, a cap with dyed media was placed on each tube and these were placed in the incubator at 25°C in a humid chamber for 48 hours. During this period flies consumed the food containing the non-absorbable dye and excreted dyed waste products inside the vial. The dye excreted was collected by adding 2 ml of water into the vials and vortexing until all excrements were completely dissolved, subsequently 90 ml of each sample were placed in a 96 well plate and absorbance was determined in a spectrophotometer (BioTek cytation 5 imaging reader) at 630 nm. Tubes where flies died during the course of the experiment were excluded from the analysis. The concentration of Blue 1 in each sample was calculated from a standard curve made by serial dilutions dye in water. Water was considered as the baseline blank.

[0309] Dye Uptake

[0310] Dye intake was adapted from previously described (Shell, B. C. et al. Measurement of solid food intake in Drosophila via consumption-excretion of a dye tracer. Sci Rep 8, 11536 (2018)). Three-to-five-day old mated females were collected and starved for 40 h in a vial on a wet Kimwipe paper. Groups of 50 females were transferred onto fly narrow vials (ZE 25 mm) containing SD with 1% w / v of Blue 1 dye (861146, Sigma). After 30 min flies were frozen, distributed in microcentrifuge tubes (7 flies / tube) and homogenized in 200 mL of distilled water. Homogenates were centrifugated during 4 min at 13.000 rpm and supernatant was transferred to a new tube, this process was repeated two consecutive times to ensure the complete removal of debris and lipids. Finally, 90 ml of each sample were transferred to a 96 well plate and the absorbance was measured at 630 nm in the spectrophotometer (BioTek cytation 5 imaging reader). The concentration of Blue 1 in each homogenate was calculated from a standard curve made by serial dilutions in water of a sample of dyed food. The supernatant of flies fed the non-dyed medium was used as a blank for spectrophotometric analysis for each condition.

[0311] Calculation of the amount of dye ingested

[0312] The amount of dye ingested per fly by study time (typically 30 min in the dye ingestion, 48h in the ExQ) was calculated by multiplying the interpolated concentration (pg / mL) by the final volume and dividing by the number of flies. Mean and standard deviation of each group was calculated taking into account all the replicates. A minimum of 3 independent experimental replicates were performed per condition. For each experiment data was standardized against the corresponding control (SD=1 and a mean=0) in order to summarize and plot together the results of multiple experimental replicates.

[0313] Body weight determination For measuring the body weight of flies, 10-day old mated females were anesthetized and weighed immediately by using an analytical balance (0.1 mg accuracy).

[0314] The Capillary Feeder (CAFE) assay to measures food intake and food preference The CAFE assay protocol utilized in this study was adapted from a previously described method (Ja, W. W. et al. Prandiology of Drosophila and the CAFE assay. Proc Natl Acad Sci U SA 104, 8253-8256 (2007)) . CAFE chambers were constructed of plastic vials (2,8 cm diameter, 11 ,4 cm length) Each chamber was equipped with small holes to facilitate air exchange, a piece of wet Whatman filter paper was placed, inside each vial, to maintain humidity levels. Four 5 pl capillaries (2-000-005; Drummond Scientific) filled with liquid medium were inserted into the chamber plug through adaptors made of truncated pipette tips. For choice experiments, two capillaries were labelled and contained a solution of 5% sucrose, while the other two contained a solution of 5% sucrose + 15% ethanol. Capillaries were filled by capillarity and were measured and replaced every day for 4 consecutive days. The chambers were placed within an outer chamber covered with wet paper and surrounded by 10 conical tubes filled with 30 ml of water to ensure constant humidity conditions. The assay was conducted at 25°C.

[0315] Fly PAD assays

[0316] UAS-TNT and UAS-imTNT were crossed with Th_Gal4 driver line in SD supplemented with 200mg / L AA or the vehicle, at 19°C. 3-5 day old flies (male+female) were collected and put in starvation in a media containing only 0,6% agar supplemented with the corresponding amount of AA or vehicle. After 24-h, flies were trained for 2h in vials where they were allowed to feed by choosing between two different kinds of food: 0,6% agar + 5% sucrose or 0,6% agar + 5% sucrose + 15% ethanol, both containing the corresponding amount of AA or vehicle. Training was performed once a day for 3 consecutive days. On the third day, flies were anesthetized before training and females were separated. The fourth day, female flies were tested in the food preference experiments performed using the FlyPAD, as previously described (PMIDXXX). Briefly, trained female flies were set in individual arenas with 2 food patches containing 1 % agarose mixed with either 5% sucrose or 5% sucrose + 15% ethanol. After introducing flies to the fly pad arenas these were allowed to feed for 40 min. FlyPAD data were acquired using the Bonsai framework (version 2.8.0), and analysed in MATLAB (version 8.2.0.701) using custom-written software. Noneating flies (those flies having less than two activity bouts per assay) were excluded from the analysis were the number of syps and the preference index was calculated.

[0317] Statistical analyses in Drosophila

[0318] Numerical data is presented as mean + / -SEM. Significance was calculated using One-way ANOVA combined with Dunnett’s test for multiple comparisons when more than two conditions are presented in one graph. T-test was performed whenever two conditions were compared to each other. For the buoyancy assay we performed a Two-way ANOVA with Sidak correction for multiple comparison. Statistics calculations were performed using the Prism-GraphPad software (Version 9.3.1).

[0319] 1.16. Statistical analyses

[0320] First, normal distribution and homogeneity of variances were tested. Results are expressed as number and frequencies for categorical variables, mean and standard deviation (SD) for normal distributed continuous variables and median and interquartile range [IQ] for non-normal distributed continuous variables. To determine differences between study groups (obesity, Microviridae or Gokushovirus WZ-2015a quartiles or quintiles) a Kruskal-Wallis test, followed by a Wilcoxon test for pairwise comparisons. Non-parametric monotonic trends were assessed by the Mann-Kendall trend test. Partial Spearman’s correlation analysis was used to determine the correlation between iqlr-transformed Microviridae or Gokushovirus WZ-2015a levels and neuropsychological tests. After controlling for covariates, plots were generated with the ranked residuals of the model after adjusting for selected covariates. These statistical analyses were performed with SPSS, version 19 (SPSS, Inc, Chicago, IL) or R. Statistics can be found in the figures and legends.

[0321] Metagenomics statistical analysis

[0322] Differential abundance (DA) analyses for bacterial families and species associated with the YFAS, STROOP-CW, and UPPS Positive urgency were performed using both the analysis of compositions of microbiomes with bias correction (ANCOM-BC) methodology , fastANCOM , and ZicoSeq . The former considers the bias due to differential sampling fractions across samples by adding a sample-specific offset to a linear regression model, that is estimated from the observed data. The linear regression model in log scale is analogous to log-ratio transformation to consider the compositional nature of metagenomics datasets, while the offset term serves as the bias correction. On the other hand, ANCOM is an additive log-ratio (air) based methodology that performs all possible DA analyses by successively using each taxon as a reference taxon. For K microbes it requires fitting K(K - 1) / 2 models for logratios of counts. Therefore, it is computationally intensive, since for each taxon, it performs air transformation using all remaining taxa. Conversely, fastANCOM is a fast implementation of ANCOM that fits only K models for log-transformed counts. For each taxon, the number of rejections, denoted by W / , is counted, and ANCOM makes use of the empirical distribution ofW to determine the cut-off value of significant taxon. The 70th percentile of W distribution was used as the cut-off, as ANCOM has shown to successfully control the FDR under the nominal level (5%) while maintaining adequate power. Finally, ZicoSeq uses a power transformation function g(x;p)=xp, which is similar to the Box-Cox transformation, as the actual relationship between the taxa abundance and the covariates in real scenarios can be more complex than the log relationship. In our case, we used a root square transformation. ZicoSeq assesses statistical significance by permutation, which depends on fewer assumptions and is more robust to model misspecifications. In addition, it infers the underlying true proportions using an empirical Bayes approach with an informative beta mixture prior, instead of using an uninformative prior or a beta prior. We adjusted all models in humans for age, sex, BMI, and country. P-values were adjusted for multiple comparisons using a Sequential Goodness of Fit as implemented in the “SGoF” R package. Unlike FDR methods, which decrease their statistical power as the number of tests increases, SGoF methods increase their power with an increasing number of tests. SGoF has proven to behave particularly better than FDR methods with a high number of tests and low sample size, which is the case of metagenomics large datasets. Statistical significance was set at Padj<0.1. Microbial functions associated with the iqlr-transformed Gokushovirus WZ-2015a were also analysed with fastANCOM. Pathway over-representation analyses were performed mapping associated KEGG orthologs to the KEGG pathways using the “enrichKEGG” function from the “ClusterProflier” R package . Pathway significance was assessed using a hypergeometric test and a Storey procedure (q-values) was applied for multiple testing correction. Machine learning analysis (radiomics, targeted tryptophan metabolomics)

[0323] Targeted tryptophan metabolomics data were first normalized using a probabilistic quotient normalisation. Then radiomics and tryptophan metabolomics data were analysed using machine learning (ML) methods. After adjusting for age, BMI, sex, and education years, we adopted an all-relevant ML variable selection strategy applying a multiple random forest (RF)-based method as implemented in the Boruta algorithm42. It performs feature selection in four steps: a) Randomization, which is based on creating a duplicate copy of the original features randomly permuted across the observations; b) Model building, based on RF with the extended data set to compute the normalized permutation variable importance (VIM) scores; c) Statistical testing, to find those relevant features with a VIM higher than the best randomly permuted variable using a Bonferroni corrected two-tailed binomial test; and d) Iteration, until the status of all features is decided. We run the Boruta algorithm with 500 iterations, a confidence level cut-off of 0.005 for the Bonferroni adjusted p-values, 5000 trees to grow the forest (ntree), and a number of features randomly sampled at each split given by the rounded down number of features / 3 (the mtry recommended for regression). VIM obtained from random forest models do not provide the sign of the association with the response variable. Therefore, to facilitate the interpretation of the models, the contribution of each metabolites for the prediction of the response variable was determined by the exact computation of SHapley Additive exPlanations (SHAP) scores by leveraging the internal structure of the random forests models. The R packages “treeshap” and “SHAPforXGBoost” were used to calculate and plot the SHAP scores.

[0324] RNA-seq analysis

[0325] Differential expression gene analyses were performed on gene counts using the “limma” R package . First, low expressed genes were filtered, so that only gene with more than 10 reads in at least 2 samples were selected. RNA-seq data were then normalized for RNA composition using the trimmed mean of M-value (TMM) as implemented in edgeR package144. Normalized counts were then converted to Iog2 count per million (logCPM) with associated precision weights to account for variations in precision between different observations using the “voom” function with donor’s age, BMI, and sex as covariates. A robust linear regression model adjusted the previous covariates was then fitted to the data using the “ImFit” function with the option method = “robust”, to limit the influence of outlying samples. Finally, an empirical Bayes method was applied to borrow information between genes with the “eBayes” function. P-values were adjusted for multiple comparisons using the Benjamini- Hochberg procedure for False Discovery Rate (pFDR). Differentially expressed genes were mapped to the Search Tool for Retrieval of Interacting Proteins / Genes (STRING) database (which integrates known and predicted protein / gene interactions) to predict functional gene-gene interaction networks. Then, functional local clusters in the interaction network were determined using a Markov Cluster algorithm (MCL) with an inflation parameter = 2. Active interacting sources including text mining, experiments, databases, co-expression, and co-occurrence and an interaction score > 0.15 were used to construct the interaction networks. The functional roles of differentially expressed genes were characterized using over-representation analyses based on the Gene Ontologoy (GO) using ClusterProfiler and KEGG database using ConsensusPathDB . Pathway significance was assessed using a hypergeometric test and a Storey procedure (q-values) was applied for multiple testing correction.

[0326] Global untargeted metabolomics profiling

[0327] For the analysis of global metabolic profiling from plasma (humans), nucleus accumbens and dorsal striatum (FMT and AA supplementation mice experiments), data was normalized by ratio of total useful signal. Then, robust linear regression models were adjusted controlling for age, bmi, sex, and education years in the case of humans and donors of the FMT experiment as described in the RNAseq analysis. P-values were adjusted for multiple comparisons using a Sequential Goodness of Fit as implemented in the “SGoF” R package. Statistical significance was set at Padj<0.1.

[0328] 2. Results

[0329] 2.1. Microviridae are associated with food addiction and obesity

[0330] We first assessed the relationships of bacterial composition with food addiction, diagnosed using the YFAS 2.0, in a discovery cohort (IRONMET-CGM, n = 88, Figure 1). The prevalence of food addiction has recently been estimated at 20% and has shown to vary from 7.9 to 25% in patients without obesity, 24-45% in patients with obesity, and reached rates up to 60% in patients with morbid obesity (Oliveira, J., Colombarolli, M. S. & Cordas, T. A. Prevalence and correlates of food addiction: Systematic review of studies with the YFAS 2.0. Obes Res Clin Pract 15, 191-204 (2021); Praxedes, D. R. S. et al. Prevalence of food addiction determined by the Yale Food Addiction Scale and associated factors: A systematic review with meta-analysis. Eur Eat Disord Rev 30, 85-95 (2022)).

[0331] Thus, a positive association with BMI has been reported. In line with these results, we found that patients with obesity had higher YFAS scores than lean subjects (Figure 2a). The three behavioural hallmarks of addiction: motivation for the drug (Figure 2b), persistence to response (Figure 2c), and compulsivity defined as an alteration of inhibitory control despite negative consequences (Figure 2d), were also positively associated with the BMI.

[0332] We next analysed the associations of the gut microbiome with the food addiction scores. We applied three different methods that take into account the underlying compositional structure of the microbiome data and use different transformations to model the relationship between taxa abundances and the covariates such a log and power transformations: ANCOM-BC (Lin, H. & Peddada, S. das. Analysis of compositions of microbiomes with bias correction. Nat Commun 11 , 3514 (2020)) fastANCOM (Zhou, C., Wang, H., Zhao, H. & Wang, T. fastANCOM: a fast method for analysis of compositions of microbiomes. Bioinformatics 38, 2039-2041 (2022);) and ZicoSeq (Yang, L. & Chen, J. A comprehensive evaluation of microbial differential abundance analysis methods: current status and potential solutions. Microbiome 10, 130 (2022)). We adjusted all models for age, sex, BMI, and education years. Using all methodologies, we found a strong positive association (the highest positive fold change) between bacteriophages from the Microviridae family and the YFAS scores (Figure 2e,f, Figure 3a). In fact, using the ZicoSeq methodology, Microviridae was the only microbial family positively associated with food addiction (Figure 3b). Microviridae bacteriophages have been recently identified as the most stable colonizers in the human gut and a transition from a chow to a high-fat diet in mice has shown to significantly increase the relative abundance of bacteriophages from the Microviridae family along with a decrease in Siphoviridae (Schulfer, A. et al. Fecal Viral Community Responses to High-Fat Diet in Mice. mSphere 5, e00833-19 (2020)). As sexual dimorphism can shape the gut microbiome, we also conducted sex- stratified analysis. Using both ANCOM-BC and ZicoSeq, we found that Microviridae were positively associated with food addiction in women (n=61 , Figure 3c, f), but not in men (n=27, Figure 3d,g). Importantly, we were able to replicate these findings in a subset of patients from the IRONMET cohort that underwent food addiction tests using the same YFAS questionnaire (n=29) (Figure 2f). Microviridae was the only microbial family consistently positively associated with food addiction. Additionally, Microviridae was amongst the microbial families most prominently associated with YFAS in women with obesity in this cohort (Figure 3e).

[0333] As Microviridae had by the most consistent association with food addiction, we further analysed the associations of the gut microbiota with the three addiction-like criteria: motivation, persistence to response, and compulsivity, using ANCOM-BC. Remarkably, Microviridae was the microbial family most strongly and positively associated with both motivation (Figure 19a) and persistence to response (Figure 19b), but not with compulsivity (Figure 19c). Microviridae were mainly associated with motivation and persistence (although not significant, Padj=0.15) in women (Figure 19d-f), whereas we did not observe any significant association in men (Figure 19g- i). Additionally, we also analysed the associations of the inter quartile log-ratio (iqlr) transformed Microviridae abundances with these 3 addiction-like criteria as well as other neuropsychological tests related to impaired control shared by overeating and substance abuse disorders such as impulsivity and reward sensitivity. Consistent with the ANCOM-BC results, we found a consistent positive association with motivation and persistence to response (Figure 2g, h), but not with compulsivity (Figure 2i), after controlling for age, sex, BMI, and education years. Moreover, the iqlr-transformed Microviridae levels were associated with the heightened sensitivity to reward and punishment (Figure 2j), which has been associated with both substance use and overeating (Adams, R. C., Sedgmond, J., Maizey, L., Chambers, C. D. & Lawrence, N. S. Food Addiction: Implications for the Diagnosis and Treatment of Overeating. Nutrients 11 , 2086 (2019)).

[0334] In light of the strong associations between Microviridae and food addiction, we searched for the most abundant species of Microviridae in the discovery cohort, which we found to be by far the Gokushovirus WZ-2015a. In line with the previous results, we found a significative positive association of the iqlr-transformed levels with the motivation (P=0.034, Figure 2I) scores, and a trend (only statistically significant at a=0.15, P<0.15) towards higher food addiction (P=0.077, Figure 2k), persistence to response (P=0.13, Figure 2m) and compulsivity (P=0.15, Figure 2n).. Again, we found significant associations with higher sensitivity to reward and punishment (P=0.006, Figure 2o), but also with higher (P=0.022, Figure 2p) assessed by the negative urgency (a reverse score item) in the LIPPS impulsive behavior scale. After stratifying by sex we observed a trend between iqlr-transformed abundances of Gokushovirus WZ-2015a and both YFAS and motivation scores (P<0.15), but also a positive association with compulsivity (P<0.05) and a trend towards higher persistence to response (P<0.1) in men (Figure 19j,k). In addition, the Gokushovirus WZ-2015a levels were strongly associated with higher sensitivity to reward and punishment (P<0.01) and impulsivity (P<0.05) in women (Figure 19j,k).

[0335] Gene and genomic analysis of assembled data from short sequences related to Gokushovirus identified five complete genomes closely related to other viruses from the Gokushovirinae subfamily. According to the genomic comparison and the obtained average genome nucleotide identity values among each other (<43%; significantly lower than 95% of identity, all these Gokushoviruses belong to different viral species and likely to different viral genera as well. Host assignment using CRISPR spacer-protospacer match identified Faecalibacterium prausnitzii, uncultured Faecalibacteirum spp., and a Coriobacteriia bacterium as the potential host for three of these Gokushovirus genomes, while the others putatively infected bacteria that could belong to Ruminococcaceae and Firmicutes (In agreement with our findings, previous CRISPR-based host prediction highlighted connections between Microviridae and highly abundant and persistent gut bacterial genus such Bacteroides and Faecalibacterium. In agreement with our results, a recent study found that patient with overeating disorders exhibited a marked loss of Faecalibacterium prausnitizii (Fan, S. et al. Microbiota-gut-brain axis drives overeating disorders. Cell Metab 35, 2011-2027. e7 (2023)). Moreover, colonization of mice with overeating disorders with F. prausnitzii significantly reduced overeating and excessive preference for high palatable foods. 2.2. Microviridae are associated with lower inhibitory control, obesity, and alterations in functional brain connectivity in a second validation cohort

[0336] Insufficient inhibitory control of the reward and the emotional-arousal networks by the executive control network has shown to play a critical role in the shift from the homeostatic to hedonic regulatory food intake (Gupta, A., Osadchiy, V. & Mayer, E. A. Brain-gut-microbiome interactions in obesity and food addiction. Nat Rev Gastroenterol Hepatol 17, 655-672 (2020)).

[0337] Improving executive function deficits has been suggested to mediate reward drive and craving, thereby restoring control over addictive behaviours. Therefore, we next sought to validate the observed associations of Microviridae and Gokushovirus \NZ- 2015a in a large-scale validation cohort (Figure 4a) including participants from the Aging Imageomics Study (aged >50 years) that underwent neuropsychological testing of executive function, specifically inhibitory control assessed with the Stroop Colour Word test (SCWT). We found a consistent positive association of both the iqlr- transformed Microviridae and Gokushovirus WZ-2015a levels with obesity status (Figure 4b, c). Remarkably, among all microbial families, Microviridae had the strongest negative association with inhibitory control in men (Figure 4d). Similar results were obtained using the ANCOM-BC methodology (Figure 5a). When we classified subjects according to both the iqlr-transformed Microviridae and Gokushovirus WZ-2015a quartiles, we found a significant decrease in inhibitory control with higher quartiles in all subjects (Figure 4e,f) and this difference was only revealed in men (Figure 4g, h).

[0338] The present century has witnessed a large increase of studies assessing neural correlates of behavioural disorders using both task-based and resting-state functional magnetic resonance imaging (fMRI). However, most of these studies have focused on static functional connectivity. To provide a better characterization of the changes associated with the presence of Gokushovirus WZ-2015a, we studied the dynamics of resting-state brain activity across the whole-brain functional network in a subset of patients applying the intrinsic-ignition framework (Escrichs, A. et al. Whole-Brain Dynamics in Aging: Disruptions in Functional Connectivity and the Role of the Rich Club. Cereb Cortex 31 , 2466-2481 (2021)). We next applied a machine learning variable selection strategy based on applying multiple random forests (Kursa, M. B. & Rudnicki, W. R. Feature selection with the boruta package. J Stat Softw 36, 1-13 (2010).), as implemented in the Boruta algorithm, to identify relevant brain networks combined with the exact computation of Shapley Additive exPlanations (SHAP) scores to facilitate model interpretation (Lundberg, S. M. et al. From Local Explanations to Global Understanding with Explainable Al for Trees. Nat Mach Intell 2, 56-67 (2020)).

[0339] The ignition measure reflects spatial diversity and the broadness of functional connectivity across the whole-brain network. We found that the iqlr-transformed Gokushovirus WZ-2015a levels were associated with higher intrinsic ignition in regions belonging to the medial frontal network (precentral; medial orbital gyrus; inferior frontal gyrus, parts triangularis and orbitalis; anterior cingulate cortex [ACC]), the subcortical network (caudate; middle frontal gyrus, orbital part; inferior frontal gyrus, part orbitalis), and the default mode network (medial orbital gyrus; superior frontal gyrus, medial orbital) (Figure 5b). In men, the iqlr-transformed Gokushovirus WZ-2015a were also associated with higher intrinsic ignition in regions belonging to the medial frontal (Precental; Inferior frontal gyrus, orbital part) and subcortical networks (Putamen; Inferior and Middle frontal gyrus, orbital part; Hippocampus and Thalamus) (Figure 4i).

[0340] Classical resting-state fMRI methods do not capture the variability over time across the whole-brain network, which can be measured by the metastability. We found that Gokushovirus levels were mainly associated with lower metastability in the medial frontal (Superior, Middle, and Inferior temporal gyrus), the subcortical (Middle cingulate cortex), and motor networks (Supplementary motor area, Middle cingulate cortex, Insula, Rolandic operculus) (Figure 5c). Similarly, they were also negatively associated with regions belonging to the subcortical network (Hippocampus, Amygdala, Anterior cingulate cortex, Superior frontal gyrus), the default network (Medial orbital gyrus; Superior frontal gyrus, medial orbital), and the motor network (Postcentral, Precentral, Middle cingulate cortex) in men (Figure 4j). Lower metastability reflects a more stable system and reduced flexible switching across time and has been associated with behavioural decline and damage to structural connectivity in traumatic brain injury. Notably most of the identified brain regions are involved in the extended reward network, which controls normal ingestive behaviour and comprises interconnecting brain networks such as: a) the salience network, which is associated with interoceptive stimuli reaching the brain; b) the cortical inhibitory control executive network, which exerts inhibitory control on both c) the reward network, which includes subcortical and cortical brain regions and is responsible for motivation and craving for reward; and d) the emotional-arousal network (Gupta, A., Osadchiy, V. & Mayer, E. A. Brain-gut-microbiome interactions in obesity and food addiction. Nat Rev Gastroenterol Hepatol 17, 655-672 (2020)). Remarkably, we found a large number of associations with orbital parts from the control executive network. The orbitofrontal cortex (OFC) contributes to impulsive choice, and functional connectivities of the OFC, especially with the ACC, have been positively associated with food reward and BMI (Rolls, E. T., Feng, R., Cheng, W. & Feng, J. Orbitofrontal Cortex Connectivity is Associated With Food Reward and Body Weight in Humans. Soc Cogn Affect Neurosci nsab083 (2021) doi:10.1093 / SCAN / NSAB083). A recent review or restingstate functional connectivity in obesity identified a consistent pattern of increased OFC connectivity and increased BOLD signal amplitude in subcortical regions (hippocampus, amygdala, putamen) in individuals with obesity compared to controls (Parsons, N., Steward, T., Clohesy, R., Almgren, H. & Duehlmeyer, L. A systematic review of resting-state functional connectivity in obesity: Refining current neurobiological frameworks and methodological considerations moving forward. Rev Endocr Metab Disord 23, 861-879 (2022)).

[0341] Our findings are also consistent with results from neuroimaging studies in drug addiction that reported increased connectivity in all extended reward networks as well as the self-directed network (composed of the precuneus and the posterior cingulate cortex) both under resting-state or after exposure to drug cues in individuals with addiction compared to healthy controls (Zilverstand, A., Huang, A. S., Alia-Klein, N. & Goldstein, R. Z. Neuroimaging Impaired Response Inhibition and Salience Attribution in Human Drug Addiction: A Systematic Review. Neuron 98, 886-903 (2018)). The most consistent findings have reported an hyperconnectivity of the reward network, specifically in the ACC, medial prefrontal cortex (PFC), middle frontal gyrus, OFC, striatum (putamen and caudate), accompanied by a significant increase of activation levels in the salience network. Resting-state studies on behavioural addiction have also revealed hyperconnectivity in the putamen, temporal and frontal lobes (Tolomeo, S. & Yu, R. Brain network dysfunctions in addiction: a meta-analysis of resting-state functional connectivity. Transl Psychiatry 12, 41 (2022)). Crucially, the dopaminergic signaling in the corpus striatum, which contains the caudate nucleus and the putamen (dorsal striatum, DS) and the nucleus accumbens (NAc, ventral striatum) plays a pivotal role in addiction and compulsive eating behavior (Gupta, A., Osadchiy, V. & Mayer, E. A. Brain-gut-microbiome interactions in obesity and food addiction. Nat Rev Gastroenterol Hepatol 17, 655-672 (2020)). Consistently, our findings also revealed an association between Gokushovirus WZ-2015a levels and Ignition within regions comprising 87% caudate in the whole cohort and 84% putamen in men.

[0342] 2.3. Food addiction and Gokushovirus WZ-2015a are associated with microbial functions and metabolites involved in the serotonin and dopamine metabolism

[0343] To identify microbial molecular functions associated with food addiction, we next performed functional analyses in the IRONMET-CGM cohort by mapping reads originating from microbial genes to the Kyoto Encyclopedia of Genes and Genomes (KEGG) orthologs and using the ANCOM-BC methodology, controlling for age, BMI, sex and years of education (Figure 6a). To facilitate interpretation of the results, we performed a gene set enrichment analysis (GSEA) to identify the relevant bacterial pathways involved in these associations. We identified the metabolism of the aromatic aminoacids (AAA) tryptophan and phenylalanine as the most significant bacterial pathways with the largest positive enrichment score associated with YFAS (Figure 6b). Among the microbial functions of the tryptophan metabolism (Figure 6c), the monoamine oxidase (MAO, K00274), which catalyzes the degradation of biogenic amines derived from tryptophan and tyrosine such as monoamine neurotramistters (e.g. dopamine, noradrenaline, adrenaline, serotonin) or other trace amines (tryptamine, 3- and 5-hydroxykynurenamine, tyramine) (Figure 6h), had one of the highest fold change, alongside AFMID (K01432) and katG (K03782) (Figure 6d), which are responsible for the formation of anthranilic acid (AA) from N- formylkynurenine and the degradation of 3-hydroxyanthranilic acid (3-HAA) derived from AA, respectively (Figure 6h). To gain better insight into the role of Gokushovirus WZ-2015a in food addiction, we performed a global untargeted metabolomics profiling of plasma samples from the IRONMET-CGM cohort by Chemical Isotope labelling LC-MS. Then, we fitted robust linear regression model adjusting for the previous covariates to identify metabolites associated with the presence (>10 counts) of Gokushovirus WZ-2015a (Figure 6e). Consistent with our metagenomics findings, a KEGG pathway over-representation analysis identified again the metabolism of tryptophan and phenylalanine (q- value<0.1) (Figure 6f). In agreement with these findings, inoculation of overeating disorder mice with Faecalibacterium prausnitzii, the host of Gokushovirus WZ-2015a, alleviated overeating symptoms through alterations in tryptophan metabolites. Remarkably, it also underscored pathways related to cocaine and amphetamine addiction. However, the most over-represented pathways included the tyrosine metabolism and the dopaminergic synapse through increases in tyrosine and the major degradation products of monoamines dopamine and norepinephrine (homovanillic acid and 3-methoxy-4-hydroxyphenylglycolaldehyde) in the presence of Gokushovirus WZ-2015a (Figure 6f,g). Tyrosine is the precursor of dopamine while homovanillic acid, which was among the most increased metabolites in the presence of Gokushovirus WZ-2015a (Figure 6e), is a major catecholamine metabolite of dopamine degradation through MAO (Figure 6h), a microbial gene that we found strongly upregulated with food addiction (Figure 6d). Importantly, plasma homovanillic acid has been proposed as a marker of brain dopamine activity, with evidence suggesting that it may reflect changes in brain dopamine metabolism Sternberg, D. E., Heninger, G. R. & Both, R. H. Plasma homovanillic acid as an index of brain dopamine metabolism: enhancement with debrisoquin. Life Sci 32, 2447- 2452 (1983). Similarly, 3-methoxy-4-hydroxyphenylglycolaldehyde is the principal MAO metabolite of epinephrine, which is produced from dopamine by dopamine b- hydroxylase (DBH). It is also worth noting that tryptophan metabolism modulates dopamine neurotransmission through serotonergic regulation (Alex, K. D. & Pehek, E. A. Pharmacologic mechanisms of serotonergic regulation of dopamine neurotransmission. Pharmacol T / 7er 113, 296-320 (2007). 2.4. Microbiota and faecal viral transplantation from human donors’ faeces rich in Gokushovirus WZ-2015a induces an addiction-like phenotype in recipient mice

[0344] To study the potential direct role of Gokushovirus WZ-2015a underpinning the susceptibility to developing food addiction-like behavior, we transferred the microbiota of human donors with (presence group) and without (absence group) Gokushovirus WZ-2015a to recipient mice that underwent an operant behavioral model of food addiction (Martin-Garcia, E., Domingo-Rodriguez, L. & Maldonado, R. An Operant Conditioning Model Combined with a Chemogenetic Approach to Study the Neurobiology of Food Addiction in Mice. Bio Protoc 10, e3777 (2020)) (Figure 7a). The operant chambers were equipped with two retractable levers: pressing the active lever resulted in a food pellet delivery, while pressing the inactive lever had no effect. We did not find differences in body weight (Figure 7b) or performance accuracy (percentage of correct responses) (Figure 7c) across sessions. However, after a training period of a fixed ratio 1 schedule of reinforcement (FR1 , one lever-press results in one pellet delivery) mice receiving microbiota from donors with Gokushovirus WZ-2015a pressed more the active lever than those without under a fixed ratio 5 (FR5, five lever-presses results in 1 pellet delivery), indicating a food- rewarded operant conditioning (Figure 7d).

[0345] Persistence to response is one of the hallmarks of food addiction. We evaluated the persistence to response by measuring the number of non-reinforced active lever presses during a 10-min pellet-free period (PFP), when no pellet was delivered and the box was illuminated and signalling the unavailability of pellet delivery. At the end of the FR5 schedule, the presence group had a larger number of non-reinforced active responses during the PFP, thereby indicating a higher persistence to response (Figure 7e). The advantage of our model lies in the ability to measure several phenotypic traits considered risk factors of vulnerability to addiction-like behavior. In particular, we measured cognitive flexibility with a reversal learning test at the end, when the levers were switched so that the previously rewarded lever was now unrewarded and vice versa. Mice in the presence group had a higher number of total inactive lever presses during the reversal test (Figure 7f), which is indicative of lower cognitive flexibility as they were not able to refrain from pressing the previously active lever. We also measured impulsivity, as the number of non-reinforced active lever responses during a 10 seconds time-out period after each pellet delivery when no pellet is provided. Mice in the presence group had higher impulsivity than those in the absence group, as indicated by the higher number of active lever presses during the FR5 time-out period (Figure 7g).

[0346] As the results from the FMT study may be influenced by multiple components of the gut microbiota, we also performed a faecal viral transplantation (FVT) of human donors with (presence group) and without (absence group) Gokushovirus WZ-2015a to recipient mice to further confirm its role in food addiction. Translational models of food addiction were created by training the mice in operant chambers under an FR1 schedule of reinforcement during 7 sessions followed by 25 sessions under FR5 to acquire an operant response maintained by chocolate-flavored pellets (Figure 7h,i) . Animals that received FVT form donors with Gokushovirus WZ-2015a showed an increase in motivation for chocolate-flavoured pellets and the persistence to response compared to mice receiving FVT from individuals without Gokushovirus WZ-2015a and the control group (Figure 7j,k). We did not find differences between groups in compulsive-like behavior (Figure 7I). Notably, these results are in line with our findings in humans, where both Microviridae and Gokushovirus WZ-2015a were positively associated with motivation and persistence to response but not compulsivity (Figure 2g-i,l-n). No differences between groups were found for impulsive-like behavior (Figure 7m). Phenotypic traits related to addiction vulnerability such as appetitive associative learning were also significantly higher in mice receiving FVT with the presence of Gokushovirus WZ-2015a (Figure 7n), however, no differences were observed in cognitive flexibility (Figure 7o) Gokushovirus WZ-2015a induces changes in tyrosine and tryptophan metabolism in the striatum and downregulates Drd2 in the mPFC of recipient mice.

[0347] 2.5. Gokushovirus WZ-2015a induces changes in tyrosine and tryptophan metabolism in the striatum and downregulates Drd2 in the mPFC of recipient mice

[0348] To investigate the potential mechanisms through which Gokushovirus WZ-2015a could exert its effects on food addiction, we performed a global metabolic profiling of the nucleus accumbens (NAc, the major component of the ventral striatum) and the dorsal striatum (DS) of recipient mice from the FMT experiment. We focused on these two brain structures due to their crucial role in the extended reward network and its dopaminergic input, which are central to addiction, reward, pleasure, and foodseeking behavior (Figure 8a). Dopamine neurons in the ventral tegmental area (VTA) and the substantia nigra (SN) predominantly project to the NAc and the DS, respectively. In fact, the reinforcing effects of drugs mostly depend on dopamine signaling in the NAc.

[0349] The presence of Gokushovirus WZ-2015a in the donor microbiota was strongly associated with increased concentrations of 3-ethymalic acid, pyridoxal, and 4,6- and 4,8-dihydroxyquinoline in the NAc of recipient mice (Figure 8b). Notably, pyridoxal, a form of vitamin B6, is an essential cofactor for the enzymatic conversion of L-DOPA to dopamine and 5-hydroxytryptophan to serotonin by and the aromatic L-amino acid decarboxylase (AADC) (Figure 6h). This aligns with serotonin being among the metabolites mostly reduced in the NAc of mice due to Gokushovirus WZ-2015a (Figure 8b). Additionally, 4,6- and 4,8-dihydroxyquinoline are two AA alkaloids due to the key role of AA in their production through the synthesis of tryptophan from chorismic acid (Figure 6h). Moreover, 4,8-dihydroxyquinoline is the product of 3- hydroxykynurenamine by MAO, which were both strongly and positively associated with Gokushovirus WZ-2015a in humans (Figure 6d,e). Consistent with these findings, a KEGG pathway enrichment analyses identified the tryptophan metabolism, the serotonergic synapse, and the vitamin B6 metabolism as significantly over- represented pathways (Figure 8c). Remarkably, tyrosine metabolism, which is crucial for dopamine regulation, emerged as the most significant pathways, mirroring findings in human plasma metabolomics, with increased tyrosine but decreased 4- hydroxyphenylacetic acid in the presence of Gokushovirus WZ-2015a (Figure 6e,g), alongside changes in other tyrosine-derived metabolites. These included increases in hydroxyphenyllactic acid and noradrenaline-4-O-glucuronide, and a reduction in the concentration of the acetylated form of 2-carboy-2,3-dihydro-5,6-dihydroxyindole (leucodopachrome), an oxidation product of the dopamine precursor L-DOPA (Figure 6h) in the presence of Gokushovirus WZ-2015a. The identification of tyrosine underscored again pathways lined to cocaine and amphetamine addiction identified in humans. Similarly, an HMDB pathway enrichment analysis identified the tryptophan metabolism as the most over-represented pathway, but also the tyrosine and vitamin B6 metabolism as well as several drug action pathways (Figure 9a, b). Notably, it also underscored pathways involved in the catecholamine biosynthesis as well as key enzymes involved in dopamine metabolism such as MAO, AADC, and DBH deficiency. Overall, these findings reveal that Gokushovirus WZ-2015a is consistently altering the dopaminergic system in the NAc by increasing the production of metabolites involved in dopamine metabolism and decreasing degradation metabolites of the dopamine precursor L-DOPA. In fact, research has consistently shown that acute expose to drugs such as morphine, cocaine, and amphetamine, usually lead to an increase in dopamine levels in the NAc, which is consistent with our findings using a short protocol to mimic the early period of food addiction (Martin- Garcia, E., Domingo-Rodriguez, L. & Maldonado, R. An Operant Conditioning Model Combined with a Chemogenetic Approach to Study the Neurobiology of Food Addiction in Mice. Bio Protoc O, e3777 (2020).).

[0350] We also identified metabolites linked to Gokushovirus WZ-2015a presence in the DS (Figure 9c). A KEGG pathway enrichment analysis underscored again the tyrosine metabolism as the most significant pathway, alongside the metabolism of the other AAA tryptophan and phenylalanine (Figure 9d,e). Notably, we found a strong increase in the DS the acetylated form of leucodopachrome but a reduction in 4- hydroxyphenylacetic acid O-glucuronide, contrary to changes observed in the NAc (Figure 8b). This is in agreement with the transition from controlled behavior to loss of control and compulsive drug seeking in addiction involving a progressive shift from ventral (NAc) to more dorsal domains of the striatum. At early periods, these adaptations are largely restricted to the more ventral, nucleus accumbens region, whereas the DS is engaged in more long-term, rather than acute, exposure. For example, reductions in D2 dopamine receptors in the DS were observed following chronic, but not acute, cocaine self-administration in monkeys (Everitt, B. J. et al. Review. Neural mechanisms underlying the vulnerability to develop compulsive drugseeking habits and addiction. Philos Trans R Soc Lond B Biol Sci 363, 3125-3135 (2008)).

[0351] The extended reward network is regulated by cognitive network regions. In particular, the executive control network, containing several regions from the prefrontal cortex (PFC), exerts inhibitory control on the reward and emotional-arousal networks. Insufficient inhibitory control of these networks plays a key role in shifting the balance from the predominant homeostatic regulation of food intake to a hedonic mechanism. In fact, dysfunction of the PFC has shown to underlie compulsive behaviors associated with addiction (Goldstein, R. Z. & Volkow, N. D. Dysfunction of the prefrontal cortex in addiction: neuroimaging findings and clinical implications. Nat Rev Neurosci 12, 652-669 (2011)) and can predict clinical outcomes in obesity (Weygandt, M. et al. Impulse control in the dorsolateral prefrontal cortex counteracts post-diet weight regain in obesity. Neuroimage 109, 318-327 (2015)). In the medial PFC (mPFC), the prelimbic area sends dense projections to cortical and subcortical regions, preferentially to the core parts of the NAc. Considering this, we performed a RNA sequencing of the mice mPFC. Remarkably, after fitting robust linear regression models we found that the dopamine receptor D2 (Drd2) was among the most downregulated genes in the mPFC due to the presence of Gokushovirus WZ-2015a in the donor’s microbiota (Figure 8e). Dopamine is the main neurotransmitter of the reward network that is under inhibitory control by the executive control network. In line with the metabolomics results (Figure 8c), a KEGG enrichment analysis of those genes significantly downregulated in the presence of Gokushovirus WZ-2015a identified the Gap junction and cocaine addiction as the most significant pathways (Figure 8f, Figure 11a). It also underscored a cluster of genes involving the GABAergic and dopaminergic synapse (Figure 8f) that included Drd2 (Figure 8g). Consistently, dopamine inputs from the VTA and SN to the NAc and DS innervate two main types of primary GABAergic projection neurons (medium spiny neurons) as well as GABAergic interneurons. Similarly, a Gene Ontology - Biological Process enrichment analysis identified several pathways related to dopamine secretion and dopamine receptor signalling pathways with Drd2 playing a central role (Figure 11 b,c). Consistently, in addiction, obesity, and binge eating disorders, brain imaging studies in humans have revealed a decrease in D2R expression and dopamine release in both the DS and NAc (Volkow, N. D., Wise, R. A. & Baler, R. The dopamine motive system: implications for drug and food addiction. Nat Rev Neurosci 18, 741- 752 (2017)), which has been linked to decreased activity in prefrontal regions. Leptin- deficient obese ob / ob mice have also reduced D2R in the striatum. High fat diets have also shown to reduce the expression of D2R in the DS and NAc in rats (Adams, W. K. et al. Long-term, calorie-restricted intake of a high-fat diet in rats reduces impulse control and ventral striatal D2 receptor signalling - two markers of addiction vulnerability. Eur J Neurosci 42, 3095-3104 (2015)), which is necessary for PFC regulation.

[0352] 2.6. Microbial-derived anthranilic acid are linked to Gokushovirus WZ-2015a

[0353] We further assessed the associations of Microviridae and Gokushovirus WZ-2015a with obesity in a third large-scale validation cohort including participants from the Health Imageomics Study (<50 years old) (figure 10a). Consistent with our previous findings, the iqlr-transformed Microviridae levels were positively associated with obesity (Figure 10b). Likewise, the quintiles of the iqlr-transformed Gokushovirus WZ- 2015a levels were positively associated with the BMI, with patients in the highest quintiles having the highest BMI (Figure 10c). In addition, we assessed food addiction using the YFAS 2.0 in a subset of participants (n=147). Remarkably, when we classified participants based on the presence or absence of food addiction, we found again that Microviridae was the only microbial increased with food addiction (Figure 10d). In women, Microviridae was strongly associated with food addiction (Figure 10e), while in men it was the family most strongly linked to motivation (Figure 10f), which was the vulnerability to addiction trait most consistently associated with Microviridae and Gokushovirus WZ-2015a in the other cohorts.

[0354] In the IRONMET-CGM cohort (discovery cohort), we found that the most robust association among metagenomics and metabolomics data, food addiction, and Gokushovirus WZ-2015a involved the tryptophan metabolism (Figure 6). In fact, tryptophan is the sole precursor of serotonin and tryptophan-derived catabolites have been identified as important mediators of the host-microbiota interactions in the context of food addiction and obesity by acting on the extended reward network. Therefore, we applied a targeted metabolomic approach on plasma samples of this second validation cohort to characterize a large number of metabolites involved in the three main tryptophan catabolic routes: the kynurenine, serotonin, and indolic pathways. To identify tryptophan catabolites predictive of the iqlr-transformed Gokushovirus levels, we applied a machine learning variable selection strategy based on multiple random forests (Kursa, M. B. & Rudnicki, W. R. Feature selection with the boruta package. J Stat Softw 36, 1-13 (2010)). Among all tryptophan-related metabolites, the microbial-derived indole-3-propionic acid (I PA) had the strongest association with Gokushovirus WZ-2015a (Figure 10g) followed by anthranilic acid (AA), a downstream metabolite of the kynurenine pathway. In addition, 2-picolinic acid and 5-hydroxy-L-tryptophan were also associated with Gokushovirus WZ-2015a. As machine learning models are usually considered as black boxes, we also determined the contribution and effect of each catabolite by the exact computation of SHAP scores (Lundberg, S. M. et al. From Local Explanations to Global Understanding with Explainable Al for Trees. Nat Mach Intell 2, 56-67 (2020)). Consistent with our previous results, I PA had the highest SHAP values for the prediction of Gokushovirus WZ-2015a, with high values having a strong impact (Figure 10h), followed closely by AA, which plasma concentrations were negatively associated with Gokushovirus \NZ- 2015a.

[0355] To further investigate the functionality of the microbiome linked to Gokushovirus \NZ- 2015Aa, we mapped reads to the KEGG orthologs (KO) to obtain functional annotation of microbial genes. Then, we used the fastANCOM methodology to identify microbial molecular functions associated with the presence of Gokushovirus \NZ- 2015a (>5 counts). In line with our metabolomics results, an over-representation analysis mapping significant KO (padj<0.05) to KEGG modules identified the tryptophan biosynthesis as one of the most enriched pathways (Figure 13e). A close look at the genes significantly associated with the Gokushovirus WZ-2015a and involved in tryptophan biosynthesis and metabolism showed a consistent negative association with genes participating in tryptophan and serotonin biosynthesis, but a positive association with genes from the indolic pathway (Figure 10i). In particular, consistent with our metabolomics findings, Gokushovirus WZ-2015a had the strongest positive association with acyl-Coa dehydrogenase (K06446), which participates in the biosynthesis of I PA from indole-3-acrylic acid, while it was negatively associated with several bacterial genes involved in the synthesis of tryptophan from AA, particularly anthranilate synthase component I (trpE) and II (frpG), anthranilate phosphoribosyltransferase (trpD), and phosphoribosylanthranilate synthase (trpF) (Figure 10i,j).

[0356] 2.7. AA supplementation decreases food addiction traits in mice through alterations in tryptophan and tyrosine metabolism Despite AA plays a crucial role in tryptophan synthesis and metabolism, it has been largely understudied compared to other tryptophan intermediates68. Given the strong association with Gokushovirus WZ-2015a, we further investigated the role of AA in food addiction by supplementing AA (0.3 mg / ml, n=12) or water (Control, n=11) to mice. Models of food addiction were developed in operant chambers under an FR1 schedule of reinforcement during 7 sessions followed by 25 sessions under FR5 to acquire an operant response maintained by chocolate-flavored pellets (Figure 12a). During FR1-FR5, all groups increased the number of reinforcers across sessions without significant differences between groups (Figure 12b). Interestingly, the number of reinforcers was significantly reduced in the AA group compared to control mice over the last FR5 sessions. These results reveal that chocolate-flavored pellets were less reinforcing for mice supplemented with AA, suggesting that these supplementations may represent a protective factor.

[0357] We also assessed the three addiction criteria. AA supplementation decreased the motivation for chocolate-flavored pellets (Figure 12c) and the persistence to response in mice compared to the control group (Figure 12d), whereas we did not find differences between groups in compulsive-like behavior (Figure 12e). Again, these results are in line with our findings in humans and mice after FVT, where Microviridae and Gokushovirus WZ-2015a were consistently positively associated with motivation and persistence to response but not compulsivity (Figure 2g-i,l-n, Figure 19). AA supplementation reduced impulsive-like behavior in mice (Figure 12f). However, no differences between groups were found in appetitive associative learning (Figure 12g) and cognitive flexibility (Figure 12h).

[0358] Supplementation with AA increased the concentrations of such metabolite both in the NAc (Figure 12i) and DS (Figure 12j). A global metabolic profiling of the NAc further highlighted an increase in AA and its product 3-HAA in the NAc of mice receiving AA compared to the control group (Figure 12k). Remarkably, those metabolites the experienced a largest reduction in the NAc of mice supplemented with AA compared to the control mice were 3-Ethylmalic acid, 4,6- and 4,8-dihydroxyquinoline (Figure 12k), which were precisely the same metabolites that increased the most with FMT from donors with presence of Gokushovirus WZ-2015a (Figure 8b). Similarly, one of the metabolites most increased with AA supplementation was leucodopachrome (Figure 12k), which we also found strongly reduced in the NAc of mice receiving microbiota with presences of Gokushovirus WZ-2015a (Figure 8b). Consistent with our results in humans (Figure 6f) and the FMT experiment (Figure 8c), a KEGG enrichment analysis revealed a significant over-representation of the tryptophan, tyrosine, glutamate, GABAergic synapse and nicotine, morphine, and cocaine addiction pathways (Figure 121). The metabolites involved in these pathways included glutamate, GABA, a- ketoglutarate, or adenosine, all reduced in the NAc after AA supplementation (Figure 12m). Notably, glutamate, the main excitatory neurotransmitter was reduced after AA supplementation, whereas GABA, the main inhibitory neurotransmitter was increased in the NAc. The NAc receives dense glutamatergic input from cortical regions (e.g. PFC; anterior cingulate cortex, ACC) and sends GABAergic projections primarily to basal ganglia nuclei Scofield, M. D. et al. The Nucleus Accumbens: Mechanisms of Addiction across Drug Classes Reflect the Importance of Glutamate Homeostasis. Pharmacol Rev 68, 816-871 (2016). Impaired glutamate homeostasis between cortical regions and the NAc has been suggested to underlie addiction-like behaviour (Kalivas, P. W. The glutamate homeostasis hypothesis of addiction. Nat Rev Neurosci 10, 561-572 (2009)). Chronic drug exposure has shown to trigger glutamatergic-mediated neuroadaptations in dopamine striato-cortical pathways, predominantly in PFC and AAC regions, whereas the GABAergic system also plays a critical role in the modulation of mesolimbic dopaminergic reward regions. In fact, a recent study has shown that neurons in the ACC projecting to the NAc are essential in reward-seeking behaviour (Fetcho, R. N. et al. A stress-sensitive frontostriatal circuit supporting effortful reward-seeking behavior. Neuron 112, 473-487. e4 (2024)). In addition, participants with compulsive behaviour exhibited elevated glutamate and lower GABA levels in the ACC (Biria, M. et al. Cortical glutamate and GABA are related to compulsive behaviour in individuals with obsessive compulsive disorder and healthy controls. Nat Commun 14, 3324 (2023)). Similarly, opioid, cocaine, or methamphetamine use in humans has shown to decrease the PFC GABA levels (Shyu, C., Chavez, S., Boileau, I. & Foil, B. Le. Quantifying GABA in Addiction: A Review of Proton Magnetic Resonance Spectroscopy Studies. Brain Sci 12, 918 (2022)). In animal models, nicotine addiction has shown to increase extracellular glutamate levels in the NAc (Reid, M. S., Fox, L., Ho, L. B. & Berger, S. P. Nicotine stimulation of extracellular glutamate levels in the nucleus accumbens: neuropharmacological characterization. Synapse 35, 129-36 (2000)).

[0359] Glutamate and a-ketoglutarate, also participate in the glyoxylate and dicarboxylate metabolism, which was the most over-represented pathways (Figure 121), that also included 3-ethylmalic acid and aconitic acid, both also decreased in the NAc of the AA group. Notably, the later pathway was connected to the tyrosine metabolism through pyruvic acid and gentisic acid (Figure 12n), both increased after AA supplementation (Figure 12k). Moreover, 3-ethylmalic is produced from glyoxylate, which in turn can also react with glutamate to produce glycine and a-ketoglutarate through the GABA shunt connecting GABA and glutamate to the TCA cycle (Figure 12n). In addition, pyruvate can produce acetyl-CoA that can also enter the TCA cycle and produce glyoxylate from isocritate, which precursors is cis-aconitic acid (Figure 12n).

[0360] The folate biosynthesis was also one of the most significant pathways. The significant metabolites in this pathway included 7,8-dihydrobiopterin (BH2), which was strongly decreased after supplementation with AA. BH2 is involved in the synthesis of tetrahydrobiopterin (BH4) from 5-methyltetrahydrofolate, (5-MTHF) generating folic acid through the folate cycle, which also involves 5-Formyltetrahydrofolate (folinic acid)75. Importantly the acetylated forms of folinic acid and 5-MTHF were altered with AA supplementation (Figure 12j). This process is coupled to the methionine cycle, which involved S-adenosylmethionine (SAM) and S-adenosylhomocysteine (SAH), with the later also associated with the presence of Gokushovirus WZ-2015a in the FMT experiment (Figure 8b). Notably, BH4 is a key cofactor for the enzymes involved in the synthesis of dopamine and serotonin (Figure 6h) (Fanet, H., Capuron, L., Castanon, N., Calon, F. & Vancassel, S. Tetrahydrobioterin (BH4) Pathway: From Metabolism to Neuropsychiatry. Curr Neuropharmacol 19, 591-609 (2021)). The folate cycle also participates in the synthesis of serine from glycine (Figure 12n), which was significantly increased after AA supplementation. Finally, an HMDB enrichment analyses highlighted similar pathways (Figure 15), but also pathways involved in DBH, MAO, and DOPA deficiency, which is in agreement with the associations observed with FMT from donors with Gokushovirus WZ-2015a. To further validate the role of AA in food addiction, we identified the plasma metabolites associated with the YFAS scores in the IRONMET-CGM cohort. Notably, we observed that some of the most significant metabolites linked to increased food addiction included aspartic acid and glutamic acid, while cis-aconitic acid had the strongest fold change (Figure 16a). Consistently, AA supplementation led to a significant decrease in the levels of glutamic acid, aconitic acid, and N-Methyl aspartic acid in the NAc of mice (Figure 12k). Conversely, our analysis revealed that the three metabolites most negatively associated with food addiction included shikimic acid, 3- ethylmalic acid, and N-carboxyethyl-g-aminobutyric acid (CE-GABA) (Figure 16a). The shikimic acid pathway is responsible for the microbial production of aromatic amino acids (tyrosine, tryptophan, phenylalanine). Moreover, we found that 3- ethylmalic acid was one of the metabolites most consistently altered in the NAc of mice after FMT (Figure 8b) or AA supplementation (Figure 12k). Finally, CE-GABA has been identified in mammalian brain and has been shown to participate in the synthesis of GABA (which was increased in the NAc of mice after AA supplementation) from spermidine (Fussi, F., Savoldi, F. & Curti, M. Identification of N-carboxyethyl gamma-aminobutyric acid in bovine brain and human cerebrospinal fluid. Neurosci Lett 77, 308-310 (1987)). A KEGG-based enrichment analysis identified again the glyoxylate and dicarboxylate metabolism as well as other pathways involved in cocaine and nicotine addiction, or the glutamatergic or GABAergic synapse (Figure 16c, d), whereas an HMDB enrichment analyses underscored the significance of the tyrosine an glutamate metabolism, and the MAO, DBH, and GABA-transaminase deficiency (Figure 16e,f).

[0361] We next performed an RNA sequencing of the mPFC, as it may influence addictive behaviors through its projections to the NAc (Figure 8a), and identified differentially expressed genes between the AA and control groups. A pathway enrichment analysis of those genes downregulated after AA supplementation revealed a downregulation of pathways involved in the release of several neurotransmitters (GABA, dopamine, glutamate, serotonin) as well as the glutamatergic synapse and nicotine and morphine addiction (Figure 12o). Importantly, the release of glutamate from the mPFC into the core of the NAc has shown to mediate reinstatement of drug-seeking behavior (McFarland, K., Lapish, C. C. & Kalivas, P. W. Prefrontal glutamate release into the core of the nucleus accumbens mediates cocaine-induced reinstatement of drug- seeking behavior. J Neurosci 23, 3531-3537 (2003)). Thus, inhibiting prefrontal cortical glutamatergic neurons that project to the NAC prevents rises in glutamate. This is consistent with the downregulation of glutamatergic synapase and glutamate release in the mPFC and reductions in glutamate concentrations in the NAc of mice supplemented with AA.

[0362] 2.8. AA regulates feeding behavior and addiction-like ethanol preference in Drosophila melanogaster potentially interacting with the dopaminergic system

[0363] We next used the model organism Drosophila melanogaster to further investigate the role of AA and the dopaminergic system in food intake. We supplemented the Drosophila standard diet with AA and fed flies during larval development and adulthood. When flies were three to five days old, food intake was measured by excreta quantification (EXQ). Similarly, we assessed starvation resilience in Drosophila by measuring the amount of dyed-food ingested in a period of 30 min after 40 hours of starvation (we refer to this method as dye uptake). AA supplementation (200 mg / L) significantly decreased food consumption in both cases (Figure 14a,b). Additionally, flies feed with AA showed a significant decrease in overall body weight (Figure 14c).

[0364] Dopamine receptor 2 (Dop2R), the orthologue of Drd2 and Drd3 in mice, has already been described to play a role in feeding behavior in Drosophila. We used the driver w; UAS-Dcr2; elav-GAL.4 to downregulate Dop2R in neurons and found a decrease in the food intake of 4-to-5 day old flies when performing EX-Q compared to controls (Figure 14d). Conversely, the enhancement of the dopaminergic signaling by overexpression of Dop2R in neurons led to overfeeding (Figure 14e).

[0365] While assays specifically tailored to determine food addiction in Drosophila are currently unavailable, the consumption of ethanol in fruit flies exhibits notable parallels with addiction. Fruit flies demonstrate a clear preference for ethanol-containing food and this preference exhibits several features reminiscent of compulsive alcohol consumption. Similar to mammals, the dopaminergic system in this organism also plays a crucial role in the development of ethanol-related behaviors. To deepen our understanding of the involvement of AA in addiction we conducted food preference two-choice CAFE (Capillary Feeder) assays in Drosophila to measure ethanol consumption and preference (Devineni, A. V. & Heberlein, II. Preferential ethanol consumption in Drosophila models features of addiction. CurrBiol 19, 2126- 2132 (2009)). In this assay, flies were presented with a choice between capillaries containing 5% sucrose or 5% sucrose + 15% ethanol (Figure 14f). Ethanol preference was assessed using a preference index (PI), which ranges from -1 to +1 , with positive values indicating preference for ethanol and negative values indicating repulsion. As expected, the PI for ethanol increased steadily over time in wild type flies (Figure 14g). However, upon supplementation with AA, flies did not increase their PI for ethanol over the course of three days (Figure 14g). This observation aligns with our previous findings in mice, indicating that AA confers protective effects against the development of addiction.

[0366] To explore a potential interaction between AA and the dopaminergic system in the control of addiction, we used the UAS-Gal4 system to selectively downregulate the Drosophila dopamine receptor 2 (DopR2) within the mushroom body, as dopaminergic input into the mushroom body mediates conditioned ethanol preference (Chvilicek, M. M., Titos, I. & Rothenfluh, A. The Neurotransmitters Involved in Drosophila Alcohol-Induced Behaviors. Front Behav Neurosci 14, 607700 (2020); Kaun, K. R., Azanchi, R., Maung, Z., Hirsh, J. & Heberlein, II. A Drosophila model for alcohol reward. Nat Neurosci 14, 612-621 (2011)). As expected, flies lacking DopR2 expression in the mushroom body did not develop a preference for ethanol over time (Figure 14h), additionally supplementation of these flies with AA did not further decrease the PI to ethanol (Figure 14h). Next, we disrupted dopaminergic synaptic transmission by inducing the expression of the active form of tetanus toxin (TNT) in dopaminergic neurons using with the dopaminergic driver ple-GAL4. The flies were kept in starvation and given a two-hour feeding window each day, during which they were offered a choice between food with or without ethanol. On the fourth day, flies underwent food preference assays using the flyPAD apparatus. Flies with an inhibited dopaminergic signaling (TNT), exhibited a significantly lower PI for ethanol compared to flies with intact dopaminergic system, expressing an inactive form of TNT (impTNT) (Figure 17a). Additionally, flies with an intact dopaminergic system treated with AA (impTNT+AA) had reduced preference for ethanol compared to control flies (impTNT) (Figure 14i), similar to the levels observed in flies with active TNT expression (Figure 8i, 17b). Finally, supplementing flies with the dopaminergic system inactivated (TNT) with AA did not further increase their preference for sucrose (Figure 17c), suggesting that AA may exert its protective effect through modulation of the dopaminergic system. Thus AA administration confers a protective effect to the development of food addiction in mice and is regulating feeding behavior, food preference and body weight in Drosophila melanogaster.

[0367] 2.9. Gokushovirus WZ-2015a upregulates expression of dopamine receptor 3 and other neurotransmitter receptors in the mPFC of recipient mice alongside decreases in Faecal i bacterium

[0368] To further study the potential mechanisms underlying vulnerability to food addiction caused by the presence of Gokushovirus, we performed a second faecal microbiota transplantation experiment (FMT). Twenty-two antibiotic-treated mice received microbiota from human donors with either presence or absence of Gokushovirus \NZ- 2015a. After four weeks, we performed an RNA-sequencing of the medial prefrontal cortex (mPFC) of recipient mice (Figure 14j). At the end of the study, we found a trend towards an increase in body weight in those recipient mice receiving microbiota from patients with higher iqlr-transformed Gokushovirus WZ-2015a levels after controlling for donor’s sex (Figure 14k). In fact, mice receiving microbiota from women donors with presence of Gokushovirus WZ-2015a in their microbiota had a trend to higher body weight compared to those receiving microbiota from women donors without Gokushovirus WZ-2015a (P=0.06, Figure 141, m). We identified several mPFC genes from recipient mice associated with the donor iqlr-transformed Gokushovirus WZ-2015a levels after controlling for donor’s age, sex, and BMI (Figure 14n).

[0369] To gain better insights in the potential mechanisms underlying the Gokushovirus \NZ- 2015a on susceptibility to food addiction, we built a gene-gene interaction network considering the 42 genes with an absolute Iog2(fold change) > 1 and pFDR<0.2 and performed over-representation analyses mapping the selected genes to the Gene Ontology terms and KEGG pathways included in the Consensus Pathway DataBase. We identified a cluster comprising genes known to regulate neurotransmitter release such as Dopamine Receptor D3 (Drd3), Histidine Decarboxylase (Hdd), Adrenoreceptor Alpha 2B (Adra2b), or Gamma-Aminobutyric Acid Type A Receptor Subunit Alpha 6 (Gabra6) (Figure 18a). Consistently, GO and KEGG overrepresentation analyses revealed that most of these genes participated in postsynaptic neurotransmitter receptor activity (Figure 18b,c) and the neuroactive ligand-receptor interaction, GABAergic synapse, nicotine and morphine and addiction (Figure 14o, Figure 18d), respectively, in strong agreement with the pathways identified in the other FMT experiment (Figure 8f,g). Remarkably, while first FMT experiment revealed a strong downregulation of Drd2 in the presence of Gokushovirus WZ-2015a, in this second FMT experiment the expression of the dopamine receptor D3 (Drd3) had the highest fold change associated with the donor iqlr-transformed Gokushovirus WZ-2015a levels (Figure 14n). Drd3 belongs to the D2-type receptors (D2R, D3R, and D4R) and has been associated with drug-addiction (Sokoloff, P. & Le Foil, B. The dopamine D3 receptor, a quarter century later. Eur J Neurosci 45, 2-19 (2017)).

[0370] Finally, we conducted shotgun metagenomics sequencing of recipient mice feces and analyzed the gut microbiota associated with the donors’ iqlr-transformed Gokushovirus WZ-2015a levels (Figure 14p). Increases in the Gokushovirus WZ- 2015a levels were associated with a strong reduction in several species from the Verrucomicrobia phylum including Akkermansia muciniphila CAG.154, Akkermansia sp.54_46, Akkermansia_uc, Akkermansia muciniphila, or Akkermansia glycaniphila. However, the most significant association was a reduction in Faecalibacterium sp. An 122. In addition, donor’s Gokushovirus WZ-2015a levels were also negatively associated with Faecalibacterium prausnitzii, which we identified as its host using CRISPR. Recently, F. prausnitzii has shown to alleviate disordered overeating behaviors by decreasing preference for highly palatable food through increase in tryptophan-related metabolites (Fan, S. et al. Microbiota-gut-brain axis drives overeating disorders. Cell Metab 35, 2011-2027. e7 (2023)). Therefore, this further supports that Gokushovirus WZ-2015a could partially exert its effects on food addiction through changes in Faecalibacterium.

[0371] 3. Conclusions Despite its predominance and the potential to regulate bacterial communities and function and consequently human health, the gut phageome remains largely unexplored. Our results show that bacteriophages from the Microviridae family, in particular the Gokushovirus WZ-2015a, are associated with obesity, food addiction, phenotypic traits considered as factors of vulnerability to food addiction, and hyperconnectivity in functional cortical and subcortical areas consistently implicated in addiction and obesity. The potential role of Gokushovirus was evaluated by FMT and FVT from human donors to mice: recipient mice receiving microbiota from donors with presence of Gokushovirus WZ-2015a displayed higher addiction criteria and addiction-related phenotypic traits as well as alterations in the tryptophan and tyrosine metabolism, the precursors of serotonin and dopamine, in the NAc and DS, and misregulation of dopamine receptors and glutamatergic and GABAergic-related genes in the mPFC. Faecalibacterium prausnitzii, which has recently shown to alleviate overeating disorders, was identified as the Gokushovirus host and its levels decreased after FMT with presence of Gokushovirus.. AA, which exhibited a strong negative association with Gokushovirus WZ-2015a in humans, also demonstrated a protective effect against food addiction traits in mice. Metabolomic and transcriptomic analyses in mice brains revealed AA's involvement in the synthesis and release of several neurotransmitters including dopamine and the modulation of genes and metabolites impacted in addiction. AA supplementation in Drosophila reduced food consumption and ethanol like-addiction behaviours. The potential modulatory effects of AA on circuits and pathways underlying addiction highlight its promising utility as a therapeutic agent in managing addictive behaviours and related conditions. Overall, our work provides novel findings on the gut virome and may lay the groundwork to illuminate new approaches for the treatment of food addiction and obesity.

Claims

CLAIMS1. Method for in vitro diagnosis of food addiction and / or obesity in a subject, comprising the following steps: a) determining the levels of Microviridae bacteriophages in an isolated biological sample from the subject; and b) comparing the levels of Microviridae determined in step a) with control values, wherein elevated levels of Microviridae compared to control values, indicate that the subject suffers from food addiction and / or obesity.

2. Method for in vitro monitoring the evolution of food addiction and / or obesity in a subject, comprising the following steps: a) determining in, at least, two different moments in time, the levels of Microviridae bacteriophages in an isolated biological sample from the subject; and b) comparing the levels of Microviridae determined in said moments with one another, wherein an increase in Microviridae levels indicates that the food addiction and / or obesity evolves negatively in the subject.

3. Method for evaluating in vitro the response of a subject to a treatment for food addiction and / or obesity, comprising the following steps: a) determining, before and after treatment, the levels of Microviridae bacteriophages in an isolated biological sample from the subject; and b) comparing the levels of Microviridae determined before the treatment with those obtained after the treatment, wherein lower levels of Microviridae after treatment than before treatment indicate that the treatment is effective and / or that the subject is responding positively to treatment.

4. The method according to any one of claims 1 to 3, wherein the subject is a human being.

5. The method according to any one of claims 1 to 4, wherein the Microviridae bacteriophages belong to Gokushovirus.

6. The method according to claim 5, wherein Gokushovirus is Gokushovirus WZ- 2015a.

7. The method according to any one of claims 1 to 6 wherein the isolated biological sample is selected from the list consisting of a sample isolated from the subject digested system, urine, blood and salive, preferably, the sample isolated from the subject digested system is a faecal / stool sample.

8. Use of the levels of Microviridae bacteriophages for the in vitro diagnosis of food addiction and / or obesity in a subject, preferably wherein the Microviridae bacteriophages belong to Gokushovirus.

9. Use of the levels of Microviridae bacteriophages for the in vitro monitoring of the evolution of food addiction and / or obesity in a subject, preferably wherein the Microviridae bacteriophages belong to Gokushovirus.

10. Use of the levels of Microviridae bacteriophages for the in vitro evaluation of the response of a subject to a treatment for food addiction and / or obesity, preferably wherein the Microviridae bacteriophages belong to Gokushovirus.

11. The use according to any one of claims 8 to 10, wherein the subject is a human being.

12. Kit comprising means for the in vitro determination of the levels of Microviridae bacteriophages, preferably Gokushovirus, in an isolated biological sample from the subject, wherein the means are primers specifically recognizing Microviridae, preferably Gokushovirus, and / or probes specifically recognizing Microviridae, preferably Gokushovirus.

13. Use of a kit according to claim 12 for the in vitro diagnosis of food addiction and / or obesity in a subject, or for the in vitro monitoring of the evolution of food addiction and / or obesity in a subject, or for the in vitro evaluation of the response of a subject to a treatment for food addiction and / or obesity.

14. Composition comprising anthranilic acid for use in the treatment and / or prevention, of food addiction and / or obesity in a subject.

15. Composition for use according to claim 14, wherein the subject has been diagnosed to suffer from food addiction and / or obesity with the method according to any one of claims 1 , 4, 5, 6 or 7.

16. Non-therapeutic use of a composition comprising anthranilic acid in body fatreduction of a subject.

17. Non-therapeutic use according to claim 16, wherein the subject has been diagnosed to suffer from food addiction with the method according to any one of claims 1 , 4, 5, 6 or 7.

Images