Compositions and methods for treating neurological disorders

Abstract

Disclosed herein are bacterial compositions and methods for treating neurological disorders, improving brain function, and reducing stress.

Description

[0001] COMPOSITIONS AND METHODS FOR TREATING NEUROLOGICAL DISORDERS

[0002] RELATED APPLICATIONS

[0003] This application claims the benefit of priority to U.S. Provisional Application No. 63 / 662,372, filed June 20, 2024, the contents of which are incorporated herein by reference in their entirety.

[0004] STATEMENT OF GOVERNMENT SUPPORT

[0005] This invention was made with government support under DK121025, DK007180, DK106528, and MD015904 awarded by the National Institutes of Health. The government has certain rights in the invention.

[0006] BACKGROUND

[0007] Upwards of $300 billion dollars is lost annually due to stress-related health care costs and missed work in the U.S. highlighting the need for greater resilience to stress. Definitions of resilience generally refer to beneficial outcomes in response to threat or stressful events. Resilience involves positive acceptance of change, tolerance of negative affect, tenacity, and the ability to recover after stressful events. Resilience can predict stress-related depression and anxiety, traumatic stress, and maladaptive coping mechanisms such as alcohol misuse. Most research investigating resilience has focused its correlations with personality traits, emotional and behavioral regulation strategies, and social factors including supportive interpersonal relationships. Resilience is also related to autonomic flexibility and adaptive neurological correlates.

[0008] The composition and function of the human microbiome, an ecosystem consisting of trillions of microorganisms residing in and on the human body, have been linked to stress- related disorders and other mental health diagnoses. The gut microbiome modulates psychological functioning by influencing the brain-gut microbiome (BGM) system, a bidirectional signaling mechanism between the central nervous system (CNS) and gastrointestinal tract. Interestingly, the microbiome has been implicated in conferring stressresilience. Evidence reveals that alterations in BGM signatures can influence prosocial behaviors related to resilience and stress-related psychopathologies. Numerous animal studies also reveal the role of the BGM in resilience versus susceptibility traits after stress exposure. For example, gut microbial composition differed in mice susceptible compared to mice resilient to chronic social defeat stress, as well as to learned helplessness after exposure to inescapable stress. This opens the intriguing possibility that the endogenous gut microbiome may house stress mitigating therapeutic metabolites supporting neurologically adaptive processes. For instance, bacterial transcriptomes are related to a number of microbiome functions and serve to maintain a balanced and diverse population of gut microbiota (eubiosis) and gut barrier integrity. Thus, a resilient phenotype involves expression of transcriptomes that ensure appropriate modulation of the BGM system, and therefore, an adaptive CNS.

[0009] Collectively, the gut microbiota produces a number of metabolites, including hormones, neurotransmitters (such as Gamma-Aminobutyric Acid; GABA, glutamate, and serotonin), and other signaling molecules within the BGM implicated in stress-related psychopathology. Via serotonin, gut microbiota through neural activation of vagal afferents in the gut, can contribute to a regulated autonomic nervous system and adaptive stress response. The main metabolites produced by gut microbiota are short-chain fatty acids (SCFAs) which are known to influence cognitive and emotional processing through effects on the brain via anti-inflammatory properties. In animal studies, low SCFA has been associated with anxiety and depression-related behaviors and also predicted resilience versus susceptibility to traumatic stress.

[0010] The gut microbiota can also shape brain structure and function, as evidenced by probiotic studies. Probiotics have resulted in decreased activity in somatosensory and viscerosensory cortices in response to emotional attentional tasks and reduced grey matter volume and increased resting state functional connectivity (rsFC) within the default mode network (DMN). A recent neuroscience review of resilience highlighted the mesolimbic reward system, the DMN, and regions involved in fear and stress, namely, the amygdala. The DMN, activated during passive rest, daydreaming, and thinking about the past or others has been linked to major depressive disorder (i.e., excessive rumination but also dispositional mindfulness). Resilience-related grey matter volume changes have been observed in cognitive and affective regions (amygdala, subgenual and rostral anterior cingulate cortex). Lower activity within the salience network but increased connectivity within the sensorimotor network and greater responses in reward circuits (basal ganglia) have been linked to resilience. There are also associations between resilience and white matter tract integrity within structures implicated in social cognition.

[0011] In view of the foregoing, new compositions and methods are needed for harnessing the power of the microbiome to treat neurological disorders and improve brain function. SUMMARY OF THE INVENTION

[0012] In certain aspects, the present disclosure provides methods for treating a neurological disorder in a subject, comprising administering a composition comprising bacteria of at least one genus selected from Prevotella, Haemophilus, Peptococcus, and Lactococcus, or a combination thereof, to the subject.

[0013] In certain aspects, the present disclosure provides methods for improving cognitive function in a subject, comprising administering a composition comprising bacteria of at least one genus selected from Prevotella, Haemophilus, Peptococcus, and Lactococcus, or a combination thereof, to the subject.

[0014] In certain aspects, the present disclosure provides methods for decreasing stress in a subject, comprising administering a composition comprising bacteria of at least one genus selected from Prevotella, Haemophilus, Peptococcus, and Lactococcus, or a combination thereof, to the subj ect.

[0015] In certain aspects, the present disclosure provides methods for increasing levels of short-chain fatty acids (SCFAs) in a subject, comprising administering a composition comprising bacteria of at least one genus selected from Prevotella, Haemophilus, Peptococcus, and Lactococcus, or a combination thereof, to the subject.

[0016] In certain aspects, the present disclosure provides methods for treating a neurological disorder in a subject, comprising administering to the subject means for increasing in the subject an amount of bacteria of at least one genus selected from Prevotella, Haemophilus, Peptococcus, and Lactococcus, or a combination thereof.

[0017] In certain aspects, the present disclosure provides methods for treating a neurological disorder in a subject, comprising administering to the subject means for increasing in the subject levels of short-chain fatty acids (SCFAs).

[0018] In certain aspects, the present disclosure provides methods for improving cognitive function in a subject, comprising administering to the subject means for increasing in the subject an amount of a bacteria of at least one genus selected from Prevotella, Haemophilus, Peptococcus, and Lactococcus, or a combination thereof.

[0019] In certain aspects, the present disclosure provides methods for decreasing stress in a subject, comprising administering to the subject means for increasing in the subject an amount of bacteria of at least one genus selected from Prevotella, Haemophilus, Peptococcus, and Lactococcus, or a combination thereof. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] FIGs. 1A-1E show a graphical summary of parameters related to the studies disclosed herein. FIG. 1A depicts demographics showing the CD-RISC (Connor-Davidson Resilience Scale) threshold for group label, and the sample size per group. FIG. IB depicts clinical and psychological variables collected from self-report questionnaires; metabolites and transcriptomics data collected from stool samples; and multimodal brain images collected via magnetic resonance imaging. FIG. 1C shows that each of the six data blocks were cleaned and prepared as individual datasets. FIG. ID depicts data integration for biomarker discovery using latent components as used to create signatures for high and low resilience. FIG. IE depicts a summary of results. Abbreviations: BMI, body mass index; MRI, magnetic resonance imaging; RNA, Ribonucleic acid; KEGG, Kyoto Encyclopedia of Genes and Genomes; NAG, N-acetylglutamate; DMG, dimethylglycine; Vol, volume; SA, surface area; ERN, emotion regulation network; SMN, sensorimotor network; DMN, default mode network; CAN, central autonomic network; CEN, central executive network; K07335, basic membrane protein A and related proteins; K03315, Na / H antiporter; K01081, 5’- nucleotidase; K07080, uncharacterized protein; K14645, serine protease; K02343, DNA polymerase III subunit gamma / tau; K01596, phosphoenolpyruvate carboxykinase; K01361, lactocepin; K01844, beta-lysine 5,6-aminomutase alpha subunit; K03388, heterodisulfide reductase subunit A2.

[0021] FIG. 2 shows boxplots of DIABLO-selected variables of importance. Standardized values equate to variables of importance for each dataset between High (blue) and Low (orange) resilience. The line inside each box represents the median standardized value. Each box itself represents the interquartile range which captures 50% of standardized values. The lines extending from each box represent the ranges (extending up to the minimum and maximum) of standardized values outside of the interquartile range. Individual data points beyond the lines are considered outliers. Abbreviations: IPIP, International Personality Item Pool; STAI, Stat-Trait Anxiety Inventory; HAD, Hospital Anxiety and Depression scale; FFM, Five Facet Mindfulness; MASQ, Multiple Ability Self-Report Questionnaire; PSS, Perceived Stress Scale; K07335, basic membrane protein A and related proteins; K03315, Na / H antiporter; K01081, 5 ’-nucleotidase; K07080, uncharacterized protein; K14645, serine protease; K02343, DNA polymerase III subunit gamma / tau; K01596, phosphoenolpyruvate carboxykinase; K01361, lactocepin; K01844, beta-lysine 5,6-aminomutase alpha subunit; K03388, heterodisulfide reductase subunit A2; K03466, DNA segregation ATPase FtsK / SpoIIIE; K02965, small subunit ribosomal protein S19; K01119, 2',3'-cyclic-nucleotide 2'-phosphodiesterase / 3 '-nucleotidase; K01834, 2,3-bisphosphoglycerate-dependent phosphoglycerate mutase; K05516, curved DNA-binding protein; K07667, two-component system, OmpR family, KDP operon response regulator KdpE; R, right; L, left; SA, surface area; Vol, volume; SbCaG, Subcallosal Gyrus; AngG, Angular Gyrus; InfPrCS, Inferior part of the Precentral Sulcus; VTA, Ventral Tegmental Area; Tha, Thalamus proper; MRF, Mesencephalic Reticular Formation; SuMarG, Supramarginal Gyrus; InfCirlns, Inferior segment of the Circular Sulcus of the Insula; Pal, Pallidum; SbCG S, Subcentral Gyrus and Sulci; SupPL, Superior Parietal Lobule; LORs, Lateral Orbital Sulcus; Pu, Putamen.

[0022] FIG. 3 shows loading Plots from the DIABLO-selected variables of importance. Loading plots of top contributing variables selected by DIABLO separated by dataset type. Blue represents high resilience; orange represents low resilience. Magnitude of loading coefficient represents the level of contribution for prediction. Greater values represent greater contribution. Abbreviations: MASQ, Multiple Ability Self-Report Questionnaire; FFM, Five Facet Mindfulness; HAD, Hospital Anxiety and Depression scale; IPIP, International Personality Item Pool; PSS, Perceived Stress Scale; STAI, Stat-Trait Anxiety Inventory; K07335, basic membrane protein A and related proteins; K03315, Na / H antiporter; K01081, 5 ’-nucleotidase; K07080, uncharacterized protein; K14645, serine protease; K02343, DNA polymerase III subunit gamma / tau; K01596, phosphoenolpyruvate carboxykinase; K01361, lactocepin; K01844, beta-lysine 5,6-aminomutase alpha subunit; K03388, heterodisulfide reductase subunit A2; K03466, DNA segregation ATPase FtsK / SpoIIIE; K02965, small subunit ribosomal protein S19; K01119, 2',3'-cyclic-nucleotide 2'-phosphodiesterase / 3'- nucleotidase; K01834, 2,3-bisphosphoglycerate-dependent phosphoglycerate mutase; K05516, curved DNA-binding protein; K07667, two-component system, OmpR family, KDP operon response regulator KdpE; MRI, magnetic resonance imaging; L, left; R, right; SbCaG, subcallosal gyrus; AngG, angular gyrus; InfPrCS, inferior part of the precentral sulcus; VTA, ventral tegmental area; Tha, thalamus; MRF, mesencephalic reticular formation; SuMarG, supramarginal gyrus; InfCirlns, inferior segment of the circular sulcus of the insula; Pal, pallidum; SupPL, superior parietal lobule; LORs, lateral orbital sulcus; Pu, putamen; SupOcG, superior occipital gyrus; Hip, hippocampus.

[0023] FIG. 4 shows connectogram depicting the correlations within the variables of importance from all the datasets. Spearman correlations between features across datasets are shown via the lines inside the circle (cutoff r = 0.55). Mean levels of each feature for each group are represented via the lines within each data block forming the circle. A higher line represents a greater mean value for that group. Abbreviations: MRI, magnetic resonance imaging; L, left; R, right; SbCaG, subcallosal gyrus; AngG, angular gyrus; InfPrCS, inferior part of the precentral sulcus; VTA, ventral tegmental area; Tha, thalamus; MRF, mesencephalic reticular formation; SuMarG, supramarginal gyrus; InfCirlns, inferior segment of the circular sulcus of the insula; Pal, pallidum; SupPL, superior parietal lobule; LORs, lateral orbital sulcus; Pu, putamen; SupOcG, superior occipital gyrus; Hip, hippocampus; K07335, basic membrane protein A and related proteins; K03315, Na / H antiporter; K01081, 5 ’-nucleotidase; K07080, uncharacterized protein; K14645, serine protease; K02343, DNA polymerase III subunit gamma / tau; K01596, phosphoenolpyruvate carboxykinase; K01361, lactocepin; K01844, beta-lysine 5,6-aminomutase alpha subunit; K03388, heterodisulfide reductase subunit A2; K03466, DNA segregation ATPase FtsK / SpoIIIE; K02965, small subunit ribosomal protein S19; K01119, 2',3'-cyclic-nucleotide 2'-phosphodiesterase / 3'- nucleotidase; K01834, 2,3-bisphosphoglycerate-dependent phosphoglycerate mutase;

[0024] K05516, curved DNA-binding protein; K07667, two-component system, OmpR family, KDP operon response regulator KdpE.

[0025] FIG. 5 shows the distribution of group labels and depicts a histogram of CD-RISC (Connor-Davidson Resilience Scale) Scores for 116 heathy subjects. 83.1 threshold is based on previous literature for overweight and obese subjects.

[0026] FIG. 6 shows Shannon diversity of those in the low and high resilience groups. Box plots showing Shannon diversity across low (pink) and high (blue) resilience groups. All samples were sequenced using 16S sequencing. The center line within each box defines the mean, boxes define the upper and lower quartiles, and whiskers define the interquartile range.

[0027] FIG. 7 shows differences in beta diversity between low and high resilience groups. Nonmetric multidimensional scaling (MDS) ordination using Bray-Curtis dissimilarity to calculate distances on the genus level between resilience groups: low (pink) and high (blue). Each point represents one sample at one time point. The x and y-axis only provide a reference in which to gauge Bray-Curtis distances in relation to the other samples but are not a meaningful value independently.

[0028] FIGs. 8A-8C show DIABLO-derived Testing Set Performance and Confusion Matrix of Testing Set. Testing set includes a total N = 35 subjects and training set includes a total N = 81 subjects. FIG. 8A depicts a receiver operating characteristic curve (ROC) for each data block. Area under the curve (AUC) is the measure the ability of a binary classifier to distinguish between the two classes and ranges from 0 to 1; Higher AUC suggests it is a better classifier. The Clinical AUC is 0.7692; Metabolites AUC is 0.6346; Transcriptome AUC is 0.7727; Structural AUC is 0.6469; Functional AUC is 0.6573; Diffusion AUC is 0.4825. The Transcriptome and Clinical data blocks contributed the most to the classification. The Metabolome and Diffusion data blocks contributed the least. FIG. 8B depicts a confusion matrix of testing set. FIG. 8C shows that the overall model AUC is 0.7692 (BER=0.1836)

[0029] DETAILED DESCRIPTION OF THE INVENTION

[0030] The brain-gut-microbiome (BGM) system plays an influential role on mental health. BGM patterns related to resilience were characterized using fecal samples and multimodal MRI. Data integration analysis using latent components showed the high resilience phenotype was associated with lower depression and anxiety symptoms, higher frequency of bacterial transcriptomes (related to environmental adaptation, genetic propagation, energy metabolism, anti-inflammation), increased metabolites (N-acetylglutamate; dimethylglycine), and cortical signatures (increased resting state functional connectivity between reward circuits and sensorimotor networks; decreased grey matter volume and white matter tracts within the emotion regulation network). These findings support a multi-omic signature involving the BGM system suggesting that resilience impacts psychological symptoms, emotion regulation and cognitive function as reflected by unique neural correlates and microbiome function supporting eubiosis and gut barrier integrity. Bacterial transcriptomes provided the highest classification accuracy suggesting that the microbiome is critical in shaping resilience and highlights that microbiome modifications can optimize mental health.

[0031] Since no study to date has investigated an integrative biological profile of resilience, this work aimed to determine how resilience is related to clinical phenomes, microbiome function, and neural characteristics (FIGs. 1A-1E). Considering the microbiome’s role in influencing psychological resilience, it was hypothesized that high resilience would be associated with: (1) lower scores on clinical measures of psychological symptoms and higher levels of adaptive coping; (2) microbiome function supporting gut health as evidenced by characteristic bacterial transcriptomes (pathways supporting gut microbial growth and diversity) and metabolome (metabolites supporting anti-inflammation and gut barrier integrity); (3) brain morphometry and connectivity signatures, reflecting increased efficiency in regions important for emotion regulation and cognitive functioning. In certain aspects, the present disclosure provides methods for treating a neurological disorder in a subject, comprising administering a composition comprising bacteria of at least one genus selected from Prevotella, Haemophilus, Peptococcus, and Lactococcus, or a combination thereof, to the subject.

[0032] In some embodiments, the neurological disorder is anxiety. In certain embodiments, the neurological disorder is depression. In some embodiments, the neurological disorder is post-traumatic stress disorder (PTSD).

[0033] In certain aspects, the present disclosure provides methods for improving cognitive function in a subject, comprising administering a composition comprising bacteria of at least one genus selected from Prevotella, Haemophilus, Peptococcus, and Lactococcus, or a combination thereof, to the subject.

[0034] In certain aspects, the present disclosure provides methods for decreasing stress in a subject, comprising administering a composition comprising bacteria of at least one genus selected from Prevotella, Haemophilus, Peptococcus, and Lactococcus, or a combination thereof, to the subj ect.

[0035] In certain aspects, the present disclosure provides methods for increasing levels of short-chain fatty acids (SCFAs) in a subject, comprising administering a composition comprising bacteria of at least one genus selected from Prevotella, Haemophilus, Peptococcus, and Lactococcus, or a combination thereof, to the subject.

[0036] In certain embodiments, the bacteria are from the genus of Prevotella. In some embodiments, the bacteria are from the genus of Haemophilus. In some embodiments, the bacteria are from the genus of Peptococcus. In certain embodiments, the bacteria are from the genus of Lactococcus.

[0037] In some embodiments, the methods disclosed herein further comprise administering N-acetylglutamate to the subject.

[0038] In certain embodiments, the methods disclosed herein further comprise administering N-acetylglutamate, arginine (e.g., L-Arginine), or creatine to the subject.

[0039] In some embodiments, the methods disclosed herein further comprise administering creatine to the subject.

[0040] In certain embodiments, the compositions disclosed herein comprise bacteria of at least two genera selected from Prevotella, Haemophilus, Peptococcus, and Lactococcus. In some embodiments, the compositions disclosed herein comprise bacteria of at least three genera selected from Prevotella, Haemophilus, Peptococcus, and Lactococcus. In certain aspects, the present disclosure provides methods for treating a neurological disorder in a subject, comprising administering to the subject means for increasing in the subject an amount of bacteria of at least one genus selected from Prevotella, Haemophilus, Peptococcus, and Lactococcus, or a combination thereof.

[0041] In certain aspects, the present disclosure provides methods for treating a neurological disorder in a subject, comprising administering to the subject means for increasing in the subject levels of short-chain fatty acids (SCFAs).

[0042] In some embodiments, the neurological disorder is anxiety. In certain embodiments, the neurological disorder is depression. In some embodiments, the neurological disorder is post-traumatic stress disorder (PTSD).

[0043] In certain aspects, the present disclosure provides methods for improving cognitive function in a subject, comprising administering to the subject means for increasing in the subject an amount of bacteria of at least one genus selected from Prevotella, Haemophilus, Peptococcus, and Lactococcus, or a combination thereof.

[0044] In certain aspects, the present disclosure provides methods for decreasing stress in a subject, comprising administering to the subject means for increasing in the subject an amount of bacteria of at least one genus selected from Prevotella, Haemophilus, Peptococcus, and Lactococcus, or a combination thereof.

[0045] Bacterial Compositions

[0046] In certain aspects, provided herein are bacterial compositions comprising the genera Prevotella, Haemophilus, Peptococcus, and Lactococcus. In some embodiments, the bacterial formulation comprises a bacterium and / or a combination of bacteria described herein and a pharmaceutically acceptable carrier.

[0047] In certain embodiments, at least 0.1%, 0.5%, 1%, 2%, 3%, 4%, 5%, 6%, 7%, 8%, 9%, 10%, 11%, 12%, 13%, 14%, 15%, 16%, 17%, 18%, 19%, 20%, 25%, 30%, 35%, 40%, 45%, 50%, 55%, 60%, 65%, 70%, 75%, 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98% or 99% of the bacteria in the bacterial composition are selected from the genera Prevotella, Haemophilus, Peptococcus, and Lactococcus, or a combination thereof. In certain embodiments, substantially all of the bacteria in the bacterial composition are selected from the genera Prevotella, Haemophilus, Peptococcus, and Lactococcus, or a combination thereof. In certain embodiments, the bacterial composition comprises at least 10 colony forming units (CFUs), 100 colony forming units (CFUs), 1 x 103colony forming units (CFUs), 1 x 104colony forming units (CFUs), 1 x 105colony forming units (CFUs), 5 x 105 colony forming units (CFUs), 1 x 106colony forming units (CFUs), 2 x 106colony forming units (CFUs), 3 x 106colony forming units (CFUs), 4 x 106colony forming units (CFUs), 5 x

[0048] 106colony forming units (CFUs), 6 x 106colony forming units (CFUs), 7 x 106colony forming units (CFUs), 8 x 106colony forming units (CFUs), 9 x 106colony forming units (CFUs), 1 x 107colony forming units (CFUs), 2 x 107colony forming units (CFUs), 3 x 107colony forming units (CFUs), 4 x 107colony forming units (CFUs), 5 x 107colony forming units (CFUs), 6 x 107colony forming units (CFUs), 7 x 107colony forming units (CFUs), 8 x

[0049] 107colony forming units (CFUs), 9 x 107colony forming units (CFUs), 1 x 108colony forming units (CFUs), 2 x 108colony forming units (CFUs), 3 x 108colony forming units (CFUs), 4 x 108colony forming units (CFUs), 5 x 108colony forming units (CFUs), 6 x 108colony forming units (CFUs), 7 x 108colony forming units (CFUs), 8 x 108colony forming units (CFUs), 9 x 108colony forming units (CFUs), 1 x 109colony forming units (CFUs), 5 x 109colony forming units (CFUs), 1 x IO10colony forming units (CFUs) 5 x IO10colony forming units (CFUs), 1 x 1011 colony forming units (CFUs) 5 x 1011 colony forming units (CFUs), 1 x 1012colony forming units (CFUs) 5 x 1012colony forming units (CFUs), 1 x 1013colony forming units (CFUs) of bacteria selected from the genera Prevotella, Haemophilus, Peptococcus, and Lactococcus.

[0050] In some preferred embodiments, the bacterial composition comprises 1 x 109to 1 x 1011colony forming units of bacteria selected from the genera of Prevotella, Haemophilus, Peptococcus, and Lactococcus, or a combination thereof. In some embodiments, the composition comprises about 15 x 109CFU of bacteria selected from the genera Prevotella, Haemophilus, Peptococcus, and Lactococcus, or a combination thereof

[0051] The selected dosage level will depend upon a variety of factors including the subject’s diet, the route of administration, the time of administration, the residence time of the particular microorganism being employed, the duration of the treatment, other drugs, compounds and / or materials used in combination with the particular composition employed, the age, sex, weight, condition, general health and prior medical history of the patient being treated, and like factors well known in the medical arts.

[0052] A physician or veterinarian having ordinary skill in the art can readily determine and prescribe the effective amount of the bacterial composition required. For example, the physician or veterinarian could prescribe and / or administer doses of the bacteria employed in the composition at levels lower than that required in order to achieve the desired therapeutic effect and gradually increase the dosage until the desired effect is achieved. In some embodiments, probiotic formulations containing bacteria selected from the genera Prevotella, Haemophilus, Peptococcus, and Lactococcus, or combinations thereof are provided as encapsulated, enteric coated, or powder forms, with doses ranging from 10 to 1011CFU (e.g., IO10CFU). In some embodiments, the capsule is enteric coated, e.g., for duodenal release at pH 5.5. In some embodiments, the composition comprises a powder of freeze-dried bacteria selected from the genera Prevotella, Haemophilus, Peptococcus, and Lactococcus, or combinations thereof which is deemed to have “Qualified Presumption of Safety” (QPS) status. In some embodiments, the composition is storage-stable at frozen or refrigerated temperature. As used herein, “stably stored” or “storage-stable” refer to a composition in which cells are able to withstand storage for extended periods of time (e.g., at least one month, or two, three, four, six, or twelve months or more) with a less than 95%, 90%, 80%, 70%, 60%, 50%, 40%, 30%, 20%, 15%, 10%, 5%, or 1% decrease in cell viability.

[0053] Methods for producing microbial compositions may include three main processing steps. The steps are organism banking, organism production, and preservation. In certain embodiments, a sample that contains an abundance bacteria selected from the genera Prevotella, Haemophilus, Peptococcus, and Lactococcus, or combinations thereof may be cultured by avoiding an isolation step.

[0054] For banking, bacteria selected from the genera Prevotella, Haemophilus, Peptococcus, and Lactococcus, or combinations thereof included in the microbial composition may be (1) isolated directly from a specimen or taken from a banked stock, (2) optionally cultured on a nutrient agar or broth that supports growth to generate viable biomass, and (3) the biomass optionally preserved in multiple aliquots in long-term storage.

[0055] In embodiments using a culturing step, the agar or broth may contain nutrients that provide essential elements and specific factors that enable growth. An example would be Brain Heart Infusion medium (BHI), a medium composed of 14.5 g / L Casein Peptone, 7.0 g / L Meat Peptone, 6.0 g / L Brain Heart Infusion Solids, 5.0 g / L sodium chloride, 2.5 g / L disodium phosphate, and 2.0 g / L dextrose in demineralized water at pH 7.4A variety of microbiological media and variations are well known in the art. Culture media can be added to the culture at the start, may be added during the culture, or may be intermi ttently / continuously flowed through the culture. The strains in the bacterial composition may be cultivated alone, as a subset of the microbial composition, or as an entire collection comprising the microbial composition. As an example, a first strain may be cultivated together with a second strain in a mixed continuous culture, at a dilution rate lower than the maximum growth rate of either cell to prevent the culture from washing out of the cultivation.

[0056] The inoculated culture is incubated under favorable conditions for a time sufficient to build biomass. For microbial compositions for human use this is often at 37°C temperature, pH, and other parameters with values similar to the normal human niche. The environment may be actively controlled, passively controlled (e.g., via buffers), or allowed to drift. For example, for anaerobic bacterial compositions, an anoxic / reducing environment may be employed. This can be accomplished by the addition of reducing agents such as cysteine to the broth, and / or stripping it of oxygen. As an example, a culture of a bacterial composition may be grown at 37°C, pH 7, in the medium above, pre-reduced with 1 g / L cysteine-HCl.

[0057] When the culture has generated sufficient biomass, it may be preserved for banking. The organisms may be placed into a chemical milieu that protects from freezing (adding ‘cryoprotectants’), drying (Tyoprotectants’), and / or osmotic shock (‘osmoprotectants’), dispensing into multiple (optionally identical) containers to create a uniform bank, and then treating the culture for preservation. Containers are generally impermeable and have closures that assure isolation from the environment. Cryopreservation treatment is accomplished by freezing a liquid at ultra-low temperatures (e.g., at or below -80°C). Dried preservation removes water from the culture by evaporation (in the case of spray drying or ‘cool drying’) or by sublimation (e.g., for freeze drying, spray freeze drying). Removal of water improves long-term microbial composition storage stability at temperatures elevated above cryogenic conditions. Microbial composition banking may be done by culturing and preserving the strains individually, or by mixing the strains together to create a combined bank. As an example of cryopreservation, a microbial composition culture may be harvested by centrifugation to pellet the cells from the culture medium, the supernatant decanted and replaced with fresh culture broth containing 15% glycerol. The culture can then be aliquoted into 1 mL cryotubes, sealed, and placed at -80°C for long-term viability retention. This procedure achieves acceptable viability upon recovery from frozen storage.

[0058] Microbial production may be conducted using similar culture steps to banking, including medium composition and culture conditions described above. It may be conducted at larger scales of operation, especially for clinical development or commercial production. At larger scales, there may be several subcultivations of the microbial composition prior to the final cultivation. At the end of cultivation, the culture is harvested to enable further formulation into a dosage form for administration. This can involve concentration, removal of undesirable medium components, and / or introduction into a chemical milieu that preserves the microbial composition and renders it acceptable for administration via the chosen route. The suspension can then be freeze-dried to a powder and titrated.

[0059] After drying, the powder may be blended to an appropriate potency, and mixed with other cultures and / or a filler such as microcrystalline cellulose for consistency and ease of handling, and the bacterial composition formulated as provided herein.

[0060] In certain aspects, provided are bacterial compositions for administration in subjects. In some embodiments, the bacterial compositions are combined with additional active and / or inactive materials in order to produce a final product, which may be in single dosage unit or in a multi-dose format.

[0061] In some embodiments, the composition comprises at least one carbohydrate. A “carbohydrate” refers to a sugar or polymer of sugars. The terms “saccharide,” “polysaccharide,” “carbohydrate,” and “oligosaccharide” may be used interchangeably. Most carbohydrates are aldehydes or ketones with many hydroxyl groups, usually one on each carbon atom of the molecule. Carbohydrates generally have the molecular formula CnH2n0n. A carbohydrate may be a monosaccharide, a disaccharide, tri saccharide, oligosaccharide, or polysaccharide. The most basic carbohydrate is a monosaccharide, such as glucose, sucrose, galactose, mannose, ribose, arabinose, xylose, and fructose. Disaccharides are two joined monosaccharides. Exemplary disaccharides include sucrose, maltose, cellobiose, and lactose. Typically, an oligosaccharide includes between three and six monosaccharide units (e.g., raffinose, stachyose), and polysaccharides include six or more monosaccharide units. Exemplary polysaccharides include starch, glycogen, and cellulose. Carbohydrates may contain modified saccharide units such as 2’ -deoxyribose wherein a hydroxyl group is removed, 2’ -fluororibose wherein a hydroxyl group is replaced with a fluorine, or N- acetylglucosamine, a nitrogen-containing form of glucose (e.g., 2 ’-fluororibose, deoxyribose, and hexose). Carbohydrates may exist in many different forms, for example, conformers, cyclic forms, acyclic forms, stereoisomers, tautomers, anomers, and isomers.

[0062] In some embodiments, the composition comprises at least one lipid. As used herein, a “lipid” includes fats, oils, triglycerides, cholesterol, phospholipids, fatty acids in any form including free fatty acids. Fats, oils, and fatty acids can be saturated, unsaturated (cis or trans) or partially unsaturated (cis or trans). In some embodiments the lipid comprises at least one fatty acid selected from lauric acid (12:0), myristic acid (14:0), palmitic acid (16:0), palmitoleic acid (16: 1), margaric acid (17:0), heptadecenoic acid (17: 1), stearic acid (18:0), oleic acid (18: 1), linoleic acid (18:2), linolenic acid (18:3), octadecatetraenoic acid (18:4), arachidic acid (20:0), eicosenoic acid (20: 1), eicosadienoic acid (20:2), eicosatetraenoic acid (20:4), eicosapentaenoic acid (20:5) (EP A), docosanoic acid (22:0), docosenoic acid (22: 1), docosapentaenoic acid (22:5), docosahexaenoic acid (22:6) (DHA), and tetracosanoic acid (24:0). In some embodiments the composition comprises at least one modified lipid, for example a lipid that has been modified by cooking.

[0063] In some embodiments, the composition comprises at least one supplemental mineral or mineral source. Examples of minerals include, without limitation: chloride, sodium, calcium, iron, chromium, copper, iodine, zinc, magnesium, manganese, molybdenum, phosphorus, potassium, and selenium. Suitable forms of any of the foregoing minerals include soluble mineral salts, slightly soluble mineral salts, insoluble mineral salts, chelated minerals, mineral complexes, non-reactive minerals such as carbonyl minerals, and reduced minerals, and combinations thereof.

[0064] In some embodiments, the composition comprises at least one supplemental vitamin. At least one vitamin can be fat-soluble or water-soluble vitamins. Suitable vitamins include but are not limited to vitamin C, vitamin A, vitamin E, vitamin B 12, vitamin K, riboflavin, niacin, vitamin D, vitamin B6, folic acid, pyridoxine, thiamine, pantothenic acid, and biotin. Suitable forms of any of the foregoing are salts of the vitamin, derivatives of the vitamin, compounds having the same or similar activity of the vitamin, and metabolites of the vitamin.

[0065] In some embodiments, the composition comprises an excipient. Non-limiting examples of suitable excipients include a buffering agent, a preservative, a stabilizer, a binder, a compaction agent, a lubricant, a dispersion enhancer, a disintegration agent, a flavoring agent, a sweetener, and a coloring agent.

[0066] In some embodiments, the excipient is a buffering agent. Non-limiting examples of suitable buffering agents include sodium citrate, magnesium carbonate, magnesium bicarbonate, calcium carbonate, and calcium bicarbonate.

[0067] In some embodiments, the excipient comprises a preservative. Non-limiting examples of suitable preservatives include antioxidants, such as alpha-tocopherol and ascorbate, and antimicrobials, such as parabens, chlorobutanol, and phenol.

[0068] In some embodiments, the composition comprises a binder as an excipient. Nonlimiting examples of suitable binders include starches, pregelatinized starches, gelatin, polyvinylpyrolidone, cellulose, methylcellulose, sodium carboxymethylcellulose, ethylcellulose, polyacrylamides, polyvinyloxoazolidone, polyvinylalcohols, C12-C18 fatty acid alcohol, polyethylene glycol, polyols, saccharides, oligosaccharides, and combinations thereof.

[0069] In some embodiments, the composition comprises a lubricant as an excipient. Nonlimiting examples of suitable lubricants include magnesium stearate, calcium stearate, zinc stearate, hydrogenated vegetable oils, sterotex, polyoxyethylene monostearate, talc, polyethyleneglycol, sodium benzoate, sodium lauryl sulfate, magnesium lauryl sulfate, and light mineral oil.

[0070] In some embodiments, the composition comprises a dispersion enhancer as an excipient. Non-limiting examples of suitable dispersants include starch, alginic acid, polyvinylpyrrolidones, guar gum, kaolin, bentonite, purified wood cellulose, sodium starch glycolate, isoamorphous silicate, and microcrystalline cellulose as high HLB emulsifier surfactants.

[0071] In some embodiments, the compositions of the present invention are combined with a carrier (e.g., a pharmaceutically acceptable carrier) which is physiologically compatible with the gastrointestinal tissue of the subject(s) to which it is administered. Carriers can be comprised of solid-based, dry materials for formulation into tablet, capsule, or powdered form; or the carrier can be comprised of liquid or gel -based materials for formulations into liquid or gel forms. The specific type of carrier, as well as the final formulation depends, in part, upon the selected route(s) of administration. The therapeutic composition of the present invention may also include a variety of carriers and / or binders. In some embodiments, the carrier is micro-crystalline cellulose (MCC) added in an amount sufficient to complete the one-gram dosage total weight. Carriers can be solid-based dry materials for formulations in tablet, capsule, or powdered form, and can be liquid or gel-based materials for formulations in liquid or gel forms, which forms depend, in part, upon the routes of administration. Typical carriers for dry formulations include, but are not limited to: trehalose, maltodextrin, rice flour, microcrystalline cellulose (MCC) magnesium stearate, inositol, FOS, GOS, dextrose, sucrose, and like carriers. Suitable liquid or gel-based carriers include but are not limited to: water and physiological salt solutions; urea; alcohols and derivatives (e.g., methanol, ethanol, propanol, butanol); glycols (e.g., ethylene glycol, propylene glycol, and the like). Preferably, water-based carriers possess a neutral pH value (i.e., approximately pH 7.0). Other carriers or agents for administering the compositions described herein are known in the art. In some embodiments, the composition comprises a disintegrant as an excipient. In some embodiments the disintegrant is a non-effervescent disintegrant. Non-limiting examples of suitable non-effervescent disintegrants include starches such as corn starch, potato starch, pregelatinized and modified starches thereof, sweeteners, clays, such as bentonite, microcrystalline cellulose, alginates, sodium starch glycolate, gums such as agar, guar, locust bean, karaya, pectin, and tragacanth. In some embodiments the disintegrant is an effervescent disintegrant. Non-limiting examples of suitable effervescent disintegrants include sodium bicarbonate in combination with citric acid, and sodium bicarbonate in combination with tartaric acid.

[0072] In some embodiments, the bacterial formulation comprises an enteric coating or micro encapsulation. In certain embodiments, the enteric coating or micro encapsulation improves targeting to a desired region of the gastrointestinal tract. For example, in certain embodiments, the bacterial composition comprises an enteric coating and / or microcapsules that dissolve at a pH associated with a particular region of the gastrointestinal tract. In some embodiments, the enteric coating and / or microcapsules dissolve at a pH of about 5.5 - 6.2 to release in the duodenum, at a pH value of about 7.2 - 7.5 to release in the ileum, and / or at a pH value of about 5.6 - 6.2 to release in the colon. Exemplary enteric coatings and microcapsules are described, for example, in U.S. Pat. Pub. No. 2016 / 0022592, which is hereby incorporated by reference in its entirety.

[0073] In some embodiments, the composition is a food product (e.g., a food or beverage) such as a health food or beverage, a food or beverage for infants, a food or beverage for pregnant women, athletes, senior citizens or other specified group, a functional food, a beverage, a food or beverage for specified health use, a dietary supplement, a food or beverage for patients, or an animal feed. Specific examples of the foods and beverages include various beverages such as juices, refreshing beverages, tea beverages, drink preparations, jelly beverages, and functional beverages; alcoholic beverages such as beers; carbohydrate-containing foods such as rice food products, noodles, breads, and pastas; paste products such as fish hams, sausages, paste products of seafood; retort pouch products such as curries, food dressed with a thick starchy sauces, and Chinese soups; soups; dairy products such as milk, dairy beverages, ice creams, cheeses, and yogurts; fermented products such as fermented soybean pastes, yogurts, fermented beverages, and pickles; bean products; various confectionery products, including biscuits, cookies, and the like, candies, chewing gums, gummies, cold desserts including jellies, cream caramels, and frozen desserts; instant foods such as instant soups and instant soy-bean soups; microwavable foods; and the like. Further, the examples also include health foods and beverages prepared in the forms of powders, granules, tablets, capsules, liquids, pastes, and jellies. The composition may be a fermented food product, such as, but not limited to, a fermented milk product. Non-limiting examples of fermented food products include kombucha, sauerkraut, pickles, miso, tempeh, natto, kimchi, raw cheese, and yogurt. The composition may also be a food additive, such as, but not limited to, an acidulent (e.g., vinegar). Food additives can be divided into several groups based on their effects. Non-limiting examples of food additives include acidulents (e.g., vinegar, citric acid, tartaric acid, malic acid, fumaric acid, and lactic acid), acidity regulators, anticaking agents, antifoaming agents, foaming agents, antioxidants (e.g., vitamin C), bulking agents (e.g., starch), food coloring, fortifying agents, color retention agents, emulsifiers, flavors and flavor enhancers (e.g., monosodium glutamate), flour treatment agents, glazing agents, humectants, tracer gas, preservatives, stabilizers, sweeteners, and thickeners.

[0074] In certain embodiments, the bacteria disclosed herein are administered in conjunction with a prebiotic to the subject. Prebiotics are carbohydrates which are generally indigestible by a host animal and are selectively fermented or metabolized by bacteria. Prebiotics may be short-chain carbohydrates (e.g., oligosaccharides) and / or simple sugars (e.g., mono- and disaccharides) and / or mucins (heavily glycosylated proteins) that alter the composition or metabolism of a microbiome in the host. The short chain carbohydrates are also referred to as oligosaccharides, and usually contain from 2 or 3 and up to 8, 9, 10, 15 or more sugar moieties. When prebiotics are introduced to a host, the prebiotics affect the bacteria within the host and do not directly affect the host. In certain aspects, a prebiotic composition can selectively stimulate the growth and / or activity of one of a limited number of bacteria in a host. Prebiotics include oligosaccharides such as fructooligosaccharides (FOS) (including inulin), galactooligosaccharides (GOS), trans-galactooligosaccharides, xylooligosaccharides (XOS), chitooligosaccharides (COS), soy oligosaccharides (e.g., stachyose and raffinose) gentiooligosaccharides, isomaltooligosaccharides, mannooligosaccharides, maltooligosaccharides and mannanoligosaccharides. Oligosaccharides are not necessarily single components and can be mixtures containing oligosaccharides with different degrees of oligomerization, sometimes including the parent disaccharide and the monomeric sugars. Various types of oligosaccharides are found as natural components in many common foods, including fruits, vegetables, milk, and honey. Specific examples of oligosaccharides are lactulose, lactosucrose, palatinose, glycosyl sucrose, guar gum, gum Arabic, tagalose, amylose, amylopectin, pectin, xylan, and cyclodextrins. Prebiotics may also be purified or chemically or enzymatically synthesized.

[0075] Definitions

[0076] Unless otherwise defined herein, scientific and technical terms used in this application shall have the meanings that are commonly understood by those of ordinary skill in the art. Generally, nomenclature used in connection with, and techniques of, chemistry, cell and tissue culture, molecular biology, cell and cancer biology, neurobiology, neurochemistry, virology, immunology, microbiology, pharmacology, genetics and protein and nucleic acid chemistry, described herein, are those well-known and commonly used in the art.

[0077] The methods and techniques of the present disclosure are generally performed, unless otherwise indicated, according to conventional methods well known in the art and as described in various general and more specific references that are cited and discussed throughout this specification. See, e.g. “Principles of Neural Science”, McGraw-Hill Medical, New York, N.Y. (2000); Motulsky, “Intuitive Biostatistics”, Oxford University Press, Inc. (1995); Lodish et al., “Molecular Cell Biology, 4th ed ”, W. H. Freeman & Co., New York (2000); Griffiths et al., “Introduction to Genetic Analysis, 7th ed ”, W. H. Freeman & Co., N.Y. (1999); and Gilbert et al., “Developmental Biology, 6th ed ”, Sinauer Associates, Inc., Sunderland, MA (2000).

[0078] Chemistry terms used herein, unless otherwise defined herein, are used according to conventional usage in the art, as exemplified by “The McGraw-Hill Dictionary of Chemical Terms”, Parker S., Ed., McGraw-Hill, San Francisco, C.A. (1985).

[0079] All of the above, and any other publications, patents and published patent applications referred to in this application are specifically incorporated by reference herein. In case of conflict, the present specification, including its specific definitions, will control.

[0080] The term “agent” is used herein to denote a chemical compound (such as an organic or inorganic compound, a mixture of chemical compounds), a biological macromolecule (such as a nucleic acid, an antibody, including parts thereof as well as humanized, chimeric and human antibodies and monoclonal antibodies, a protein or portion thereof, e.g., a peptide, a lipid, a carbohydrate), or an extract made from biological materials such as bacteria, plants, fungi, or animal (particularly mammalian) cells or tissues. Agents include, for example, agents whose structure is known, and those whose structure is not known. The ability of such agents to inhibit AR or promote AR degradation may render them suitable as “therapeutic agents” in the methods and compositions of this disclosure. A “patient,” “subject,” or “individual” are used interchangeably and refer to either a human or a non-human animal. These terms include mammals, such as humans, primates, livestock animals (including bovines, porcines, etc.), companion animals (e.g., canines, felines, etc.) and rodents (e.g., mice and rats).

[0081] “Treating” a condition or patient refers to taking steps to obtain beneficial or desired results, including clinical results. As used herein, and as well understood in the art, “treatment” is an approach for obtaining beneficial or desired results, including clinical results. Beneficial or desired clinical results can include, but are not limited to, alleviation or amelioration of one or more symptoms or conditions, diminishment of extent of disease, stabilized (i.e., not worsening) state of disease, preventing spread of disease, delay or slowing of disease progression, amelioration or palliation of the disease state, and remission (whether partial or total), whether detectable or undetectable. “Treatment” can also mean prolonging survival as compared to expected survival if not receiving treatment.

[0082] The term “preventing” is art-recognized, and when used in relation to a condition, such as a local recurrence (e.g., pain), a disease such as cancer, a syndrome complex such as heart failure or any other medical condition, is well understood in the art, and includes administration of a composition which reduces the frequency of, or delays the onset of, symptoms of a medical condition in a subject relative to a subject which does not receive the composition. Thus, prevention of cancer includes, for example, reducing the number of detectable cancerous growths in a population of patients receiving a prophylactic treatment relative to an untreated control population, and / or delaying the appearance of detectable cancerous growths in a treated population versus an untreated control population, e.g., by a statistically and / or clinically significant amount.

[0083] “Administering” or “administration of’ a substance, a compound or an agent to a subject can be carried out using one of a variety of methods known to those skilled in the art. For example, a compound or an agent can be administered, intravenously, arterially, intradermally, intramuscularly, intraperitoneally, subcutaneously, ocularly, sublingually, orally (by ingestion), intranasally (by inhalation), intraspinally, intracerebrally, and transdermally (by absorption, e.g., through a skin duct). A compound or agent can also appropriately be introduced by rechargeable or biodegradable polymeric devices or other devices, e.g., patches and pumps, or formulations, which provide for the extended, slow, or controlled release of the compound or agent. Administering can also be performed, for example, once, a plurality of times, and / or over one or more extended periods. Appropriate methods of administering a substance, a compound or an agent to a subject will also depend, for example, on the age and / or the physical condition of the subject and the chemical and biological properties of the compound or agent (e.g., solubility, digestibility, bioavailability, stability, and toxicity). In some embodiments, a compound or an agent is administered orally, e.g., to a subject by ingestion. In some embodiments, the orally administered compound or agent is in an extended release or slow-release formulation or administered using a device for such slow or extended release.

[0084] As used herein, the phrase “conjoint administration” refers to any form of administration of two or more different therapeutic agents such that the second agent is administered while the previously administered therapeutic agent is still effective in the body (e.g., the two agents are simultaneously effective in the patient, which may include synergistic effects of the two agents). For example, the different therapeutic compounds can be administered either in the same formulation or in separate formulations, either concomitantly or sequentially. Thus, an individual who receives such treatment can benefit from a combined effect of different therapeutic agents.

[0085] A “therapeutically effective amount” or a “therapeutically effective dose” of a drug or agent is an amount of a drug or an agent that, when administered to a subject will have the intended therapeutic effect. The full therapeutic effect does not necessarily occur by administration of one dose and may occur only after administration of a series of doses. Thus, a therapeutically effective amount may be administered in one or more administrations. The precise effective amount needed for a subject will depend upon, for example, the subject’s size, health and age, and the nature and extent of the condition being treated, such as cancer or MDS. The skilled worker can readily determine the effective amount for a given situation by routine experimentation.

[0086] The term "percent sequence identity" or "percent identity" between two polynucleotide or polypeptide sequences refers to the number of identical matched positions shared by the sequences over a comparison window, taking into account additions or deletions (i.e., gaps) that must be introduced for optimal alignment of the two sequences. A matched position is any position where an identical nucleotide or amino acid is presented in both the target and reference sequence. Gaps presented in the target sequence are not counted since gaps are not nucleotides or amino acids. Likewise, gaps presented in the reference sequence are not counted since target sequence nucleotides or amino acids are counted, not nucleotides or amino acids from the reference sequence. The percentage of sequence identity is calculated by determining the number of positions at which the identical amino-acid residue or nucleic acid base occurs in both sequences to yield the number of matched positions, dividing the number of matched positions by the total number of positions in the window of comparison and multiplying the result by 100 to yield the percentage of sequence identity. The comparison of sequences and determination of percent sequence identity between two sequences can be accomplished using readily available software programs. Suitable software programs are available from various sources, and for alignment of both protein and nucleotide sequences. One suitable program to determine percent sequence identity is bl2seq, part of the BLAST suite of program available from the U.S. government's National Center for Biotechnology Information BLAST web site (at world wide web at blast.ncbi.nlm.nih.gov). B12seq performs a comparison between two sequences using either the BLASTN or BLASTP algorithm. BLASTN is used to compare nucleic acid sequences, while BLASTP is used to compare amino acid sequences. Other suitable programs are, e.g., Needle, Stretcher, Water, or Matcher, part of the EMBOSS suite of bioinformatics programs and also available from the European Bioinformatics Institute (EBI) at world wide web at ebi.ac.uk / Tools / psa.

[0087] “Operational taxonomic units” and “OTU(s)” refer to a terminal leaf in a phylogenetic tree and is defined by a nucleic acid sequence, e.g., the entire genome, or a specific genetic sequence, and all sequences that share sequence identity to this nucleic acid sequence at the level of species. In some embodiments the specific genetic sequence may be the 16S sequence or a portion of the 16S sequence. In other embodiments, the entire genomes of two entities are sequenced and compared. In another embodiment, select regions such as multilocus sequence tags (MLST), specific genes, or sets of genes may be genetically compared. For 16S, OTUs that share > 97% average nucleotide identity across the entire 16S or some variable region of the 16S are considered the same OTU. See e.g., Claesson MJ, Wang Q, O’Sullivan O, Greene-Diniz R, Cole JR, Ross RP, and O’Toole PW. 2010. Comparison of two next-generation sequencing technologies for resolving highly complex microbiota composition using tandem variable 16S rRNA gene regions. Nucleic Acids Res 38: e200. Konstantinidis KT, Ramette A, and Tiedje JM. 2006. The bacterial species definition in the genomic era. Philos Trans R Soc Lond B Biol Sci 361 : 1929-1940. For complete genomes, MLSTs, specific genes, other than 16S, or sets of genes OTUs that share > 95% average nucleotide identity are considered the same OTU. See e.g., Achtman M, and Wagner M. 2008. Microbial diversity and the genetic nature of microbial species. Nat. Rev. Microbiol. 6: 431-440. Konstantinidis KT, Ramette A, and Tiedje JM. 2006. The bacterial species definition in the genomic era. Philos Trans R Soc Lond B Biol Sci 361 : 1929-1940. OTUs are frequently defined by comparing sequences between organisms. Generally, sequences with less than 95% sequence identity are not considered to form part of the same OTU. OTUs may also be characterized by any combination of nucleotide markers or genes, in particular highly conserved genes (e.g., “house-keeping” genes), or a combination thereof. Operational Taxonomic Units (OTUs) with taxonomic assignments made to, e.g., genus, species, and phylogenetic clade are provided herein.

[0088] As used herein, the terms “optional” or “optionally” mean that the subsequently described event or circumstance may occur or may not occur, and that the description includes instances where the event or circumstance occurs as well as instances in which it does not.

[0089] The term “modulate” as used herein includes the inhibition or suppression of a function or activity (such as cell proliferation) as well as the enhancement of a function or activity.

[0090] “Pharmaceutically acceptable salt” or “salt” is used herein to refer to an acid addition salt or a basic addition salt which is suitable for or compatible with the treatment of patients.

[0091] “CFU” is a term well-known in the art and refers to a colony forming unit of bacteria. As used herein the term “healthy weight” refers to an individual with a body mass index between 18.5 and 24.9.

[0092] EXAMPLES

[0093] The invention now being generally described, it will be more readily understood by reference to the following examples which are included merely for purposes of illustration of certain aspects and embodiments of the present invention and are not intended to limit the invention.

[0094] Example 1: Preparation and administration of bacterial compositions

[0095] Results

[0096] Demographics and clinical results

[0097] Comparisons of clinical variables between the high resilience group (HR) and low resilience group (LR) are depicted in Table 1. The distribution of BMI that justifies the threshold for resilience group labels is illustrated in FIG. 5.

[0098] Microbiome analysis on group differences The alpha and beta diversity did not show significant differences between HR and LR (FIGs. 6 and 7). Performance of on the 16S-identified amplicon sequence variants (ASVs) dataset was not found to be a significant contributor in the DIABLO model (AUC=0.524).

[0099] DIABLO identifies a multi-omic signature able to classify high and low resilience DIABLO Model Performance

[0100] DIABLO identified a highly correlated ‘omics signature capable of discriminating between individuals with high versus low psychological resilience. A receiver operating characteristic (ROC) curve showing the performance of the classification model at all classification thresholds had an area under the ROC Curve (AUC; overall model) of 0.77 and the BER was 0.18. The training set AUC was 68% and the achieved overall testing set (N = 81) AUC was 68%. The training set (N=35) BER was 0.18. Area under the curve by dataset type was 77% for clinical, 63% for metabolome, 77% for transcriptome, 65% for structural MRI, 66% for resting-state functional MRI, and 48% for diffusion MRI. A summary of the final model’s performance is in FIG. 8.

[0101] DIABLO-selected variables

[0102] The standardized values of omics variables selected by the DIABLO model are depicted by resilience group in FIG. 2. A total of 45 features, 13 clinical features, including survey sub-scores, 3 metabolome, 16 transcriptome, 6 structural MRI, 5 resting-state functional MRI, and 2 diffusion MRI variables were included in the final model and depicted in the loading plots of FIG. 3. Variables from each dataset type are listed below in order of importance based on their loading vectors, which represent the magnitude of contribution to the DIABLO model. Higher loading vectors suggest greater importance in predicting resilience.

[0103] Clinical variables in order of importance included IPIP Neuroticism, STAI Anxiety, HAD Anxiety, FFM Total Score and the Describing FFM subscale, MASQ Verbal Memory, PSS Score, IPIP Extraversion, MASQ language, HAD Depression, FFM Non-judgement subscale, MASQ Visual Perception and Attention subscores. HR had higher mean levels of extraversion and mindfulness than their LR counterparts. Further, HR had lower mean levels of anxiety (HADS; STAI), neuroticism, perceived stress, and difficulties with verbal memory, visual perception, language, and attention than LR.

[0104] Metabolome variables in order of importance included N-acetylglutamate (NAG), dimethylglycine, and creatine. HR had higher mean levels of NAG and dimethylglycine, and similar mean levels of creatine compared to LR. Bacterial transcriptomes in order of importance are depicted in FIG. 3 and their functions are listed in FIG. IE. In brief, mean levels of transcriptomes linked to environmental adaptation, genetic propagation, metabolism, and anti-inflammation (i.e., increased frequency of a transcriptome involved in SCFA production) were higher in HR compared to LR.

[0105] Structural MRI features in order of importance included right subcallosal gyrus (SbCaG) volume, SbCaG surface area (SA), left angular gyrus (AngG) volume, AngG SA, right inferior part of the precentral sulcus (InfPrCS) SA, and InfPrCS volume. The restingstate functional MRI features in order of importance included connectivity between the ventral tegmental area (VTA) and right thalamus, between the left mesencephalic reticular formation (MRF) and the right supramarginal gyrus(SuMarG), between the right inferior segment of the circular sulcus of the Insula (InfCirlns) and the left pallidum, between the right subcentral gyrus and sulci (SbCG S) and the left superior parietal lobule (SupPL), and between the right lateral orbital sulcus (LORs) and the left putamen. Diffusion MRI features in order of importance included connections between the right and left SbCaG, and between the right lateral orbital gyrus (SupOcG) and right hippocampus. HR had lower mean levels of all DIABLO-selected structural MRI features but higher mean levels of all resting-state functional MRI features than LR. Regarding diffusion MRI features, HR had lower mean levels in bilateral SbCaG connections but higher connections between the right SupOcG and right hippocampus.

[0106] There were also associations between different classes of DIABLO-selected omics variables that were unique to HR. For example, anxiety (HADS) was negatively associated with a transcriptome involved in SCFA production. Although there were positive associations between NAG and transcriptomes involved in anti-inflammatory response and environmental adaptation in both HR and LR, only HR had an additional positive association between NAG and a transcriptome involved in genetic propagation.

[0107] The entire model is depicted in the connectogram (FIG. 4), where mean levels for each feature are visualized by resilience group in the data blocks forming the circle, and correlations between datasets are depicted inside the circle. Connectograms are built on a similarity matrix and represent the correlation between variables from different datasets. A cutoff was chosen as r > 0.55 as this is universally considered a “moderate” correlation.

[0108] Spearman correlations

[0109] Spearman correlations revealed two CD-RISC factors most robustly associated with DIABLO selected variables: control (sense of control and purpose in life, knowledge of where to turn for help) and persistence (personal competence, high standards, and tenacity; see Table 2). Additionally, Spearman’s rank correlations were performed to assess relations between all ASV’s on the genus level and DIABLO selected metabolites. Results showed that NAG was negatively associated with the genus Bacteroides rs (114) =-0.32, p=.004, q=.O2.

[0110] Discussion

[0111] This study shows that several key BGM markers distinguish HR from LR. HR exhibits adaptive psychological features, microbiome function facilitating gut health, and neurological signatures supporting emotion regulation and cognitive-emotional connections. Notably, among all data blocks, bacterial transcriptomes most strongly differentiated high from low resilience phenotypes. Furthermore, these findings suggest resilient individuals, particularly those demonstrating tenacity and perceived ability to control life outcomes, possess a microbiome that supports gut barrier integrity and eubiosis and a cortical signature that reflects adaptive emotional and cognitive regulation.

[0112] Association of resilience with clinical symptoms

[0113] In addition to higher levels of extraversion, HR was associated with lower scores of depression, anxiety, perceived stress, and neuroticism, which is consistent with previous literature. HR was also higher in trait-like mindfulness, particularly in abilities to express emotions in words, be non-judgmental, and express empathy, which are characteristics linked to increased resilience to stress. Additionally, HR was related to better self-reported cognitive abilities including verbal fluency, verbal memory, comprehension, visual memory, and sustained attention. Such cognitive skills have been associated with coping ability. Collectively, these findings suggest low resilience individuals may deplete psychological and cognitive resources when confronted with stressful events while resilient individuals may instead reappraise stressful events to promote advantageous psychological outcomes and coping.

[0114] Association of resilience with gut transcriptomics and metabolites

[0115] The functional categories of transcriptomes associated with HR included adaptive response to environmental changes, genetic propagation, energy metabolism, and antiinflammatory response. HR exhibited increased frequency of pathways that improve gut bacteria adaptation in unfavorable conditions (e.g., uptake of essential nutrients or extrusion of toxic substances; adjustment to fluctuations in pH and osmotic changes; communication within cellular community to adjust gene expression accordingly and cooperatively behave in virulence regulation, resource utilization, or even antibiotic resistance). In support of bacterial adaptation, HR also exhibited increased frequency of pathways facilitating genetic proliferation and pathways providing bacterial energy sources (e.g., carbohydrate metabolism). In addition, HR exhibited increased frequency of pathways related to antiinflammatory response; therefore, maintaining gut barrier integrity (e.g., metabolic degradation of lysine into SCFAs such as acetate and butyrate). See Table 3 for other transcriptomes associated with high resilience. These findings were consistent with animal studies showing that high resilience phenotypes are associated with increased frequency of pathways related to SCFA production, as well as carbohydrate metabolism and genetic information processing. Overall, HR possessed a microbiome that functions to maintain eubiosis and gut barrier integrity, thereby promoting intact communication between the gut and brain to optimize psychological functioning.

[0116] In the metabolome, HR was associated with higher levels of N-acetylglutamate (NAG) and dimethylglycine (DMG). Bacteria use NAG, derived from the amino acid glutamate, to synthesize arginine. NAG’s role in the BGM system is unclear. Increased levels of NAG may be secondary to endogenous metabolism or altered dietary intake, at least in healthy individuals. In the present study, increased levels of NAG in HR were associated with higher frequency of metabolic pathways related to environmental adaptation and antiinflammation. Increased levels of NAG may be associated with HR secondary to stimulating arginine synthesis which has anti-inflammatory effects in the gut. On the other hand, DMG is derived from dietary amino acid glycine and is used in methylation reactions crucial for various metabolic pathways, including energy metabolism, anti -oxi dative activity, and DNA synthesis and metabolism. DMG has also been shown to increase gut microbiota strains involved in anti-inflammatory response (e.g., SCFA production). See Table 3 for other metabolites associated with HR. These findings show the HR group may host a gut microbiome that can withstand perturbations as evidenced by an increase in metabolites that quell inflammation which, in turn, may support optimal neurological processes.

[0117] Association of resilience with multimodal brain signatures

[0118] Regarding brain morphometry, HR was associated with reduced SbCG grey matter volume and surface area. The SbCG, an important node in a network including the limbic system and thalamus, is strongly associated with cognitive-emotional processing and inhibiting fear responses. Similarly, reduced anterior cingulate cortex volume (subcomponents of which include the SbCG) was associated with PTSD remission after treatment completion, showing that volume reduction may indicate neuroplastic changes evidencing increased emotional regulation and extinction of maladaptive cognitive-emotional connections. Reduced SbCG volume was also positively associated with greater resilience subscale scores. These morphological signatures suggest dampened but adaptive reactivity towards acute emotional challenges and greater brain efficiency, highlighting that resilient individuals may require less cognitive effort in emotion and fear modulation. Gut microbiota alterations can also influence grey matter changes. Taken together, a resilient individual demonstrates microbiome functions that exert characteristic CNS changes, which then provide the physiological means for adaptive coping.

[0119] HR was associated with decreased anatomical connectivity involving the right and left SbCG, regions linked to dysphoric mood. Decreased anatomical connectivity between the bilateral SbCG regions may imply less activation of distressing emotions. In addition, HR was associated with decreased anatomical connectivity involving the hippocampus, a region together with the amygdala that is involved in fear and anxiety. Based on these findings, HR demonstrates a stress-resilient neurological signature.

[0120] Several functional connections were significant in differentiating HR vs LR, including the reward circuit (e.g., basal ganglia), sensorimotor network (SMN), DMN, and brainstem. HR had increased rsFC between the reward circuit (ventral tegmental area, pallidum) and SMN (thalamus, insula). Connections between these regions are related to fear-related motor neurocircuitry and thought to be involved in maladaptive fear emotions. Abnormal connectivity between these areas may be a vulnerability factor. For example, restricted range of emotion is a common symptom in stress-related mental health conditions like PTSD and depression, which highlights a possible link to reward circuitry alterations. Other brain networks associated with HR involved the DMN (supramarginal gyrus) and brainstem (MRF). Normal connectivity between the DMN and MRF may be a resilient characteristic given that self-monitoring (DMN) of threatening stimuli (MRF) facilitates readiness to react to stressors appropriately. The DMN may also play a role in resilience given its involvement in treatment recovery and “bouncing back” from trauma. DMN connectivity has also been linked to an abundance of certain gut microbiota; therefore, it is plausible the microbiome optimizes stress response by modulating relevant areas of brain connectivity. These findings highlight that a resilient individual demonstrates an intact BGM system that shapes connectivity within the CNS to allow for adaptive coping. Association of resilience with integrated brain-gut interactions

[0121] Although SCFAs were nonsignificant predictors of resilience within the DIABLO model, lower anxiety scores in HR were associated with higher frequency of a transcriptome involved in SCFA production. Thus, an individual’s level of resilience may modulate microbiome function to produce SCFAs and consequently mitigate stress-related conditions such as anxiety. Importantly, SCFAs are involved in maintaining intestinal barrier integrity and anti-inflammation. Lower anxiety scores in HR were also associated with a lower degree of neuroticism and perceived stress in life. In addition, increased frequency of certain bacterial transcriptomes in HR was associated with increased production of the bacterial metabolite, NAG. Such transcriptomic pathways may facilitate overall gut health by supporting increased production of NAG, a metabolite implicated in having antiinflammatory effects in the gut. Furthermore, resilient individuals may be able to mitigate development of stress-related psychopathology due to bacterial transcriptomes aiding in maintaining overall gut health.

[0122] Conclusion and Clinical Implications

[0123] This study is the first to identify a HR BGM phenotype and reveals promising pathways by which the onset and severity of stress-related psychiatric conditions might be prevented or mitigated. This cross-sectional study provides theoretical support for the development of longitudinal studies needed to establish causality. While the current study focuses on healthy samples to facilitate identification of potential factors that enhance and maintain health, future studies should consider comparing microbial profiles of healthy controls to those presenting with psychiatric diagnoses such as depression or PTSD. Future work should also consider metabolomics analysis of plasma samples and, importantly, vagal tone as it may mediate the relation between stress-resistance and microbial composition. Interestingly, the afferent vagus nerve can differentiate between pathogenic and non- pathogenic bacteria and can transmit signals that either exacerbate or mitigate stress responses depending on the bacterial stimulus making it a promising avenue for future inquiry. Some clinical implications to explore are whether dietary modifications, prebiotics, probiotics, or other clinical interventions (e.g., fecal transplantation) may improve coping and resilience to stress. Collectively, these findings support numerous avenues for novel inquiries and suggest that features of the brain and the gut microbiome work together to build stressresilience. Methods and Materials

[0124] Participants

[0125] A cohort of 116 healthy individuals (71 females) were recruited from the Los Angeles community through advertisements. Premenopausal women were included as determined by self-report of the last day of the previous menstrual cycle, and enrolled women were scanned during the follicular phase of the menstrual cycle. Participants were excluded if they had any major medical / neurological conditions, current or past psychiatric illnesses, gastroenterological issues, abdominal surgeries, substance use, tobacco dependence (half a pack or more daily), or metal implants (due to MRI contraindications); regularly used medications that interfere with the CNS; regularly used analgesics; were pregnant or breastfeeding; performed extreme strenuous exercise (> 8 hours of continuous exercise per week); weighed over 400 pounds; or used antibiotics or probiotics in the past 3 months. All procedures complied with institutional guidelines and were approved by the Institutional Review Board at UCLA’s Office of Protection for Research Subjects. All participants provided written informed consent.

[0126] Study design

[0127] In this cross-sectional study, all participants underwent multimodal MRI brain imaging, provided a stool sample within 2-3 days prior to scans, and answered questionnaires including detailed diet information during the week before the MRI scan (consistent with published studies). This study was a secondary data analysis on existing data and was pooled from two studies: (IRB#s 16-000281, 16-000187).

[0128] Questionnaires

[0129] Questionnaire data included the following:

[0130] Resilience was measured using the Connor-Davidson Resilience Scale (CD-RISC), which is a self-reported scale that consists of 25 items, evaluated on a five-point Likert scale ranging from 0-4; not true at all (0), rarely true (1), sometimes true (2), often true (3), and true nearly all the time (4) resulting in a number between 0-100 with higher scores indicating higher resilience. The resilience total score is comprised of 5 factors: 1) personal competence, high standards, and tenacity; 2) trust in one’s instincts, tolerance of negative affect, and strengthening effects of stress; 3) positive acceptance of change and secure relationships; 4) control; 5) spiritual influences. The mean resilience score of the general US population is 80.7. However, given our study’s participants having a median body mass index (BMI) of 28.06, we based the threshold for high and low resilience using the mean score (83.1) of a study involving mostly overweight patients (normally distributed data with Cronbach’s alpha of 0.92). In this study, a CD-RISC score > 83.1 was labeled as “High” resilience (N = 50) and < 83.1 was “Low” Resilience.

[0131] Other measures included BMI, socioeconomic status (SES), Early Trauma Inventory (ETI), Adverse Childhood Experiences (ACE) questionnaire, Hospital Anxiety and Depression Scale (HADS), Coping Strategies Questionnaire (CSQ), Perceived Stress Scale (PSS), State-Trait Anxiety Inventory (STAI), Positive and Negative Affect Schedule (PANAS), 12-item Short Form (SF12) Survey, International Physical Activity Questionnaires (IPAQ), Behavioral Inhibition System, Behavioral Approach System (BISBAS), Everyday Discrimination Score (EDS), Brief COPE (BCope), Patient Health Questionnaire (PHQ), Multiple Ability Self-Report Questionnaire (MASQ), Mindful Attention Awareness Scale (MAAS), Five Facet Mindfulness (FFM), Patient-Reported Outcomes Measurement Information System Sleep Scale (PROMIS_Sleep9), Visceral Sensitivity Index (VSI), Pain Vigilance and Awareness Questionnaire (PVAQ), Pain Catastrophizing Scale (PCS), Normal Personality Assessment (NEO), International Personality Pool (IP IP), Diet Questionnaire (used in our previous studies).

[0132] Gut Microbiome

[0133] The methods used for sample collection, processing and analysis are described in detail in published papers. Collection and storage

[0134] Participants were given “at-home kits” with specified instructions for when to collect and how to store their stool sample. Stool was collected 2-3 days before the MRI scan. 2-3 consecutive diet diaries were collected from the time of enrollment to the time of the MRI scan and stool collection (1-2 weekdays and 1 weekend). Participants were asked to collect the stool before the first meal of the day and immediately freeze the fresh stool after collection. If participants were on antidiarrheal or laxatives, they were asked to refrain from use for 2-3 days before the sample collection. Any deviations from the stool sample collection were documented to account for in the analysis. Fecal samples were stored at - 80°C, then ground to coarse powder by mortar and pestle under liquid nitrogen and aliquoted for nucleic acid extraction and metabolomic profiling.

[0135] Fecal microbial profiling

[0136] DNA extraction with bead beating was performed using the QIAGEN Powersoil DNA Isolation Kit (MO BIO Laboratories, Carlsbad, CA), following the manufacturer’s protocol. The V4 hypervariable region of the 16S rRNA gene was then amplified using 515F and 806R primers to generate a sequencing library according to a published protocol. The PCR products were purified with a commercial kit. The library underwent 2x250 sequencing on an Illumina HiSeq 2500 to a mean depth of 250,000 merged sequences per sample. The DADA2 pipeline was used for quality filtering, merging paired-end reads, removing chimera, and assigning taxonomy to each amplicon sequence variant (ASV) using the SILVA 138 reference database.

[0137] Alpha and Beta Diversity

[0138] Bacterial counts were derived using the DADA2 package vl. 29.0 and analyzed using R statistical software. To preserve statistical power, only bacterial genera that were present in at least 10% of the samples were included in analyses resulting in the exclusion of 146 genera and inclusion of 125 genera for analyses. Microbial alpha diversity was assessed on datasets rarefied to 10000 using the Shannon index to assess richness and evenness. Permutational analysis of variance analyses (PERMANOVA) of Bray-Curtis dissimilarities was used to quantify variation in genus between groups. Specifically, beta diversity was calculated using the adonis2 function in Vegan package 2.6-4 to determine Bray-Curtis distance matrices and conduct PERMANOVA significance testing for compositional data with 999 permutations. Fecal metabolomics processing

[0139] Aliquots of fecal samples were shipped to and processed by Metabolon, Inc and run as a single batch through their global HD4 Metabolomics platform, which involves running methanol-extracted samples through ultrahigh performance liquid chromatography -tandem mass spectroscopy under four separate chromatography and electrospray ionization conditions, separating the compounds by their chemical properties. The amount of missing data was low (<3%). However, missing values in the raw data were median imputed using median values, and ineffective peaks were dropped through interquartile range denoising, and an internal standardization normalization method was employed. The metabolites data block was compiled from the metabolite profiling results, and a 3 -dimensional matrix with metabolite numbers, sample names, and normalized peak intensities processed with the Metab oAnly st web software 3.0 (http: / / www.metaboanalyst.ca).

[0140] Fecal Transcriptomics Processing

[0141] Fecal samples were submitted to Viome Life Sciences, Inc., where RNA extraction, metatranscriptomics sequencing and annotation were conducted. A detailed description of these procedures is provided in previously published work. In sum, RNA extraction by beading was performed, DNA degraded by DNase, and 16S / 23S ribosomal RNA depleted by subtractive hybridization. Sequencing libraries were prepared from the resulting RNA and underwent 150x2 paired-end sequencing on Illumina NovaSeq. Taxonomy was assigned by aligning sequencing reads to a precomputed database of unique k-mers; 898 taxa were identified, including bacteria, fungi, viruses, and bacteriophages. Functional annotation was performed by aligning sequencing reads to the integrated gene catalog from the MetaHIT consortium then mapping these genes to the KEGG database; 5,896 distinct transcripts annotated as KOs were identified.

[0142] Multimodal neuroimaging

[0143] Magnetic resonance imaging acquisition

[0144] Each subject underwent imaging in a 3.0T Prisma MRI Scanner with a 20-channel head coil (Siemens Healthcare, Erlangen, Germany) for a high resolution T1 structural scan, a resting state functional scan, and a diffusion weighted scan. Participants were asked to fast for an average of 6 hours prior to scanning. The acquisition parameters are as follows: T1 weighted MP -RAGE scans acquired to assess brain structure (TR: 2300ms, TE: 2.98ms, TI: 900ms, flip angle: 9°, field of view: 240 x 256 mm, acquisition matrix: 240 x 256, slice thickness: 1 mm, voxel resolution: I x l x l mm). A 10-minute resting-state fMRI scan was acquired to assess resting-state functional connectivity (TR: 2000ms, TE: 28ms, flip angle: 77°, acquisition matrix: 64 x 64, slice thickness: 4 mm, voxel resolution: 3.44 x 3.44 x 4 mm, 300 volumes). A diffusion weighted image was acquired to assess white matter anatomical connectivity (64 noncollinear directions, b = 1000 s / mm2, 9 b = 0 s / mm2 images, TR: 9500ms, TE: 88ms, field of view: 2304 x 2304, acquisition matrix: 128 x 128, slice thickness: 2mm, spacing between slices: 2mm).

[0145] MRI preprocessing

[0146] Scans from each neuroimaging modality was considered as separate datasets and processed separately with the appropriate respective modality specific pipelines. Structural scans were processed and passed quality control using Statistical Parametric Mapping 12. Processing included motion correction, skull stripping, segmentation into gray matter, white matter, and cerebral spinal fluid (CSF), and normalization onto a NMI153 T1 template. Structural image processing

[0147] Cortical reconstruction and volumetric segmentation was done using the FreeSurfer6 analysis suite. All participants’ T1 structural data was first parcellated using the Destrieux cortical atlas and the Harvard-Oxford subcortical atlas and the Harvard Ascending Arousal Network (AAN). FreeSurfer was used to compute values of cortical thickness, surface area, mean curvature and volume for cortical ROIs and volume for subcortical ROIs. Multimodal scans were processed similar to previously published studies.

[0148] Functional image processing

[0149] Functional scans were preprocessed using the volume-based rs-FC analysis pipeline in the functional connectivity toolbox (CONN). All scans underwent realignment and unwarping, slice-timing correction, and outlier identification (advanced retrospective technique-based identification of outlier scans, ART) for scrubbing. Functional and structural data (T1 scans) were normalized and segmented into grey matter, white matter and CSF tissue. Denoising was done using ordinary least squares regression of potential confounding effects and temporal band-pass filtering. The default anatomical componentbased noise correction procedure (aCompCor) includes noise components from white matter, estimated subject-motion parameters, outlier scans or scrubbing based on frame-wise displacement, and effect of rest repressing potential ramping effects (at the start of the session). The influence of physiological, head-motion and other noise sources were minimized using a 0.008 Hz to 0.09 Hz after regression temporal band-pass filter. Fisher transformed correlations (Z) were computed between the functional time series of all the FreeSurfer parcellated regions to derive a 165x165 matrix for each participant. A single vector representing the correlation strength between each ROI pair was concatenated from the bottom half of the undirected matrix for each subject.

[0150] Diffusion image processing

[0151] Diffusion-weighted images, corrected for eddy current-induced distortions and movement with FSL’s eddy correct tool, along with the associated b-vectors and v-values were converted into Camino data formats with Camino’s fsl2scheme and image2voxel. Weighted linear least squares regression was used to fit a diffusion tensor on the voxel order data in Camino wdtfif). Deterministic and probabilistic tensor-based approaches have shown to have similar performance. The track command in Camino Euler algorithm performed whole-brain deterministic tractography with a step size of 0.5 and curve threshold of 76. Connectivity matrices were constructed using the conmat command in Camino and produced a 165x165 matrix that represents the number of streamlines connecting each ROI-to-ROI pair. Every subject’s matrix went through within-subject normalization by taking a sum of all counts between each ROI, then dividing each pair’s count by that total. The bottom half of the undirected matrix was then concatenated into one vector for each subject representing every ROI pair.

[0152] Data Integration Analysis for Biomarker discover using Latent components (DIABLO) Approach Overview

[0153] Integrating multiple omics approaches is essential for a thorough understanding of stress resilience and its associated phenotypes. Studying multimodal brain imaging, microbiome, transcriptome, metabolome, and clinical / behavioral variables in isolation cannot provide a comprehensive biological insight. DIABLO facilitates employing integrative multi- omics analysis, examining the interplay between various datasets to unravel the intricacies of stress-resilient phenotypes.

[0154] DIABLO was used to accomplish the Integrative Multi-omics Analyses goal to elucidate the interactions among central (brain), peripheral (microbiome, metabolome), and clinical / behavioral markers linked to resilience phenotypes. This advanced multi-block integration strategy simultaneously models outcomes based on multiple data matrices, identifies key predictive variables, and unveils relationships between different dataset types. DIABLO calculates linear combinations (multi-omic signatures) maximally correlated with a specified outcome while performing variable selection, controlling for relevant covariates (e.g., gender, age, BMI), and correcting for multiple comparisons. The result is a yield of a minimal subset of variables associated with resilience outcomes. Additionally, a limited number of multi-omics signatures were identified distinguishing high resilience from low resilience in both training and test sets.

[0155] DIABLO Analysis

[0156] DIABLO with a supervised learning framework was conducted to determine a sparse subset of correlated phenotypic and behavioral features from the 6 high-dimensional input data blocks (Q) that predict resilience group given the 83.1 score threshold. Data was split into a 70% training (N = 81) and 30% testing set (N = 35). Training data is used to calculate the design matrix and train the model. DIABLO extends sparse generalized canonical correlation analysis, a generalization of partial least squares (PLS), to a supervised machine learning framework with scarcity constraints for variable selection. Pairwise sparse PLS (sPLS) models were run (e.g. clinical versus resting-state functional MRI; metabolome versus diffusion MRI) prior to the DIABLO analysis to gauge the overall correlation structure between the dataset types and guide the data integration in DIABLO. A weighted Q x Q design matrix, a DIABLO input that informs the integration process, was calculated taking the correlation of the first principal component of individual sPLS models between pairwise data blocks. The data-driven design matrix contains values from 0 to 1 and represents if and by how much each data block Q should be correlated to one another. The final model consists of a limited number of features across datatypes that show high correlation with one another, which gives insight to both which ‘omic types’ are relevant to the discriminatory process and how different ‘omic types,’ or datasets interact with one another.

[0157] Data Preparation

[0158] Following the independent processing of the six data blocks (clinical, microbiome, metabolite, morphological MRI, rs-FC MRI, DTI MRI), involved the examination of each block to eliminate variables not suitable for analysis. Due to the nature of the data integration algorithm, variables exhibiting near zero variance (NZV) were detected and excluded from further consideration. Regarding the clinical dataset, categorical variables of interest were replaced with dummy variables before identifying and 16 variables with NZV were eliminated. For the fecal metabolome dataset, 38 NZV variables were dropped before performing median normalization. All clinical and metabolome variables with 50% or more NA’s were removed.

[0159] Neuroimaging data preparation for DIABLO Measures of brain morphometry, resting state functional connectivity and anatomical connectivity were derived for each individual and pairwise ROI similar to previous published studies. Data from the three neuroimaging modalities was considered as separate datasets and went through their own specific preprocessing methods. To construct the structural MRI dataset, the observations that used the ascending arousal network (AAN) and Destrieux Harvard-Oxford atlas parcellations were merged and measures of surface area and volume were residualized by the estimated total intracranial volume (eTIV) to control for effects driven by brain size. The NZV features in the resting-state functional MRI and diffusion MRI dataset, which can consist of many variables with zeros (i.e., regions that do not share anatomic connections), were dropped. Data blocks for all three neuroimaging datasets were resilience group median imputed.

[0160] The final clinical dataset had 99 features, metabolome dataset had 714 features, transcriptome dataset had 1,424 features, structural MRI dataset had 626 features, restingstate functional MRI dataset had 15,753 features, and diffusion MRI dataset had 3,408 features. All six datasets were then scaled and centered separately by calculating mean and standard deviation of each vector, then “scaling” each element by subtracting the mean and dividing by the standard deviation within the DIABLO function call. These datasets were used in subsequent analyses.

[0161] Variable selection

[0162] An initial DIABLO model with 10 components and all features from each dataset was fit on the training subset and the global performance was assessed with 5-fold cross validation to identify the number of components that produces the lowest Balanced Error Rate (BER) with mahalanobis distance. One component produced the lowest BER. Mahalanobis distance was set as the distance metric parameter as it is robust at handling imbalanced groups. Manual tuning was then performed to determine the optimal number of variables for each data block for the selected number of components to obtain the lowest BER on the unseen test subset. The testing set’s performance was used to determine the final model. DIABLO outputs a set of components (i.e., latent variables), a set of loading vectors (i.e., coefficients assigned to each variable) and a subset of selected variables in each dataset associated with each component. The outputs are obtained by maximizing the covariance between a linear combination of X variables and Y labels and the magnitude of the loading vectors represents the importance of that variable in the model with higher values suggesting greater importance. Loading plots show the importance of each variable within each component. Connectograms represent the correlation between variables from each datatype selected by the final DIABLO model. Performance metrics, including a confusion matrix, BER, and area under the receiver operating characteristic curve (AUC) curve were calculated to determine how well the model performed for each data block and overall. AUC is a value ranging from 0 to 1 that summarizes the overall diagnostic accuracy of a model. 1 represents a perfectly accurate classification model, 0.5 represents a poor model with no discriminatory ability, and values between 0.7 and 0.8 are considered acceptable.

[0163] Spearman Correlations

[0164] Spearman correlations were conducted to calculate associations between the five factors that compose the total resilience score and DIABLO selected variables.

[0165] Table 1. Demographic, Clinical, Metabolome, Transcriptome, and Multimodal Brain Differences Between the High and Low Resilience Groups.

[0166]

[0167] Table 1. Demographic, Clinical, Metabolome, Transcriptome, and Multimodal Brain Differences Between the High and Low Resilience Groups. Demographic, clinical, metabolic, transcriptomic, and neurological characteristics of high and low resilience groups. Abbreviations: BMI, body mass index; CD-RISC, Connor-Davidson Resilience Scale; MASQ, Multiple Ability Self-Report Questionnaire; FFM, Five Facet Mindfulness; HADS, Hospital Anxiety and Depression scale; IPIP, International Personality Item Pool; PSS, Perceived Stress Scale; STAI, Stat-Trait Anxiety Inventory; K07335, basic membrane protein A and related proteins; K03315, Na / H antiporter; K01081, 5 ’-nucleotidase; K07080, uncharacterized protein; K14645, serine protease; K02343, DNA polymerase III subunit gamma / tau; K01596, phosphoenolpyruvate carboxykinase; K01361, lactocepin; K01844, beta-lysine 5,6-aminomutase alpha subunit; K03388, heterodisulfide reductase subunit A2; K03466, DNA segregation ATPase FtsK / SpoIIIE; K02965, small subunit ribosomal protein S19; K01119, 2',3'-cyclic-nucleotide 2'-phosphodiesterase / 3 '-nucleotidase; K01834, 2,3- bisphosphoglycerate-dependent phosphoglycerate mutase; K05516, curved DNA-binding protein; K07667, two-component system, OmpR family, KDP operon response regulator KdpE. R, right; L, left; SA, surface area; Vol, volume; SbCaG, Subcallosal Gyrus; AngG, Angular Gyrus; InfPrCS, Inferior part of the Precentral Sulcus; VTA, Ventral Tegmental Area; Tha, Thalamus proper; MRF, Mesencephalic Reticular Formation; SuMarG, Supramarginal Gyrus; InfCirlns, Inferior segment of the Circular Sulcus of the Insula; Pal, Pallidum; SbCG S, Subcentral Gyrus and Sulci; SupPL, Superior Parietal Lobule; LORs, Lateral Orbital Sulcus; Pu, Putamen; SupOcG, Superior Occipital Gyrus; Hip, Hippocampus.^ / N (%); Mean (standard deviation)2Pearson's Chi-squared test; Welch Two Sample t-test. Table 2, Correlations of Participant Characteristics and Omics Variables Selected by the DIABLO Model with CD-RISC Sub-scales

[0168] Table 2, Cont’d.

[0169] Table 2. Correlations of Participant Characteristics and Omics Variables Selected by the DIABLO Model with CD-RISC Sub-scales. Spearman’s correlations of participant demographics and DIABLO selected variables with Total CD-RISC score, and Adaptability, Control, Meaning, Persistence, and Emotional Cognitive control sub-scales. Adaptability represents ability to bounce back and relates to the positive acceptance of change and secure relationships. Control meaning refers to individual’s perception of control over their life circumstances. Higher scores represent having a sense of control and a tendency to view challenges as manageable and belief that one has some influence over outcomes which enhance resilience by promoting problem-solving and adaptive coping strategies. Meaning relates to spiritual influences. Persistence refers to a sense of self-efficacy and reflects a sense of personal competence and high standards. Emotional Cognitive control scores control under pressure, trust in one’s instincts and tolerance of negative affects. P- value < 0.05 (Significant) Abbreviations: BMI, body mass index; CD-RISC, Connor- Davidson Resilience Scale; FFM, Five Facet Mindfulness; HADS, Hospital Anxiety and Depression Scale; IP IP, International Personality Pool; MASQ, Multiple Ability Self-Report Questionnaire; PSS, Perceived Stress Scale; STAI, State-Trait Anxiety Inventory; K07335, basic membrane protein A and related proteins; K03315, Na / H antiporter; K01081, 5’- nucleotidase; K07080, uncharacterized protein; K14645, serine protease; K02343, DNA polymerase III subunit gamma / tau; K01596, phosphoenolpyruvate carboxykinase; K01361, lactocepin; K01844, beta-lysine 5,6-aminomutase alpha subunit; K03388, heterodisulfide reductase subunit A2; K03466, DNA segregation ATPase FtsK / SpoIIIE; K02965, small subunit ribosomal protein S 19; KOI 119, 2',3'-cyclic-nucleotide 2'-phosphodiesterase / 3'- nucleotidase; K01834, 2,3-bisphosphoglycerate-dependent phosphoglycerate mutase;

[0170] K05516, curved DNA-binding protein; K07667, two-component system, OmpR family, KDP operon response regulator KdpE; MRI, magnetic resonance imaging; L, left; R, right; SbCaG, subcallosal gyrus; AngG, angular gyrus; InfPrCS, inferior part of the precentral sulcus; VTA, ventral tegmental area; Tha, thalamus; MRF, mesencephalic reticular formation; SuMarG, supramarginal gyrus; InfCirlns, inferior segment of the circular sulcus of the insula; Pal, pallidum; SupPL, superior parietal lobule; LORs, lateral orbital sulcus; Pu, putamen; SupOcG, superior occipital gyrus; Hip, hippocampus.

[0171] Table 3, Loading Values of DIABLO-selected Variables

[0172] Table 3. Loading Values of DIABLO-selected Variables. Magnitude of the loading values represent the relative importance of each variable in the final model. Variables are grouped by data type and ranked from most to least important. Abbreviations: FFM, Five Facet Mindfulness; HADS, Hospital Anxiety and Depression Scale; IP IP, International Personality Pool; MASQ, Multiple Ability Self-Report Questionnaire; PSS, Perceived Stress Scale; STAI, State-Trait Anxiety Inventory; MRI, magnetic resonance imaging; L, left; R, right; SbCaG, subcallosal gyrus; AngG, angular gyrus; InfPrCS, inferior part of the precentral sulcus; VTA, ventral tegmental area; Tha, thalamus; MRF, mesencephalic reticular formation; SuMarG, supramarginal gyrus; InfCirlns, inferior segment of the circular sulcus of the insula; Pal, pallidum; SupPL, superior parietal lobule; LORs, lateral orbital sulcus; Pu, putamen; SupOcG, superior occipital gyrus; Hip, hippocampus

[0173] INCORPORATION BY REFERENCE

[0174] All publications and patents mentioned herein are hereby incorporated by reference in their entirety as if each individual publication or patent was specifically and individually indicated to be incorporated by reference. In case of conflict, the present application, including any definitions herein, will control.

[0175] EQUIVALENTS

[0176] While specific embodiments of the subject invention have been discussed, the above specification is illustrative and not restrictive. Many variations of the invention will become apparent to those skilled in the art upon review of this specification and the claims below. The full scope of the invention should be determined by reference to the claims, along with their full scope of equivalents, and the specification, along with such variations.

Claims

We claim:

1. A method for treating a neurological disorder in a subject, comprising administering a composition comprising bacteria of at least one genus selected from Prevotella, Haemophilus, Peptococcus, and Lactococcus, or a combination thereof, to the subject.

2. The method of claim 1, wherein the neurological disorder is anxiety.

3. The method of claim 1, wherein the neurological disorder is depression.

4. The method of claim 1, wherein the neurological disorder is post-traumatic stress disorder (PTSD).

5. A method for improving cognitive function in a subject, comprising administering a composition comprising bacteria of at least one genus selected from Prevotella, Haemophilus, Peptococcus, and Lactococcus, or a combination thereof, to the subject.

6. A method for decreasing stress in a subject, comprising administering a composition comprising bacteria of at least one genus selected from Prevotella, Haemophilus, Peptococcus, and Lactococcus, or a combination thereof, to the subject.

7. A method for increasing levels of short-chain fatty acids (SCFAs) in a subject, comprising administering a composition comprising bacteria of at least one genus selected from Prevotella, Haemophilus, Peptococcus, and Lactococcus, or a combination thereof, to the subject.

8. The method of any one of claims 1-7, wherein the bacteria are from the genus of Prevotella.

9. The method of any one of claims 1-8, wherein the bacteria are from the genus of Haemophilus.

10. The method of any one of claims 1-9, wherein the bacteria are from the genus of Peptococcus.

11. The method of any one of claims 1-10, wherein the bacteria are from the genus of Lactococcus.

12. The method of any one of claims 1-11, wherein the method further comprises administering N-acetylglutamate to the subject.

13. The method of any one of claims 1-12, wherein the method further comprises administering arginine (e.g., L- Arginine) to the subject.

14. The method of any one of claims 1-11, wherein the method further comprises administering creatine to the subject.

15. The method of any one of claims 1-14, wherein the composition comprises bacteria of at least two genera selected from Prevotella, Haemophilus, Peptococcus, and Lactococcus.

16. The method of any one of claims 1-15, wherein the composition comprises bacteria of at least three genera selected from Prevotella, Haemophilus, Peptococcus, and Lactococcus.

17. A method for treating a neurological disorder in a subject, comprising administering to the subject means for increasing in the subject an amount of bacteria of at least one genus selected from Prevotella, Haemophilus, Peptococcus, and Lactococcus, or a combination thereof.

18. A method for treating a neurological disorder in a subject, comprising administering to the subject means for increasing in the subject levels of short-chain fatty acids (SCFAs).

19. The method of claim 17 or 18, wherein the neurological disorder is anxiety.

20. The method of claim 17 or 18, wherein the neurological disorder is depression.

21. The method of claim 17 or 18, wherein the neurological disorder is post-traumatic stress disorder (PTSD).

22. A method for improving cognitive function in a subject, comprising administering to the subject means for increasing in the subject an amount of bacteria of at least one genusselected from Prevotella, Haemophilus, Peptococcus, and Lactococcus, or a combination thereof.

23. A method for decreasing stress in a subject, comprising administering to the subject means for increasing in the subject an amount of bacteria of at least one genus selected from Prevotella, Haemophilus, Peptococcus, and Lactococcus, or a combination thereof.

Images