RELATED APPLICATIONS

This application claims priority under 35 U.S.C. § 119(e) to U.S. Provisional

Application No. 62/816,021, filed March 8, 2019, and entitled“Human Gut Microbiome- Derived Biosynthetic Enzymes for Production of Fatty Acid Amides,” which is incorporated herein by reference in its entirety for all purposes.

BACKGROUND OF THE INVENTION

Fatty acid amides (FAAs) represent a class of compounds produced by humans and other mammals that have been implicated as important signaling molecules. ¹ FAAs comprise a hydrophobic acyl tail group and a head group and these groups may contribute to the bioactivity of FAAs. ² The first FAA identified was arachidonoyl ethanolamide (anandamide), a linear hydrophobic FAA that activates the CBi and CB2 endocannabinoid G protein coupled receptors (GPCRs). ³ It is present at low levels with short half-life due to hydrolysis by the enzyme fatty acid amide hydrolase (FAAH). ⁴ Nonetheless, it has been suggested to induce motivation and pleasure via CBi activation, ³ suppress immune function via CB2 activation, ⁵ and promote obesity via an unknown mechanism. ⁶ Since then, a second family of FAAs has been found with fatty acids linked to monoamine neurotransmitters. One member is oleoyl dopamine, a nanomolar activator of vanilloid receptor 1 involved in sensing pain, ⁷ which has also been proposed to act via other mechanisms including CBi agonism, ⁷ FAAH antagonism, ⁷ calcium channel antagonism, ⁸ glucose homeostasis receptor agonism, ⁹ dopamine transporter antagonism, ¹⁰ and a as general carrier of dopamine across the blood-brain barrier. ¹¹ Other examples include docosahexaenoyl serotonin that attenuates cytokine production ¹² and arachidonoyl g-aminobutryic acid (GABA) that suppresses pain and motor activity in mice via an uncharacterized mechanism. ¹³ Moreover, a third family of FAAs with amino acid conjugates is known, and includes arachidonoyl serine which shows vascular modulation activity via interaction with a noncannabinoid GPCR receptor. ¹⁴ Although the primary biological roles for these FAAs are still largely left for investigation, they have been shown to target diverse molecular targets to control various physiopathological conditions. Given the diverse molecular targets of FAAs, efficient and high throughput methods of producing FAAs are needed. SUMMARY OF THE INVENTION

Aspects of the present disclosure relate to methods of producing fatty acid amides, comprising contacting a composition, comprising one or more exogenous fatty acids and one or more amines, with a set of biosynthetic enzymes which are human gut microbiome-derived clostridia biosynthetic enzymes in an effective amount to produce a fatty acid amide.

In some embodiments, the biosynthetic enzymes comprise a fatty-acyl transferase. In some embodiments, the biosynthetic enzymes comprise an acyl carrier protein (ACP). In some embodiments, the biosynthetic enzymes comprise a fatty acyl- ACP synthetase. In some embodiments, the method is performed in vitro. In some embodiments, the composition is a culture broth. In some embodiments, the set of biosynthetic enzymes are isolated from human gut microbiome-derived clostridia. In some embodiments, the method further comprises isolating acyl carrier protein from a human gut microbiome-derived clostridia. In some embodiments, the biosynthetic enzymes comprise a hydrolase. In some embodiments, the biosynthetic enzymes comprise a lipid transfer protein. In some embodiments, the biosynthetic enzymes comprise a glycosyltransferase.

In some embodiments, the set of biosynthetic enzymes are synthetic proteins. In some embodiments, the set of biosynthetic enzymes are recombinant proteins expressed from one or more vectors. In some embodiments, the set of biosynthetic enzymes are recombinant proteins expressed from a vector comprising a single operon, wherein the operon comprises genes encoding at least a fatty acyl transferase, an acyl carrier protein, and a fatty acyl- ACP synthetase. In some embodiments, the set of biosynthetic enzymes are recombinant proteins expressed from a plurality of vectors comprising a first vector, a second vector, and a third vector, wherein each of the plurality of vectors comprises a gene in a single operon and under the control of a regulatory element.

In some embodiments, the one or more exogenous fatty acids are selected from a group consisting of acetic acid, propionic acid, butyric acid, valeric acid, hexanoic acid, octanoic acid, capric acid, lauric acid, myristic acid, pentadecanoic acid, palmitic acid, stearic acid, arachidic acid, iso-pentadecanoic acid, palmitoleic acid, oleic acid, vaccenic acid, linoleic acid, g-linolenic acid, a-linolenic acid, dihomo-y-linolenic acid, arachidonic acid, eicosapentaenoic acid, docosapentaenoic acid, docosahexaenoic acid, 8-methyl-6-nonenoic acid, octynoic acid, myristic acid alkyne, and palmitic acid alkyne. In some embodiments, the one or more exogenous amines are selected from a group consisting of phenylalanine, tryptophan, tyrosine, histidine, lysine, glycine, alanine, valine, leucine, isoleucine, methionine, proline, serine, threonine, cysteine, asparagine, glutamine, aspartic acid, glutamic acid, arginine, ornithine, b-alanine, L-DOPA, creatine, citmlline, phenylacetylglutamine, phenylethylamine, tryptamine, tyramine, histamine, serotonin, dopamine, epinephrine, norepinephrine, g-aminobutryic acid (GABA), aminovaleric acid, ethanolamine, cadaverine, putrescine, spermine, spermidine, agmatine, propylamine, butylamine, dimethylamine, pyrollidine, piperidine, homocysteine, cysteamine,

homocysteamine, taurine, hypotaurine, glutathione, octopamine, 3-iodothyronamine, melatonin, and vanillylamide.

Further aspects of the present disclosure relate to methods of producing a fatty acid amide, comprising: (i) administering to a host a plurality of vectors comprising a first vector, a second vector, and a third vector, wherein each of the plurality of vectors comprises a human gut microbiome-derived bacterium gene under the control of a regulatory element; and wherein the gene on the first vector, the second vector and the third vector encode a fatty acyl transferase, an acyl carrier protein, and fatty acyl-ACP synthetase, respectively; and (ii) contacting the host with a composition in an effective amount to produce a fatty acid amide in the host.

In some embodiments, the composition is a culture broth. In some embodiments, the gene is in a single operon. In some embodiments, the regulatory element is an inducible promoter, optionally wherein the inducible promoter is an IPTG-inducible promoter. In some embodiments, the regulatory element is a constitutive promoter. In some embodiments, the regulatory element is a repressible promoter, optionally, wherein the repressible promoter comprises a lac operon. In some embodiments, the regulatory element comprises a T7 promoter. In some embodiments, the method further comprises administering a vector encoding T7 RNA polymerase. In some embodiments, the vector further comprises a ribosomal binding site. In some embodiments, at least one gene is codon-optimized. In some embodiments, the gene is an open reading frame. In some embodiments, the host is an E. coli. In some embodiments, the gut microbiome-derived bacterium is from Clostridia. In some embodiments, the gene is codon-optimized for production in E. coli.

In some embodiments, the fatty acid amide is palmitoleyl-putrescine, oleoyl aminovaleric acid, a-linolenoyl aminovaleric acid, oleoyl g-aminobutyric acid, oleoyl dopamine, oleoyl tyramine, palmitoleoyl dopamine, a-linolenoyl tyramine, oleoyl phenethylamine, lauroyl tyrptamine, lineoleoyl tryptamine, lauroyl tyramine, a-linolenoyl homocysteine, oleoyl homocysteine, a-linolenoyl cysteine, linoleoyl homocysteine, oleoyl- aminopentanoic acid, or a-linolenoyl homocysteamine.

In some embodiments, the plurality of vectors further comprises a vector having a gene encoding Sfp 4'-phosphopantcthcinyl transferase. In some embodiments, the first vector, the second vector, or the third vector further comprises a gene encoding Sfp 4'- phosphopantetheinyl transferase.

In some embodiments, the composition does not comprise an exogenous fatty acid. In some embodiments, the composition does not comprise an exogenous amine. In some embodiments, the composition comprises one or more exogenous fatty acids selected from a group consisting of acetic acid, propionic acid, butyric acid, valeric acid, hexanoic acid, octanoic acid, capric acid, lauric acid, myristic acid, pentadecanoic acid, palmitic acid, stearic acid, arachidic acid, iso-pentadecanoic acid, palmitoleic acid, oleic acid, vaccenic acid, linoleic acid, g-linolenic acid, a-linolenic acid, dihomo-y-linolenic acid, arachidonic acid,

eicosapentaenoic acid, docosapentaenoic acid, docosahexaenoic acid, 8-methyl-6-nonenoic acid, octynoic acid, myristic acid alkyne, and palmitic acid alkyne.

In some embodiments, the composition comprises one or more exogenous amines selected from a group consisting of phenylalanine, tryptophan, tyrosine, histidine, lysine, glycine, alanine, valine, leucine, isoleucine, methionine, proline, serine, threonine, cysteine, asparagine, glutamine, aspartic acid, glutamic acid, arginine, ornithine, b-alanine, L-DOPA, creatine, citrulline, phenylacetylglutamine, phenylethylamine, tryptamine, tyramine, histamine, serotonin, dopamine, epinephrine, norepinephrine, g-aminobutryic acid (GABA), aminovaleric acid, ethanolamine, cadaverine, putrescine, spermine, spermidine, agmatine, propylamine, butylamine, dimethylamine, pyrollidine, piperidine, homocysteine, cysteamine,

homocysteamine, taurine, hypotaurine, glutathione, octopamine, 3-iodothyronamine, melatonin, and vanillylamide.

Further aspects of the present disclosure provide vectors for the production of fatty acid amides, comprising genes, wherein each of the genes is under the control of a regulatory element, wherein each of the genes is derived from gut microbiome-derived bacterium, and wherein the genes encode at least a fatty acyl transferase, an acyl carrier protein, and a fatty acyl-ACP synthetase. In some embodiments, the regulatory element is an inducible promoter. In some embodiments, the regulatory element is a constitutive promoter. In some

embodiments, the regulatory element is a repressible promoter. In some embodiments, the regulatory element comprises a T7 promoter. In some embodiments, the vector further encodes a T7 RNA polymerase. In some embodiments, the vector further comprises a ribosomal binding site.

In some embodiments, the genes are codon-optimized. In some embodiments, the genes are open reading frames. In some embodiments, the genes are codon-optimized for production in E. coli. In some embodiments, the gut microbiome-derived bacterium is from Clostridia. In some embodiments, the genes also encode a hydrolase. In some embodiments, the genes also encode a lipid transfer protein. In some embodiments, the genes also encode a glycosyltransferase. In some embodiments, the genes encode Sfp 4'-phosphopantcthcinyl transferase. In some embodiments, the genes are in a single operon.

Further aspects of the present disclosure provide methods of producing a fatty acid amide comprising: (i) administering to a host a vector comprising genes, wherein each of the genes is under the control of a regulatory element, wherein each of the genes is derived from a human gut microbiome-derived bacterium, and wherein the genes encode a fatty acyl transferase, an acyl carrier protein, and fatty acyl-ACP synthetase; and (ii) contacting the host with a composition in an effective amount to produce a fatty acid amide in the host.

In some embodiments, the composition is a culture broth. In some embodiments, the regulatory element is an inducible promoter. In some embodiments, the regulatory element is a constitutive promoter. In some embodiments, the regulatory element is a repressible promoter. In some embodiments, the regulatory element is a T7 promoter. In some embodiments, the vector further encodes a T7 RNA polymerase. In some embodiments, the vector further comprises a ribosomal binding site. In some embodiments, wherein the genes are codon- optimized. In some embodiments, the genes are open reading frames. In some embodiments, the host is E. coli. In some embodiments, the gut microbiome-derived bacterium is from Clostridia. In some embodiments, the genes are codon-optimized for production in E. coli.

In some embodiments, the fatty acid amide is palmitoleyl-putrescine, oleoyl aminovaleric acid, a-linolenoyl aminovaleric acid, oleoyl g-aminobutyric acid, oleoyl dopamine, oleoyl tyramine, palmitoleoyl dopamine, a-linolenoyl tyramine, oleoyl

phenethylamine, lauroyl tyrptamine, lineoleoyl tryptamine, lauroyl tyramine, a-linolenoyl homocysteine, oleoyl homocysteine, a-linolenoyl cysteine, linoleoyl homocysteine, oleoyl- aminopentanoic acid, or a-linolenoyl homocysteamine.

In some embodiments, the genes further comprise a gene encoding Sfp 4'- phosphopantetheinyl transferase. In some embodiments, the genes are in a single operon. In some embodiments, the composition does not comprise an exogenous fatty acid. In some embodiments, the composition does not comprise an exogenous amine.

In some embodiments, the composition comprises one or more exogenous fatty acids selected from a group consisting of acetic acid, propionic acid, butyric acid, valeric acid, hexanoic acid, octanoic acid, capric acid, lauric acid, myristic acid, pentadecanoic acid, palmitic acid, stearic acid, arachidic acid, iso-pentadecanoic acid, palmitoleic acid, oleic acid, vaccenic acid, linoleic acid, g-linolenic acid, a-linolenic acid, dihomo-y-linolenic acid, arachidonic acid, eicosapentaenoic acid, docosapentaenoic acid, docosahexaenoic acid, 8- methyl-6-nonenoic acid, octynoic acid, myristic acid alkyne, and palmitic acid alkyne.

In some embodiments, the composition comprises one or more exogenous amines selected from a group consisting of phenylalanine, tryptophan, tyrosine, histidine, lysine, glycine, alanine, valine, leucine, isoleucine, methionine, proline, serine, threonine, cysteine, asparagine, glutamine, aspartic acid, glutamic acid, arginine, ornithine, b-alanine, L-DOPA, creatine, citmlline, phenylacetylglutamine, phenylethylamine, tryptamine, tyramine, histamine, serotonin, dopamine, epinephrine, norepinephrine, g-aminobutryic acid (GABA), aminovaleric acid, ethanolamine, cadaverine, putrescine, spermine, spermidine, agmatine, propylamine, butylamine, dimethylamine, pyrollidine, piperidine, homocysteine, cysteamine,

homocysteamine, taurine, hypotaurine, glutathione, octopamine, 3-iodothyronamine, melatonin, and vanillylamide.

Further aspects of the present disclosure relate to engineered cells comprising any of the vectors described herein.

Further aspects of the present disclosure relate to methods of producing fatty acid amides, the method comprising: contacting a composition, comprising one or more exogenous fatty acids and one or more amines, with a set of biosynthetic enzymes which are human gut microbiome-derived clostridia biosynthetic enzymes in an effective amount to produce a fatty acid amide, wherein the set of biosynthetic enzymes comprise a fatty acyl transferase, an acyl carrier protein, and a fatty acyl-ACP synthetase.

The details of one or more embodiments of the invention are set forth in the description below. Other features or advantages of the present invention will be apparent from the following drawings and detailed description of several embodiments, and also from the appended claims. BRIEF DESCRIPTION OF THE DRAWINGS

The following drawings form part of the present specification and are included to further demonstrate certain aspects of the present disclosure, which can be better understood by reference to one or more of these drawings in combination with the detailed description of specific embodiments presented herein. For purposes of clarity, not every component may be labeled in every drawing. It is to be understood that the data illustrated in the drawings in no way limit the scope of the disclosure. In the drawings:

FIGS. 1A-1F show FAA pathways from human gut microbiome-derived Clostridia a, Computational workflow from DNA sequencing data to gut Clostridia NRPS pathways b, Arrow diagram of the eight NRPS pathways with the source Clostridia strain indicated on the left. Darker arrows = conserved biosynthetic genes with homology to condensation (C), thiolation (T), and adenylation (A) domains from NRPS. Fighter arrows = accessory genes c, MS chromatogram (EIC ESI+ m/z 325.3213, [M + H] ⁺ of palmitoeyl putrescine, as shown) of the in vivo extract of E. coli expressing different combinations of E. rectale biosynthetic genes and sfp gene. * denotes peak corresponding to the shown compound. The fatty acid olefin is in cis (Z) conformation d, Adenylation activity assay based on absorbance of pyrophosphate (PPi) fluorogenic dye (Abs36o), normalized to no adenylation enzyme control e, Condensation activity assay, based on MS ion intensity of the final product, normalized to product level with no condensation enzyme f, Biosynthetic scheme for FAA formation, where T is an acyl carrier protein, A is a fatty acyl-ACP synthetase, and C is a - fatty acyltransferase. Substrate labels for d and e: 06:1 = palmitoleic acid, 06: palmitic acid, Put = putrescine, Orn = ornithine,

Arg = argninine, Agm = agmatine, Cad = cadaverine, Fys = lysine, Phe = phenylalanine, Trp = tryptophan, Tyr = tyrosine, Pha = phenethylamine, Tpa = tryptamine, Tya = tyramine. All error bars are the s.d. of three replicates conducted on different days.

FIGS. 2A-2E show in vitro substrate screening assay a, Experimental workflow of the assay b, Adenylation activity with panel of fatty acid substrates, based on absorbance of tryptamine conjugated product (Abs2so). c, Condensation activity with panel of amine substrates, based on MS ion intensity of fatty acid conjugated product, normalized to product level with no condensation enzyme d, Panel of 25 fatty acids (labeled F1-F25) and 53 biogenic amines (labeled A1-A53) tested for substrate screening assay, with indication of the primary source in the human gut (bacteria, human, food). The fatty acid olefins are all in cis (Z) conformation e, Feft: Bar graphs representing adenylation and condensation enzyme activity on a panel of fatty acid and amine substrates, respectively, for the eight pathways. Error bars are the s.d. of three replicates conducted on different days. Right: Major compound from each pathway based on the fatty acid and amine substrate with highest activity level (and also surpasses 2 fold-difference threshold), as well as examples of minor compounds based on substrates with second highest activity level. The fatty acid olefins are all in cis (Z) conformation.

FIG.3 shows a pathway similarity network map. Network map of 346 human gut- derived and actively transcribed pathways, represented as nodes. Edges are calculated using BiG-SCAPE based on the protein sequence- and domain-level (dis)similarity of pairs of pathway with a cutoff of 0.75. Clostridia NRPS group is circled. Groups labeled 1 and 2 show saccharides, 3 and 4 show NRPS products, 5 shows PKS products, 6 and 7 show hybrid products, 8 and 9 show RiPP products, and 10 shows other products. FIG. 4 shows pathway phytogeny analysis. Left: A phylogenetic tree (MUSCLE alignment, Neighbor Joining, best tree, MacVector) based on the adenylation gene of 148 Clostridia FAA pathways. Eight representative HMP-derived pathways and their close homologs (>98% sequence identity) are labeled with the originating species. Right: A phylogenetic tree (phyloT) based on NCBI taxon ID of the bacterial strain harboring the pathway. The HMP-derived pathways are labeled according to species or the pathway to which they are homologous (“horn.”)· Pathways without unique NCBI taxon ID are not included.

1

FIGS. 5A-5C show structural elucidation of E. coli in vivo compound a, H NMR

(600 MHz) of major compound from E. rectale- derived pathway harboring E. coli BAP1.

1 1

b, Key spectral data in structure determination c, NMR data summary. H- H COSY show two coupling networks. The first network consists of four methylene systems with deshielded carbons (539.8, 42.7) at each end with one end connected to an amide proton system, suggestive of a 1,4-butanediamine substructure. The second network consists of two olefin, twelve methylene, and one methyl systems, which are correlated to form a hexadecene substructure (double bond position inferred based on its biosynthetic origin). The two networks are connected by a ¾- ¹³ C HMBC correlation from H6 and H7 to the same remaining carbonyl carbon (5177.1). The purified compound was found to be spectroscopically identical to an authentic standard that has been chemically synthesized based on the elucidated structure.

FIGS. 6A-6B show E. coli heterologous production a, MS chromatogram (EIC

+

ESI+ m/z 325.3213, [M + H] of palmitoeyl putrescine, as shown) of the in vivo extract of E. coli expressing the E. rectale biosynthetic genes (C, T, A), sfp gene, and with or without the 2 accessory genes encoding alpha/beta hydrolase and sterol transfer proteins. * denotes peak corresponding to the shown compound, palmitoeyl putrescine. b, MS chromatogram of the in vivo extract of E. rectale pathway-expressing E. coli with either tryptamine (Tpa) or octanoic acid (C8) fed into culture broth. Top: EIC ESI+ m/z 397.3213. Bottom: EIC ESI+ m/z 215.2118. * denotes peak corresponding to the shown product compound.

FIG. 7 shows in vitro compound production. MS chromatogram of the reaction in amine in vitro substrate panel assay at EIC ESI+ corresponding to m/z of the preferred product with highest enzymatic activity (1: 382.3316, 2: 418.3316, 3: 402.3367, 4:

343.2744, 5: 396.2567). Labels: standard = chemically synthesized authentic standard; C = in vitro extract with condensation enzyme; none = in vitro extract without condensation enzyme.

FIG. 8 shows inter-pathway domain swapping. Adenylation (left graph) and condensation activity (right graph) level and corresponding enzyme-catalyzed product formation (if any) for different combinations of biosynthetic proteins (C, T, A) from the C. eutactus (lighter shade) and E. rectale (darker shade) pathways. The fatty acid olefin is in cis (Z) conformation. Substrate labels: C12 = lauric acid, C18: 1 oleic acid, Tpa = tryptamine, Apa = aminovaleric acid. All error bars are the s.d. of three replicates conducted on different days.

FIGS. 9A-9B show structural homology of Clostridia and human FAAs. The major and minor compounds characterized from the Clostridia FAA pathways (left) in

comparison to known human FAAs (right), suggesting structural and functional mimicry.

FIGS. 10A-10C show expression vector design. Cloning into pET28a (Novagen) for constructs containing a, re-designed complete pathways, b, gene knockout pathways, and c, single gene expression for protein purification. Sequences of genes and vector components are inTables 1 and 2.

FIG. 11 A-l IB shows a substrate panel assay time course. Fatty acid (left; 25 substrates) and amine (right; 53 substrates) substrate panel assay for C. eutactus pathway proteins over the course of ten time points (10, 20, 30, 45, 60, 90, 120, 240, 480, and 960 min) a, For eight representative fatty acid (left) or amine (right) substrates, unnormalized data of product level with (unmarked curves) and without (curves marked with *) adenylation protein is shown (left), or product level with (unmarked curves) and without (curves marked with *) condensation protein is shown (right) b, For all 25 fatty acid (left) or 53 amine (right) substrates, normalized activity data is shown over the ten time points. The time point with the largest difference between the highest point and the rest is marked with an arrow. All error bars are the s.d. of three replicates conducted on different days.

FIG. 12 shows a non-limiting example of an acyl carrier protein (ACP or T protein) sequence logo. The sequence logo was created using Web Logo. See, e.g., Crooks el al, Genome Research, 14:1188-1190, (2004); Schneider et al., Nucleic Acids Res. 18:6097-6100 (1990).

FIGS. 13A-13D show a non-limiting example of a fatty acyl- ACP synthetase (A protein) sequence logo. The sequence logo was created using Web Logo. See, e.g., Crooks et al, Genome Research, 14:1188-1190, (2004); Schneider et al, Nucleic Acids Res. 18:6097- 6100 (1990).

FIGS. 14A-14D show a non-limiting example of a fatty acyl-transferase (C protein) sequence logo. The sequence logo was created using Web Logo. See, e.g., Crooks et al, Genome Research, 14:1188-1190, (2004); Schneider et al., Nucleic Acids Res. 18:6097-6100 (1990).

FIGS. 15A-15C show computational identification of Clostridia-derived NRPS-like pathways from human gut metagenome a, Genetic diagrams for the eighteen Clostridia pathways ordered based on their 16S rRNA phylogeny (left tree). Eight pathways were screened from the Human Microbiome Project human gut metagenome (indicated with *) and ten were selected from NCBI nr database. Biosynthetic genes: condensation domain protein (large diagonally-hashed arrows), thiolation domain protein (small solid arrows), adenylation domain protein (large solid arrows). Accessory genes (horizontally-striped arrows): alpha/beta fold hydrolase, PPTase, sterol transfer protein b, Similarity network map of 3531 human gut- associated pathways represented as nodes and connected with edges based on protein sequence- and domain-level similarity (BiG-SCAPE) at a cutoff of 0.9. c, Similarity network map of 346 human gut-derived and actively transcribed pathways, with edges at a cutoff of 0.75. The Clostridia NRPS group is circled.

FIGS. 16A-16D show FAA production from Clostridia-derived pathways a,

Comparison of biosynthetic mechanisms. The Clostridia condensation, thiolation, or adenylation-like protein contains the Pfam condensation (PF00668), PP-binding (PF00550), or AMP-binding (PF00501) domain that is present in its corresponding NRPS homolog. The Clostridia proteins show no sequence homology to known FAA biosynthetic enzymes.

Abbreviations: FA = fatty acid substrate, FadD = fatty acyl-CoA synthetase, N-AT = N- acyltransferase, A = adenylation, T = thiolation, C = condensation, TE = thioesterase, PPTase = phosphopantetheinyl transferase, AA = amino acid substrate b, In vivo production of palmitoleoyl putrescine FAA in E. coli host. MS chromatogram (EIC ESI+ [M + H]+ of palmitoleoyl putrescine: theoretical = 325.3213, experimental = 325.2874) of the in vivo extract of E. coli expressing different combinations of biosynthetic genes and sfp gene. The * denotes the palmitoleoyl putrescine peak c, In vitro adenylation activity assay on possible fatty acid or amine substrates, as measured by absorbance (Abs360) of pyrophosphate (PPi) fluorogenic dye relative to no adenylation enzyme control (TA/T ratio) d, Left bar graph: In vitro condensation activity assay on possible amine substrates, as measured by MS ion intensity of the resulting palmitoleoylated FAA product relative to no condensation enzyme control (CTA/TA ratio). Right bar graph: In vitro condensation assay of lauroyl tryptamine production with the condensation domain protein (EreC), without the protein, or with the H144V mutated condensation domain protein. For c and d, the data was taken in triplicate per day. The reported data is the average of the mean triplicate normalized values conducted on three different days, with the error bar being the s.d. of the three mean triplicate values.

FIG. 17 shows condensation domain protein sequence alignment. MUSCLE alignment of the eight condensation domain (Pfam: PF00668) containing proteins from the eight HMP- derived Clostridia pathways, along with the closest NRPS homolog in the NCBI nr database (NCBI Accession: AKL84684).

FIG. 18 shows thiolation domain protein sequence alignment. MUSCLE alignment of the eight thiolation-like PP-binding domain (Pfam: PF00550) containing proteins from the eight HMP-derived Clostridia pathways, along with the closest NRPS homolog in the NCBI nr database (NCBI Accession: WP_104149350).

FIG. 19 shows adenylation domain protein sequence alignment. MUSCLE alignment of the eight adenylation-like AMP-binding domain (Pfam: PF00501) containing proteins from the eight HMP-derived Clostridia pathways, along with the closest NRPS homolog in the NCBI nr database (NCBI Accession: Q9R9J0).

FIGS. 20A-20B show pathway boundary analysis a, Genetic diagrams for the eighteen Clostridia pathways, consisting of the eight HMP-derived pathways (top) and the ten NCBI nr database (bottom), with pathway boundaries predicted by antiSMASH. b, Annotation of the E. recale pathway including two additional genes outside of the antiS MAS H-predicted boundary. Each gene was investigated for the presence of a homolog in other FAA pathways. FIGS. 21A-21B show expression vector design. Cloning into pET28a (Novagen) for constructs containing a, re-designed pathways and b, single gene expression for protein purification.

FIGS. 22A-22B show LC-MS detection of thiotemplated peptide a, LC-MS detection of phosphopantetheinylated peptide conjugated with lauric acid. The peptide was generated by tryptic digestion of the E. rectale T protein upon reaction with the E. rectale A protein, Sfp, and lauric acid for substrate loading, and confirmed to be absent in the negative control reaction mixture without ATP. b, Summary of ions observed by MS/MS-based fragmentation of the peptide extending from the amino (b ions) and carboxyl (y ions) termini.

FIG. 23 shows the adenylation domain catalytic region sequence alignment. MUSCLE alignment of the catalytic region of the eight adenylation-like AMP-binding domain (Pfam: PF00501) containing proteins from the eight HMP-derived Clostridia pathways, along with the closest NRPS homolog in the NCBI nr database (NCBI Accession: Q9R9J0) and the reference NRPS A domain PheA (GrsA). Sequence location of aspartic acid (D) residues of a canonical NRPS adenylation domains that interact with the a-amino group is indicated with an arrow.

FIG. 24 shows condensation domain sequence alignment and point mutation. MUSCLE alignment of the catalytic region of the condensation domain (Pfam: PF00668) containing proteins from the eight HMP-derived Clostridia pathways, along with the closest NRPS homolog in the NCBI nr database (NCBI Accession: AKL84684) and the reference NRPS A domain TycB (ProCAT). Sequence location of histidine (H) residues known to be essential for catalytic activity is indicated with an arrow.

FIGS. 25A-25D show the in vitro substrate panel assay a, Adenylation activity on a panel of fatty acid substrates, as measured by absorbance (Abs280) of tryptamine conjugated product relative to no adenylation control (TA/T ratio), analyzed on LCMS. b, Condensation activity on a panel of amine substrates, as measured by MS ion intensity of FAA product (consisting of fatty acid with highest activity on adenylation assay) relative to no condensation control (CTA/TA ratio) c, Panel of 25 fatty acids and 53 amines tested. The sources of the fatty acids in the human gut are indicated (symbols on the right of each compound; multiple symbols displayed for multiple sources, with the top symbol signifying the primary source) d, Six Clostridia-derived pathways encoding for unique major FAA compounds. Top: Major compound based on fatty acid and amine substrate with highest activity level in the panel assay. Bottom: Bar graph representing adenylation and condensation activity from the substrate panel assay. The data was taken in triplicate per day. The reported data is the average of the mean triplicate normalized values conducted on three different days, with the error bar being the s.d. of the three mean triplicate values. A threshold level is marked (dotted line) at TA/T ratio > 50 for fatty acid incorporation activity and CTA/TA ratio > 5 for amine incorporation activity.

FIG. 26 shows in vitro substrate screening for pathways producing same major compounds. Six Clostridia-derived pathways encoding for major FAA compounds, each bar graph labeled with the major compound based on fatty acid and amine substrate with highest activity level in the panel assay. Bar graphs represent adenylation and condensation activity from the substrate panel assay. The data was taken in triplicate per day. The reported data is the average of the mean triplicate normalized values conducted on three different days, with the error bar being the s.d. of the three mean triplicate values.

FIGS. 27A-27B show fatty acid specificity profile a, Overall average of product peak area of the eighteen Clostridia A domain proteins on the following fatty acid substrates (from left to right data point): stearic acid (F12), palmitic acid (Fl l), arachidonic acid (F22), docosahexaenoic acid (F25), linoleic acid (F18), a-linolenic acid (F20), and oleic acid (F16). This is correlated with the dissociation constant of human brain fatty acid binding protein measured by isothermal titration calorimetry (TIC), as reported previously (Hanhoff, T., et ah, Mol. Cell. Biochem. 239, 45-54 (2002)). Spearman correlation test (calculated using PRISM version 7): r = -0.9319, P(two-tailed) = 0.0048. b, Substrate specificity profile of the Clostridia adenylation protein by the length of fatty acid substrate structure. The reported value of product peak is the overall average from the eighteen Clostridia A domain proteins. The structure length is calculated by importing the structure from PubChem onto Chem3D version 15 (in the U-bent configuration for unsaturated fatty acids) and measuring the distance from the carboxylic acid carbon to the most distal carbon.

FIGS. 28A-28B show pathways with no observable major amine substrate a, Six Clostridia-derived pathways that showed no major FAA compound production based on not meeting the threshold of CTA/TA ratio > 5 for amine incorporation activity. Top: Compound based on fatty acid and amine substrate with highest activity level in the panel assay. Bottom: Bar graph representing adenylation and condensation activity from the substrate panel assay. The data was taken in triplicate per day. The reported data is the average of the mean triplicate normalized values conducted on three different days, with the error bar being the s.d. of the three mean triplicate values. A threshold level is marked (dotted line) at TA/T ratio > 50 for fatty acid incorporation activity and CTA/TA ratio > 5 for amine incorporation activity b, An expanded list of amines tested for these pathways.

FIG. 29 shows a comparison of major compounds with chemically synthesized standards. MS chromatogram of the in vitro reaction containing the specific pair of substrates yielding the highest enzyme activity, with EIC ESI+ corresponding to m/z of the expected FAA product (oleoyl dopamine: theoretical = 418.3316, experimental = 418.2784); (oleoyl tyramine: theoretical = 402.3367, experimental = 402.2901); (oleoyl aminovaleric acid:

theoretical = 382.3316, experimental = 382.2176); (a-linolenoyl phenylethylamine: theoretical = 382.3104, experimental = 382.2031) (caproyl tryptamine: theoretical 315.2431, experimental = 315.2239) (lauroyl tryptamine: theoretical = 343.2744, experimental = 343.2381). The samples are injected alongside a chemically synthesized and structurally verified standard. The * denotes peak corresponding to the shown compound.

FIG. 30 shows structural homology of Clostridia and human FAAs. The major compounds characterized from the Clostridia FAA pathways (left) in comparison to known human FAAs (middle), suggesting structural and functional mimicry. GPCR targets with which known human FAAs are reported to interact are summarized to the right.

FIGS. 31A-31H show GPCR activity assay of lauroyl tryptamine. a, Cell-based b- arrestin reporter assay (DiscoveRx) with a panel of 168 GPCRs with known ligands in agonist mode and antagonist mode, and also 73 orphan GPCRs in agonist mode at 10 uM. Agonist mode measures % activity of target GPCR by the compound, relative to the baseline value (0% activity) and maximum value with a known ligand, or twofold increase over baseline for orphan GPCR (100% activation). Antagonist mode measures % inhibition of target GPCR by the compound in the presence of a known ligand, relative to the value at the EC 80 (0% inhibition) and basal value (100% inhibition). GPCR targets with activity/inhibition higher than the empirical threshold provided by DiscoverX (30%, 35%, or 50% for GPCR agonist, GPCR antagonist, or orphan GPCR agonist, respectively; plotted as dotted line) suggest that the interaction is potentially significant. EBI2 antagonist activity (61%) is marked with an asterisk (*). b, Concentration-dose response curves of the inhibitory activity of lauroyl tryptamine, tryptamine, and lauric acid on EBI2 in the presence of the native agonist 7a, 25- dihydroxycholesterol (7a,25-diHC) at 0.232 uM (EC80). IC50: lauroyl tryptamine = 0.98 uM; tryptamine > 100 uM; lauric acid > 100 uM. c, Concentration-dose response curves of the inhibitory activity of lauroyl tryptamine, tryptamine, and lauric acid on P2RY4 in the presence of the native agonist UTP at 2.79 uM (EC80). IC50 > 100 uM for all three compounds d, Concentration-dose response curves of the stimulatory activity of lauroyl tryptamine, tryptamine, and lauric acid on GPR132. EC50: lauroyl tryptamine = 1.45 uM; lauric acid = 25.2 uM; tryptamine > 100 uM. For b, c, d, the curves span ten concentration points in duplicates from 100 to 0.0051 uM, with the error bar being the s.d. from duplicates e, MS chromatogram (EIC ESI+ [M + H] ⁺ m/z of lauroyl tryptamine, theoretical = 343.2744, experimental = 343.2369) of the in vivo extract of E. rectale in RCM media (with no fed substrates). The samples are injected alongside a chemically synthesized and structurally verified standard. The * denotes lauroyl tryptamine. The production titer was determined to be 0.04 mg/L. f, Product titer of fatty acyl tryptamine from the in vivo extract of E. rectale in RCM, when fed with exogenous fatty acid (lauric acid, linoleic acid, oleic acid,

docosahexaenoic acid, or a-linolenic acid) g, Mapping of human gut metatranscriptomic reads from subject subX316701492 (accession: SRX247340) onto a 80 kb region of the E. rectale genome centered on the FAA pathway (darker arrows, labeled“FAA”). The mean coverage of the 80 kb segment was 0.8 (SD = 3.3), while the mean coverage of the FAA pathway by itself was 1.5 (SD = 1.1). Housekeeping genes within this segment include tRNA synthetase (tR), carbohydrate metabolism (CM), and sugar transporter (ST) genes h, Immunological and development role of the EBI2-oxy sterol axis and its suppression by E. rectale- derived lauroyl tryptamine to possibly protect against IBD.

FIG. 32 shows GPCR activity assay of oleoyl dopamine. Cell-based b-arrestin reporter assay (DiscoveRx) with a panel of 168 GPCRs with known ligands in agonist mode and antagonist mode, and also 73 orphan GPCRs in agonist mode at 10 uM. Agonist mode measures % activity of target GPCR by the compound, relative to the baseline value (0% activity) and maximum value with a known ligand, or twofold increase over baseline for orphan GPCR (100% activation). Antagonist mode measures % inhibition of target GPCR by the compound in the presence of a known ligand, relative to the value at the EC 80 (0% inhibition) and basal value (100% inhibition). GPCR targets with activity/inhibition higher than the empirical threshold provided by DiscoveRx (30%, 35%, or 50% for GPCR agonist, GPCR antagonist, or orphan GPCR agonist, respectively; plotted as dotted line) suggest that the interaction is potentially significant.

FIG. 33 shows a GPCR activity assay of oleoyl tyramine. Cell-based b-arrestin reporter assay (DiscoveRx) with a panel of 168 GPCRs with known ligands in agonist mode and antagonist mode, and also 73 orphan GPCRs in agonist mode at 10 uM. Agonist mode measures % activity of target GPCR by the compound, relative to the baseline value (0% activity) and maximum value with a known ligand, or twofold increase over baseline for orphan GPCR (100% activation). Antagonist mode measures % inhibition of target GPCR by the compound in the presence of a known ligand, relative to the value at the EC 80 (0% inhibition) and basal value (100% inhibition). GPCR targets with activity/inhibition higher than the empirical threshold provided by DiscoveRx (30%, 35%, or 50% for GPCR agonist, GPCR antagonist, or orphan GPCR agonist, respectively; plotted as dotted line) suggest that the interaction is potentially significant.

FIG. 34 shows a GPCR activity assay of oleoyl aminovaleric acid. Cell-based b-arrestin reporter assay (DiscoveRx) with a panel of 168 GPCRs with known ligands in agonist mode and antagonist mode, and also 73 orphan GPCRs in agonist mode at 10 uM. Agonist mode measures % activity of target GPCR by the compound, relative to the baseline value (0% activity) and maximum value with a known ligand, or twofold increase over baseline for orphan GPCR (100% activation). Antagonist mode measures % inhibition of target GPCR by the compound in the presence of a known ligand, relative to the value at the EC 80 (0% inhibition) and basal value (100% inhibition). GPCR targets with activity/inhibition higher than the empirical threshold provided by DiscoveRx (30%, 35%, or 50% for GPCR agonist, GPCR antagonist, or orphan GPCR agonist, respectively; plotted as dotted line) suggest that the interaction is potentially significant.

FIG. 35 shows a lauroyl tryptamine concentration standard curve. Standard curve of 280 nm absorbance (0.5 nm tolerance) peak area by injection of the chemically synthesized lauroyl tryptamine standard spanning nine concentration points in triplicates from 100 pg to 1 Pg·

FIG. 36 shows a purified biosynthetic protein gel. Purified biosynthetic enzymes (C, T, A proteins) ran on 16% Tricine protein gel and stained with SimplyBlue SafeS tain (Thermo Fisher Scientific) to check purity. Fadder = Precision Plus Protein Fadder (Bio-Rad).

DETAIFED DESCRIPTION OF THE INVENTION

Fatty acid amides comprise a diverse class of organic compounds and have been implicated as neuromodulatory lipids and as regulators of human physiopathological conditions. Production of fatty acid amides, however, using conventional laboratory host cells often has low yield. For example, in laboratory strains of E. coli, exogenous fatty acids are often converted into fatty acyl-CoAs, which are rapidly degraded and rebuilt by E. coli's fatty acid biosynthesis machinery (FASH) into their endogenous pool of fatty acyl-ACPs. Without being bound by a particular theory, in E. coli, the endogenous fatty acyl-transferase enzyme that conducts the lipidation of these molecules does not use free fatty acid as its substrate, but instead uses a fatty acid conjugated onto either coenzyme A (fatty acyl-CoA) or acyl carrier protein (fatty acyl-ACP). Therefore, when a foreign biosynthetic pathway comprising a fatty acyl-transferse is expressed in E. coli, the addition of an exogenous fatty acid in the culture broth generally does not lead to the incorporation of the desired fatty acid onto the final product, but rather a fatty acid that is endogenously made by the host.

This disclosure is premised, at least in part, on the unexpected finding that host cells may be engineered to efficiently produce fatty acid amides from free fatty acids. Accordingly, provided herein, in some embodiments, are vectors encoding biosynthetic enzymes derived from gut microbiome bacteria, engineered cells comprising the same, and methods of using biosynthetic enzymes derived from gut microbiome bacteria to produce fatty acid amides. Biosynthetic enzymes are enzymes that are capable of promoting synthesis of organic compounds and include fatty acyl-transferase, acyl carrier protein (ACP), fatty acyl-ACP synthetase, hydrolases, lipid transfer proteins, and glycosyltransferases. In some instances, a biosynthetic enzyme is a metabolic enzyme.

Fatty acid amides (FAAs) comprise a hydrophobic acyl tail group and a head group and are formed from a fatty acid and an amine. Non-limiting examples of fatty acid amides include oleoyl aminovaleric acid, a-linolenoyl aminovaleric acid, oleoyl g-aminobutyric acid, oleoyl dopamine, oleoyl tyramine, palmitoleoyl dopamine, a-linolenoyl tyramine, oleoyl

phenethylamine, lauroyl tyrptamine, lineoleoyl tryptamine, lauroyl tyramine, a-linolenoyl homocysteine, oleoyl homocysteine, a-linolenoyl cysteine, linoleoyl homocysteine, oleoyl- aminopentanoic acid, palmitoleyl-putrescine, and a-linolenoyl homocysteamine. FAAs produced by humans include arachidonoyl g-aminobutryic acid, oleoyl dopamine,

docosahexaenoyl serotonin, and arachidonoyl serine. See also, FIGs. 2E, 9 and 32. In some embodiments, the fatty acid amides produced by the methods described herein are structural analogs (also referred to herein as structural homologs) of FAAs that are naturally produced by a cell or subject. See, e.g., FIGs. 9 and 32. In some instances, the FAAs produced by the methods described herein are functional analogs that share one or more biological activities with a FAA that is naturally produced by a cell or subject. As a non-limiting example, a FAA produced by the methods described herein may activate a particular cellular pathway that is activated by a naturally occurring structural analog. In some instances, the subject is a human. Other than human cells, the human gut microbiome can likewise produce diffusible metabolites that interact with human cellular targets to modulate human signaling

pathways. ^15,16 For example, a phenotypic assay of DNA damage-inducing Escherichia coli or colitis-inducing Klebsiella oxytoca led to discovery of pathway responsible for the production of the compounds colibactin ^17,18 and tilivalline, ¹⁹ respectively. Human microbiome sequencing studies show that, while healthy individuals differ remarkably in the bacterial strains that constitute the microbiome at a taxonomical level, the metabolic pathways remain fairly constant. ²⁰ Taxonomical identification of the human microbiota is therefore not sufficient in associating them with different health states. Elucidating the chemical modulators and connecting them back to specific bacterial genes provide the means to glean functional outputs from DNA sequencing data and ultimately control the underlying mechanisms by therapeutic interventions.

A forward genetics-based approach, like phenotypic assay, has been instrumental in compound discovery, but it is intrinsically designed to be characterized one molecule or bacterial strain at a time. Alternatively, metabolites can be elucidated in a reverse genetics- based manner by bioinformatically mining putative compound-encoding biosynthetic pathways from DNA sequencing data. As an example, a focused search on biosynthetic genes encoding the N-acyl synthase protein domain (PF13444), followed by E. coli heterologous expression of representative genes, led to the isolation of FAAs that modulate human health by structurally mimicking human endogenous FAAs, primarily of the amino acid conjugate type from Gram negative gut commensal bacteria. ²¹ A more generalized computational search using a detection algorithm (antiSMASH) ²² based on pattern recognition from known pathways identified 14,000 putative pathways from microbiome sequencing data, select groups of which have been characterized to produce compound classes, like lactocillin-based antibiotics ²³ and peptide aldehyde-based protease inhibitors. ²⁴ Without being bound by a particular theory, sequence- based methods provide the means to leverage the growing number of available genomic datasets to identify large unexplored groups of pathways that are likely to produce compounds significant in the human gut and systematically characterize them en masse.

Putative pathways identified from sequence data have largely been characterized in vivo by fermentation of the pathway-harboring native or heterologous bacterial strain in laboratory media. However, such laboratory in vivo systems often fail to elucidate pathways that rely on substrates that are absent from or found in insufficient levels in the host strain or culture broth, including bacterial pathways that convert bile acids endogenously produced by human liver cells into more cytotoxic secondary bile acids, like deoxycholic acid and lithocholic acid. ²⁵ The human gut lumen contains a complex mixture of metabolites from gut microbes, human cells, and dietary intake, which is difficult to recapitulate in vivo. On the other hand, pathways can be investigated with precise control of substrates and enzymes by in vitro reconstitution, like the characterization of colibactin biosynthesis. ²⁶ However, this requires a priori biochemical knowledge of the pathways to predict the substrates and enzymatic conditions, which are often lacking for pathways gleaned from sequence data.

One natural product superclass that is widely used by bacteria is the nonribosomal peptide synthetase (NRPS) class. ²⁷ NRPS is known to produce peptide -based compounds of various structures (e.g. cyclic, D-amino acids) and bioactivities (e.g. antibiotics,

immunosuppressants). The biosynthetic machinery consists of multiple modules, each of which is generally responsible for incorporating a specific amino acid subunit. The minimal module contains three protein domains: a carrier protein or thiolation domain (T; PF00550) that requires a 4'-phospho-pantetheine (PPT) arm, added by a separate PPTase (e.g. Sfp); ²⁸ an adenylation domain (A; PF00501) that uses ATP to load a specific amino acid onto the T domain to form aminoacyl-T thiotemplated intermediate; and a condensation domain (C; PF00668) that catalyzes amide bond formation between the upstream and downstream aminoacyl-T to generate peptidyl-T thiotemplated intermediate. A termination domain then releases the elongated peptidyl-T from the machinery to form the final product. The chemical ecology in various habitats evolutionarily subjects the bacteria to produce unique secondary metabolites made by specialized NRPS machineries, such as the iterative module of gut E. coli enterobactin biosynthesis ²⁹ and the bisintercalator loading domain of marine Micromonospora thiocoraline biosynthesis. ³⁰

As described herein, in some embodiments, genome mining was coupled with in vitro reconstitution to characterize a group of noncanonical NRPS-like pathways from the Gram positive human gut Clostridia. Putative biosynthetic pathways were computationally detected from publically available human-associated bacterial genomes, and further mapped with metagenomic and metatranscriptomic reads from human gut samples of the Human

Microbiome Project to identify eight unprecedented NRPS-like pathways that are ubiquitously present and actively transcribed in the human gut.

Each of the three conserved biosynthetic genes encodes for exactly one protein domain (C, T, or A) that together constitute the NRPS minimal module. Surprisingly, E. coli heterologous expression resulted in the production of a FAA. By in vitro reconstitution of the pathway, the C domain was found to load an exogenous fatty acid, instead of an amino acid as predicted from conventional NRPS enzymology, onto the T domain. Furthermore, contrary to NRPS C domain that catalyzes the conjugation of two thiotemplated substrates, ³¹ this pathway's C domain was characterized to conjugate a thiotemplated fatty acid with a free amine, providing the flexibility to incorporate monoamine neurotransmitter as substrate.

Without being bound by a particular theory, because the FAA isolated from E. coli was found to be an artifact, the pathways were screened for their preferred substrates with a panel of bacteria-, human-, and food-derived fatty acids and amines known to be abundant in the human gut, resulting in the characterization of four major neurotransmitter conjugate FAA products.

In some instances, the chemical structures of the gut Clostridia FAAs are either identical or otherwise homologous to known human FAAs, making them likely to interact with similar human endogenous cellular targets involved in various signaling pathways and serve as a way to modulate human health.

Without being bound by a particular theory, genome mining across multiple sequencing datasets serves as an effective approach for the systematic discovery of human microbiome- derived chemical effectors that scales with the exponential growth of publically available sequence data with minimal prior knowledge. ⁵³ However, the conventional method for sequence-to-compound production relies on heterologous expression, which may fail due to the lack of proper substrates in the heterologous host's metabolic pool. ⁵⁴ Coupling of sequence mining with in vitro substrate panel screening to analyze enzyme activity with defined substrates to systematically characterize human gut microbiome-derived biosynthetic pathways encoding for chemical compounds is described herein. Without being bound by a particular theory, this allowed for the characterization of the highly abundant and actively transcribed gut Clostridia NRPS-like pathways that preferentially incorporated substrates abundant in the human gut that may not be readily available in conventional laboratory host strains and culture media. Laurie acid and a-linolenic acid are examples of such substrates, found mostly in plants as part of human diet. ⁵⁵ Although oleic acid occurs naturally across diverse biological systems, the fact that it is the most abundant fatty acid in human adipose tissue ⁵⁶ suggests that Clostridia pathways have also likely incorporated fatty acids from the human gut lumen. For the amine substrates, dopamine, tyramine, tryptamine, and aminovaleric acid may be absent from or found in insufficient levels in E. coli host strains, but are found endogenously in humans, as well as secreted by gut commensal bacteria. ⁵⁷ Thus, instead of encoding for substrate biosynthesis, the FAA pathways encode a mechanism to incorporate substrates from the host, a phenomenon that has been observed for known biosynthetic pathways in symbiotic systems.

The Clostridia pathways were characterized to produce four major compounds of the FAA type, the same class of signaling molecules produced by human cells to interact with various signaling pathways involved in human health (FIG. 9). One Clostridia-derived compound, oleoyl dopamine, is a known human endogenous FAA. Two of the other compounds, oleoyl tyramine and lauroyl tryptamine, are structurally analogous to known human FAAs, including oleoyl dopamine and docosahexanoyl serotonin. These FAAs are aryl neurotransmitter conjugates that have anandamide-like fatty acid tail and dopamine-like aryl head group, thereby providing the potential to interact with various neurotransmitter receptor sites. The remaining Clostridia-derived FAA, oleoyl aminovaleric acid, is structurally similar to the known human FAA, arachidonoyl GABA. These neurotransmitter conjugates have a GABA-like linear head group, allowing for the interaction with neurotransmitter receptor sites distinct from the arylamine conjugates.

The human gut is referred to as the "second brain" because of the multitude of neurons, otherwise called the enteric nervous system, that embed the walls of the gastrointestinal tract. ⁵⁹ Neurotransmitters within the gut have been investigated to interact with these receptors to not only influence host neurological, but also immune, metabolic, and vasculature states. ⁶⁰ The known human endogenous FAAs have also been suggested to interact with these receptors to modulate conditions, such as pain, mood, inflammation, and obesity. ⁶¹ Considering the structural mimicry, these Clostridia-derived FAAs are likely the missing chemical link that enables human gut Clostridia to interact with a diverse set of host signaling pathways to influence a variety of physiopathological conditions. The systematic elucidation of these functional outputs will ultimately allow for the deterministic manipulation of these chemical modulators and their targets as an effective mode for human microbiome-based therapy.

Reference numbers in the preceding section correspond to those listed in Reference Section 1.

Gut microbiome bacterium

Any of the biosynthetic enzymes ( e.g ., fatty acyl-transferase, acyl carrier protein (ACP), fatty acyl-ACP synthetase, hydrolases, lipid transfer proteins, and glycosyltransferases) described herein may be derived from a gut microbiome bacterium. A human gut microbiome bacterium is a bacterium that is found in the gastrointestinal tract of a human. The gut microbiome bacterium may be from any species. For example, the gut microbiome bacterium may be Bacteroides spp., Enterococcus spp., Escherichia spp.,

Enterobacter spp., Klebsiella spp., Bifidobacterium spp., Staphylococcus spp., Lactobacillus, Clostridium spp., Proteus spp., Pseudomonas spp., Coprococcus spp., Eachnoclostridum spp., Eubacterium spp., Salmonella spp., Faecalibacterium spp., Peptostreptococcus spp., or Peptococcus spp. See also, e.g., Lloyd-Price Nature. 2017 Oct 5;550(7674):61-66.

In some instances, a biosynthetic enzyme is a human gut microbiome-derived

Clostridia enzyme. Clostridia comprise a class of anaerobic bacteria that are polyphyletic. Non-limiting orders of Clostridia include Clostridiales, Halanaerobiales, Natranaerobiales, Thermoanaerobacteriales, and Negativicutes. Genera of Clostridia include Coprococcus, Marvinbryantia, Lachnoclostridum, Blautia, Ruminococcaceae, Eubacterium, and Clostridium. For example, a biosynthetic enzyme may be derived from Coprococcus eutactus,

Marvinbryantia formatexigens, Lachnoclostridium clostridioforme, Blautia producta,

Rumiococcaceae bacterium, Clostridiales sp., Eubacterium rectale, or Clostridium celatum. Non-limiting examples of human gut-associated bacteria include Bacteroides fragilis NCTC 9343, Bacteroides fragilis YCH46, Bacteroides thetaiotaomicron VPI-5482, Bifidobacterium longum NCC2705, Enterococcus faecalis V583, Escherichia coli sv. 06:K15:H31 536, Escherichia coli 0157:H7 EDL933 (EHEC), Escherichia coli 0157:H7 Sakai (EHEC), Escherichia coli UTI89 (UPEC), Helicobacter hepaticus 3B1, ATCC 51449, Helicobacter pylori 26695, Helicobacter pylori HPAG1, Helicobacter pylori J99, Lactobacillus acidophilus NCFM, Lactobacillus delbrueckii bulgaricus sv. E Lbl4, Lactobacillus johnsonii NCC 533, Lactobacillus salivarius salivarius UCC118, Lawsonia intracellularis PHE/MNl-00, Listeria monocytogenes sv. l/2a EGD-e, Listeria monocytogenes sv. 4b F2365, Porphyromonas gingivalis W83, Vibrio cholerae sv. 01 bv. El Tor N16961, Campylobacter jejuni jejuni HB93- 13, Campylobacter upsaliensis RM3195, Escherichia coli 0148:H28 B7A

(CS6:LT+:ST+)(ETEC), Escherichia coli 011 LH9 El 10019 (EPEC), Escherichia coli

0139:H28 Fl l (ETEC), Listeria monocytogenes sv. l/2a F6854, Listeria monocytogenes 4b H7858, Yersinia bercovieri ATCC 43970, Bifidobacterium adolescentis ATCC 15703, Lactobacillus brevis ATCC 367, Lactobacillus casei ATCC 334, Lactobacillus delbrueckii bulgaricus ATCC BAA-365, Lactobacillus gasseri ATCC 33323, Collinsella aerofaciens ATCC 25986, Bacteroides vulgatus ATCC 8482, Campylobacter concisus 13826,

Campylobacter curvus 525.92, Campylobacter hominis ATCC BAA-381, Citrobacter koseri ATCC BAA-895, Escherichia coli sv. 0139:H28 E24377A (ETEC), Escherichia coli sv. 09 HS, Parabacteroides distasonis ATCC 8503, Yersinia pseudotuberculosis sv. 0:1b IP 31758, Vibrio cholerae sv. 014 MZO-2, Pseudoflavonifractor capillosus ATCC 29799,

Bifidobacterium adolescentis L2-32, Parabacteroides merdae ATCC 43184, Eubacterium ventriosum ATCC 27560, Bacteroides caccae ATCC 43185, Blautia obeum ATCC 29174, Ruminococcus torques ATCC 27756, Campylobacter jejuni CG8486, Dorea longicatena DSM 13814, Listeria monocytogenes sv. 4b HPB2262, Ruminococcus gnavus ATCC 29149, Listeria monocytogenes sv. l/2al LSL N3-165, Lactobacillus helveticus DPC 4571, Laecalibacterium prausnitzii M21/2, Parvimonas micra ATCC 33270, Coprococcus eutactus ATCC 27759, Clostridium leptum DSM 753, Lachnoclostridium bolteae ATCC BAA-613, Absiella dolichum DSM 3991, Bacteroides uniformis ATCC 8492, Clostridium sp. L2-50, Bacteroides ovatus ATCC 8483, Escherichia coli C ATCC 8739, Escherichia coli DH10B, Linegoldia magna ATCC 29328, Intestinibacter bartlettii DSM 16795, Burkholderia oklahomensis E0147, Dorea formicigenerans ATCC 27755, Escherichia albertii TW07627, Clostridium spiroforme DSM 1552, Streptococcus infantarius infantarius ATCC BAA-102, Providencia stuartii ATCC 25827, Eubacterium siraeum DSM 15702, Bifidobacterium dentium ATCC 27678,

Anaerofustis stercorihominis DSM 17244, Bacteroides stercoris ATCC 43183,

Lachnoclostridium scindens ATCC 35704, Alistipes putredinis DSM 17216, Escherichia coli 0157:H7 EC508, Anaerostipes caccae DSM 14662, Bifidobacterium animalis lactis HN019, Listeria monocytogenes FSL J2-071, Listeria monocytogenes sv. l/2b FSL J2-064,

Clostridium sp. SS2/1, Anaerotruncus colihominis DSM 17241, Akkermansia muciniphila ATCC BAA-835, Bifidobacterium longum DJO10A, Helicobacter pylori Shi470,

Lactobacillus casei casei BL23, Ruminococcus lactaris ATCC 29176, Bacteroides coprocola M16, DSM 17136, Clostridium sporogenes ATCC 15579, Bacteroides intestinalis 341, DSM 17393, Bifidobacterium catenulatum DSM 16992, JCM 1194, LMG 11043, Desulfovibrio piger ATCC 29098, Providencia alcalifaciens sv. 019:H2 DSM 30120, Providencia rustigianii DSM 4541, Collinsella stercoris DSM 13279, Anaerococcus hydrogenalis DSM 7454,

Helicobacter pylori HPKX_438_AG0C1, Helicobacter pylori 98-10, Helicobacter pylori B128, Helicobacter pylori HPKX_438_CA4C1, Bacteroides eggerthii DSM 20697, Bacteroides pectinophilus ATCC 43243, Bacteroides plebeius M12, DSM 17135, Parabacteroides johnsonii DSM 18315, Lachnoclostridium hylemonae DSM 15053, Holdemanella biformis DSM 3989, Lactobacillus rhamnosus HN001, Tyzzerella nexilis DSM 1787, Bacteroides dorei DSM 17855, Clostridium hiranonis TO-931 DSM 13275, Bacillus cereus AH187 (F4810/72), Bacillus cereus G9842, Bifidobacterium animalis lactis AD011, Bifidobacterium longum infantis ATCC 15697, Escherichia coli 081 EDla, Escherichia coli SE11, Helicobacter pylori G27, Helicobacter pylori P12, Listeria monocytogenes sv. 4a HCC23, Roseburia inulinivorans DSM 16841, Clostridiales sp. 1_7_47FAA, Lactobacillus acidophilus ATCC 4796,

Bifidobacterium bifidum NCIMB 41171, Lactobacillus hilgardii ATCC 8290, Lactobacillus ultunensis DSM 16047, Lactobacillus salivarius H066, ATCC 11741, Lactobacillus paracasei ATCC 25302, Acidaminococcus intestini D21, Lactobacillus brevis gravesensis ATCC 27305, Lactobacillus fermentum ATCC 14931, Lactobacillus buchneri ATCC 11577, Enterococcus faecalis TX1322, Enterococcus faecalis TX0104, Bifidobacterium breve DSM 20213, JCM 1192, Holdemania filiformis VPI J 1-3 IB- 1 , DSM 12042, Leuconostoc mesenteroides cremoris ATCC 19254, Catenibacterium mitsuokai DSM 15897, Bacteroides cellulosilyticus DSM 14838, Lachnoclostridium asparagiforme DSM 15981, Coprococcus comes ATCC 27758, Proteus penneri ATCC 35198, Enterococcus faecium TX1330, Lactobacillus rhamnosus LMS2-1, Lactobacillus paracasei 8700:2, Bifidobacterium longum longum ATCC 55813, Helicobacter cinaedi CCUG 18818, Bacteroides coprophilus DSM 18228, JCM 13818, Blautia hydrogenotrophica DSM 10507, Eubacterium hallii DSM 3353, Clostridium methylpentosum R2, DSM 5476, Helicobacter canadensis MGG 98-5491, ATCC 700968, Bifidobacterium longum longum CCUG 52486, Helicobacter pullorum MGG 98-5489, Bifidobacterium animalis lactis Bl-04, ATCC SD5219, Bifidobacterium animalis lactis DSM 10140, Eubacterium eligens ATCC 27750, Eubacterium rectale ATCC 33656, Helicobacter pylori B38,

Lactobacillus plantarum JDM1, Lactobacillus rhamnosus GG, Acinetobacter baumannii AB900, Parabacteroides sp. 2_1_7, Bacteroides fragilis 3_1_12, Weissella paramesenteroides ATCC 33313, Fusobacterium nucleatum subsp. animalis Dl l, Escherichia coli D9,

Fusobacterium gonidiaformans ATCC 25563, Faecalibacterium prausnitzii A2-165,

Butyrivibrio crossotus DSM 2876, Hungatella hathewayi DSM 13479, Lactobacillus antri DSM 16041, Lactobacillus helveticus DSM 20075, Fusobacterium gonidiaformans 3-1-5R, Oxalobacter formigenes HOxBLS, Coprobacillus sp. D7, Fusobacterium nucleatum animalis 7_1, Fusobacterium nucleatum subsp. vincentii 4_1_13, Fusobacterium mortiferum ATCC 9817, Helicobacter winghamensis ATCC BAA-430, Citrobacter portucalensis 30_2,

Bacteroides dorei 5_1_36/D4, Escherichia sp. 3_2_53FAA, Clostridium sp. 7_2_43FAA, Escherichia sp. 1_1_43, Fusobacterium periodonticum 2_1_31, Bacteroides sp. 9_1_42FAA, Bacteroides sp. Dl, Oxalobacter formigenes OXCC13, Bacteroides sp. 2_2_4, Bacteroides thetaiotaomicron 1_1_6, Bacteroides sp. 4_3_47FAA, Fusobacterium varium ATCC 27725, Parabacteroides sp. D13, Clostridioides difficile CD196, Clostridioides difficile R20291, Escherichia coli K-12, MG1655, Mycoplasma hominis PG21, ATCC 23114, Veillonella parvula Te3, DSM 2008, Bifidobacterium longum longum JDM301, Escherichia coli 055:H7 CB9615, Lactobacillus johnsonii FI9785, Bifidobacterium animalis lactis BB-12,

Bifidobacterium animalis lactis V9, Escherichia coli 044:H18 042 (EAEC), Escherichia coli 018:K1:H7 IHE3034, Escherichia coli O150:H5 SE15, Helicobacter pylori 51, Helicobacter pylori v225, Lactobacillus rhamnosus GG, Acetomicrobium hydrogeniformans ATCC BAA- 1850, Bacteroides ovatus SD CC 2a, Bacteroides ovatus SD CMC 3f, Bacteroides vulgatus PC510, Bacteroides xylanisolvens SD CC lb, Clostridioides difficile NAP07, Clostridioides difficile NAP08, Corynebacterium ammoniagenes DSM 20306, Edwardsiella tarda ATCC 23685, Enterococcus faecium PC4.1, Enterococcus faecium E1071, Grimontia hollisae CIP 101886, Lactobacillus amylolyticus DSM 11664, Turicibacter sanguinis PC909, Vibrio cholerae CT 5369-93, Vibrio cholerae sv. Inaba INDRE 91/1, Vibrio cholerae sv. 01 RC27, Vibrio furnissii sv. II CIP 102972, Vibrio mimicus MB-451, Bacteroides fragilis 2_1_16, Bacteroides sp. 2_1_22, Bacteroides sp. 2_1_33B, Bacteroides sp. 3_1_33FAA, Bacteroides sp. D20, Enterococcus faecalis X98, ATCC 27276, Enterococcus faecium 1,141,733, Enterococcus faecium 1,231,501, Enterococcus faecium Coml2, Enterococcus faecium Coml5, Fusobacterium periodonticum 1_1_41FAA, Klebsiella variicola 1_1_55, Pediococcus acidilactici 7_4, Streptococcus sp. 2_1_36FAA, Veillonella sp. 3_1_44, Veillonella sp.

6_1_27, Escherichia coli 026:H11 11368, Helicobacter pylori PeCan4, Helicobacter pylori SJM180, Campylobacter jejuni jejuni Ml, Edwardsiella tarda FL6-60, Escherichia coli AIEC UM146, Helicobacter pylori 908, Helicobacter pylori Cuz20, Helicobacter pylori Sat464, Campylobacter coli JV20, Enterococcus faecalis TX2134, Escherichia coli MS 107-1, Escherichia coli MS 115-1, Escherichia coli MS 116-1, Escherichia coli MS 119-7,

Escherichia coli MS 124-1, Escherichia coli MS 146-1, Escherichia coli MS 175-1,

Escherichia coli MS 182-1, Escherichia coli MS 185-1, Escherichia coli MS 187-1,

Escherichia coli MS 196-1, Escherichia coli MS 198-1, Escherichia coli MS 200-1,

Escherichia coli MS 21-1, Escherichia coli MS 45-1, Escherichia coli MS 69-1, Escherichia coli MS 78-1, Escherichia coli MS 84-1, Bacteroides sp. 1_1_14, Parabacteroides sp. 20_3, Bacteroides sp. 3_1_19, Bacteroides sp. 3_1_23, Bacteroides sp. D22, Burkholderiales bacterium 1_1_47, Bifidobacterium bifidum PRL2010, Bifidobacterium bifidum S17, Bifidobacterium longum infantis 157F-NC, Bifidobacterium longum longum BBMN68, Bifidobacterium longum longum JCM 1217, Campylobacter jejuni jejuni ICDCCJ07001, Helicobacter felis CS1, ATCC 49179, Helicobacter pylori B8, Lactobacillus delbrueckii bulgaricus ND02, Odoribacter splanchnicus 1651/6, DSM 20712, Anaerostipes sp.

3_2_56FAA, Arcobacter butzleri JV22, Bacteroides eggerthii 1_2_48FAA, Bacteroides sp. 3_1_40A, Bacteroides sp. 4_1_36, Bifidobacterium sp. 12_1_47BFAA, Campylobacter upsaliensis JV21, Clostridium sp. HGF2, Lachnoclostridium symbiosum WAL-14163, Lachnoclostridium symbiosum WAL-14673, Coprobacillus cateniformis 29_1, Eggerthella sp. 1_3_56FAA, Erysipelotrichaceae bacterium sp. 3_1_53, Escherichia coli MS 145-7,

Faecalibacterium cf. prausnitzii KLE1255, Helicobacter suis HS1, Helicobacter suis HS5, Lachnospiraceae bacterium sp. 5_1_63FAA, Lachnospiraceae bacterium sp. 8_1_57FAA, Pediococcus acidilactici DSM 20284, Phascolarctobacterium succinatutens YIT 12067, Prevotella salivae DSM 15606, Ralstonia sp. 5_7_47FAA, Streptococcus anginosus

1_2_62CV, Streptococcus equinus ATCC 9812, Succinatimonas hippei YIT 12066, Sutterella wadsworthensis 3_1_45B, Alistipes shahii WAL 8301, Bacteroides fragilis 638R, Bacteroides xylanisolvens XB1A, Bifidobacterium longum longum F8, Butyrivibrio fibrisolvens 16/4, Lachnoclostridium cf. saccharolyticum K10, Coprococcus catus GD/7, Enterobacter cloacae cloacae NCTC 9394, Enterococcus faecalis 62, Enterococcus faecalis 7L76, Faecalitalea cylindroides T2-87, Eubacterium rectale DSM 17629, Eubacterium rectale M 104/1,

Eubacterium siraeum 70/3, Faecalibacterium prausnitzii SL3/3, Faecalibacterium prausnitzii L2-6, Gordonibacter pamelaeae 7-10-1-bT, DSM 19378, Helicobacter pylori Gambia94/24, Helicobacter pylori India7, Helicobacter pylori Lithuania75, Helicobacter pylori SouthAfrica7, Lactobacillus delbrueckii bulgaricus 2038, Listeria monocytogenes sv. 4a L99, Megamonas hypermegale ART12/1, Roseburia intestinalis M50/1, Roseburia intestinalis XB6B4,

Ruminococcus bromii L2-63, Blautia obeum A2-162, Ruminococcus sp. SRI/5,

Ruminococcus sp. 18P13, Ruminococcus torques L2-14, Clostridiales sp. SS3/4, Anaerostipes hadrus SSC/2, Victivallis vadensis ATCC BAA-548, Clostridium sp. SY8519, Eggerthella sp. YY7918, Helicobacter bizzozeronii CIII-1, Bifidobacterium animalis lactis CNCM 1-2494, Bifidobacterium breve UCC2003 (NCIMB8807), Bifidobacterium longum infantis JCM 1222, Escherichia coli DH1 (ME8569), Escherichia coli LF82, Helicobacter pylori 2017,

Helicobacter pylori 2018, Helicobacter pylori 83, Helicobacter pylori FI 6, Helicobacter pylori F30, Helicobacter pylori F32, Helicobacter pylori F57, Lactobacillus amylovorus GRL 1118, Lactobacillus johnsonii DPC 6026, Yersinia enterocolitica palearctica sv. 0:3 bt. 4 Y11, Clostridium sp. D5, Bacteroides clarus YIT 12056, Bacteroides fluxus YIT 12057, Bacteroides ovatus 3_8_47FAA, Bacteroides sp. 1_1_30, Bacteroides fragilis 2_1_56FAA, Eggerthella sp. HGA1, Fusobacterium nucleatum subsp. animalis 11_3_2, Klebsiella sp. MS 92-3, Lachnospiraceae bacterium 1_1_57FAA, Lachnospiraceae bacterium 1_4_56FAA,

Lachnospiraceae bacterium 2_1_58FAA, Lachnospiraceae bacterium 3_1_46FAA,

Lachnospiraceae bacterium 6_1_37FAA, Lachnospiraceae bacterium 5_1_57FAA,

Lachnospiraceae bacterium 6_1_63FAA, Lachnospiraceae bacterium 9_1_43BFAA, Neisseria macacae ATCC 33926, Paenibacillus sp. HGF5, Paenibacillus sp. HGF7, Paraprevotella xylaniphila YIT 11841, Parasutterella excrementihominis YIT 11859, Turicibacter sp. HGF1, Lactobacillus reuteri CF48-3A, Bifidobacterium pseudocatenulatum D2CA, Bdellovibrio bacteriovorus W, Alistipes finegoldii AHN 2437, DSM 17242, Microcystis aeruginosa PCC 7806, Megasphaera elsdenii DSM 20460, Acidaminococcus intestini RyC-MR95, Roseburia hominis A2-183, DSM 16839, Enterococcus faecalis OG1RF, ATCC 47077, Helicobacter pylori Punol35, Pseudomonas aeruginosa NCGM2.S1, Helicobacter pylori 35A, Listeria monocytogenes sv. l/2a 10403S, Listeria monocytogenes FSL R2-561, Listeria

monocytogenes sv. l/2a J0161, Listeria monocytogenes Finland 1988, Helicobacter pylori Punol20, Corynebacterium appendicis CIP 107643, Alistipes sp. HGB5, Bifidobacterium breve SC95, Bifidobacterium longum infantis SC 142, Bifidobacterium breve SC 154, Bifidobacterium bifidum SC555, Bifidobacterium longum longum SC596, Bifidobacterium longum longum SC664, Bifidobacterium breve JCM 7019, Helicobacter cetorum MIT 99- 5656, Helicobacter cetorum MIT 00-7128, Helicobacter pylori Shil 12, Helicobacter pylori Shil69, Helicobacter pylori Shi417, Helicobacter pylori PeCanl8, Bifidobacterium animalis animalis ATCC 25527, Pseudomonas sp. 2_1_26, Klebsiella sp. 4_1_44FAA, Citrobacter freundii 4_7_47CFAA, Tannerella sp. 6_1_58FAA_CT1, Hungatella hathewayi WAL-18680, Alistipes indistinctus YIT 12060, Odoribacter laneus YIT 12061, Flavonifractor plautii ATCC 29863, Campylobacter sp. 10_1_50, Synergistes sp. 3_l_synl, Subdoligranulum sp.

4_3_54A2FAA, Enterococcus saccharolyticus 30_1, Lachnoclostridium clostridioforme 2_1_49FAA, Prevotella stercorea DSM 18206, Enterococcus faecalis PC 1.1, Bilophila sp. 4_1_30, Desulfovibrio sp. 6_1_46AFAA, Bacillus sp. 7_6_55CFAA_CT2, Lachnospiraceae bacterium sp. 7_1_58FAA, Fusobacterium necrophorum funduliforme 1_1_36S,

Erysipelotrichaceae bacterium sp. 21_3, Dorea formicigenerans 4_6_53AFAA, Collinsella tanakaei YIT 12063, Erysipelotrichaceae sp. 2_2_44A, Eubacterium sp. 3_1_31, Dialister succinatiphilus YIT 11850, Lachnoclostridium citroniae WAL-17108, Coprobacillus sp. 8_2_54BFAA, Clostridium sp. 7_3_54FAA, Sutterella parvirubra YIT 11816, Coprobacillus sp. 3_3_56FAA, Hafnia alvei ATCC 51873, Bacillus smithii 7_3_47FAA, Propionibacterium sp. 5_U_42AFAA, Paraprevotella clara YIT 11840, Lactobacillus sp. 7_1_47FAA, Megamonas funiformis YIT 11815, Blautia producta ATCC 27340, Acetomicrobium hydrogenif ormans ATCC BAA- 1850, Listeria monocytogenes NKB04_01, Listeria monocytogenes sv. 4b NKB04_02, Listeria monocytogenes sv. 4b NKB04_03, Listeria monocytogenes sv. l/2b NKB06_01, Listeria monocytogenes NKB06_03, Vibrio cholerae sv. 01 Ogava P-18785, Escherichia coli HM605, Helicobacter pylori XZ274, Eubacterium siraeum DSM 15702, Lactobacillus helveticus R0052, Lawsonia intracellularis N343, Escherichia coli DEC15E, Escherichia coli DEC15D, Escherichia coli DEC15C,

Desulfitobacterium hafniense DP7, Bifidobacterium longum D2957, Vibrio cholerae

ZWU0020, Helicobacter pylori 26695, Clostridium celatum DSM 1785, Helicobacter pylori Hp P-15b, Helicobacter pylori Hp P-25c, Helicobacter pylori Hp P-25d, Helicobacter pylori Hp P-28b, Escherichia coli DEC15A, Escherichia coli DEC15B, Escherichia coli DEC14D, Escherichia coli DEC13C, Escherichia coli DEC13D, Escherichia coli DEC14A, Escherichia coli DEC9E, Escherichia coli DEC5B, Escherichia coli DEC13A, Escherichia coli DEC11D, Escherichia coli DEC11C, Escherichia coli DEC12A, Escherichia coli DEC12C, Escherichia coli DEC12B, Escherichia coli DEC12D, Helicobacter pylori NQ4076, Helicobacter pylori NQ4044, Helicobacter pylori NQ4110, Helicobacter pylori Hp A-5, Helicobacter pylori Hp A- 4, Helicobacter pylori Hp A-9, Escherichia coli DECIOE, Escherichia coli DECIOF,

Escherichia coli DEC IOC, Escherichia coli DEC10D, Escherichia coli DECIOA, Escherichia coli DECIOB, Escherichia coli DEC9D, Escherichia coli DEC11A, Escherichia coli DEC1 IB, Helicobacter pylori Hp H-43, Escherichia coli DEC9A, Escherichia coli DEC8D, Escherichia coli DEC8C, Escherichia coli DEC9B, Listeria monocytogenes sv. l/2c SLCC 2372, Listeria monocytogenes sv. l/2b SLCC 2755, Listeria monocytogenes sv. 4e SLCC 2378,

Bifidobacterium bifidum BGN4, Listeria monocytogenes sv. 3b SLCC 2540, Bifidobacterium longum E18, Escherichia coli DEC14B, Escherichia coli DEC14C, Escherichia coli DEC13E, Enterococcus faecalis TX2137, Enterococcus faecalis TX4244, Escherichia coli MS 57-2, Escherichia coli DEC5D, Escherichia coli DEC5E, Escherichia coli DEC6A, Escherichia coli DEC6B, Escherichia coli DEC6C, Escherichia coli DEC6D, Escherichia coli DEC6E, Escherichia coli DEC7A, Escherichia coli DEC7B, Escherichia coli DEC7C, Aeromonas veronii AMC34, Aeromonas hydrophila SSU, Escherichia coli DEC ID, Escherichia coli DEC1E, Escherichia coli DEC1B, Escherichia coli DEC1C, Escherichia coli DEC1A, Bamesiella intestinihominis YET 11860, Slackia piriformis YIT 12062, Helicobacter pylori GAM118Bi, Helicobacter pylori GAM115Ai, Helicobacter pylori GAM114Ai, Helicobacter pylori GAM112Ai, Helicobacter pylori GAM105Ai, Helicobacter pylori GAM103Bi, Helicobacter pylori GAMlOlBiv, Helicobacter pylori GAM120Ai, Helicobacter pylori GAM119Bi, Campylobacter jejuni jejuni 2008-979, Shigella dysenteriae CDC 74-1112, Helicobacter pylori CPY6311, Helicobacter pylori NQ4216, Helicobacter pylori NQ4200, Helicobacter pylori NQ4228, Helicobacter pylori CPY6081, Helicobacter pylori CPY6261, Helicobacter pylori CPY6271, Helicobacter pylori NQ4099, Helicobacter pylori NQ4053, Escherichia coli DEC3D, Escherichia coli DEC3E, Escherichia coli DEC2C, Escherichia coli DEC2D, Escherichia coli DEC2A, Escherichia coli DEC2B, Escherichia coli DEC3C, Escherichia coli DEC3B, Escherichia coli DEC2E, Escherichia coli DEC3A, Escherichia coli O104:H4 11-4522, Helicobacter pylori Hp H-41, Helicobacter pylori Hp H-36, Helicobacter pylori Hp H-30, Helicobacter pylori Hp H-27, Helicobacter pylori Hp H-28, Helicobacter pylori Hp H-24, Helicobacter pylori HP250ASi, Helicobacter pylori HP250AFiV, Helicobacter pylori GAMchJsl24i, Helicobacter pylori GAMchJsl l7Ai, Helicobacter pylori GAMchJsl l4i, Helicobacter pylori HP250AFiii, Helicobacter pylori HP250AFii, Helicobacter pylori

HP116Bi, Helicobacter pylori GAMchJsl36i, Bifidobacterium longum longum 44B,

Escherichia coli HM605, Helicobacter pylori GAM260Bi, Helicobacter pylori GAM260ASi, Helicobacter pylori GAM263BFi, Helicobacter pylori GAM260BSi, Helicobacter pylori GAM252Bi, Helicobacter pylori GAM250T, Helicobacter pylori GAM254Ai, Helicobacter pylori GAM252T, Helicobacter pylori GAM265BSii, Helicobacter pylori GAM264Ai, Bacteroides sp. HPS0048, Enterococcus faecium ERV165, Enterococcus faecium TX1337RF, Escherichia coli MS 85-1, Escherichia coli MS 79-10, Listeria monocytogenes sv. l/2b FSL R2-503, Listeria monocytogenes FSL J 1-194, Bacteroides salyersiae WAL 10018, DSM 18765, JCM 12988, Enterococcus faecalis TX1341, Enterococcus faecalis TX1302,

Escherichia coli 4_1_47FAA, Erysipelotrichaceae bacterium sp. 6_1_45, Helicobacter pylori Hp H-19, Helicobacter pylori Hp H-21, Bifidobacterium longum longum 2-2B, Escherichia coli MS 60-1, Sutterella wadsworthensis 2_1_59BFAA, Helicobacter pylori Hp P-1,

Helicobacter pylori Hp P-3, Helicobacter pylori Hp P-4, Helicobacter pylori Hp H-23,

Helicobacter pylori Hp H-34, Helicobacter pylori Hp P-8, Acinetobacter baumannii OIFC047, Helicobacter pylori Hp H-l 1, Helicobacter pylori Hp H-24b, Helicobacter pylori Hp H-24c, Helicobacter pylori Hp H-5b, Helicobacter pylori Hp P-74, Helicobacter pylori Hp P-41, Helicobacter pylori Hp P-62, Mycobacterium avium paratuberculosis S397, Shigella boydii serotype 7 AMC 4006, ATCC 9905, Anaerostipes hadrus comb. nov. VPI 82-52, DSM 3319, Helicobacter pylori Hp Ml, Escherichia coli MS 117-3, Escherichia coli MS 110-3,

Escherichia coli MS 16-3, Escherichia coli MS 153-1, Escherichia coli DEC13B, Escherichia coli DEC11E, Escherichia coli DEC12E, Enterococcus faecalis TX1346, Enterococcus faecalis TX1342, Enterococcus faecalis TX1467, Escherichia coli DEC8B, Escherichia coli DEC8A, Escherichia coli DEC7E, Escherichia coli DEC7D, Escherichia coli DEC8E, Escherichia coli DEC9C, Bacteroides oleiciplenus YIT 12058, Helicobacter pylori CPY1124, Lactobacillus rhamnosus ATCC 21052, Helicobacter pylori CPY 1662, Helicobacter pylori CPY 1313, Helicobacter pylori CPY3281, Helicobacter pylori CPY 1962, Escherichia coli DEC5C, Escherichia coli DEC4E, Escherichia coli DEC4D, Escherichia coli DEC5A, Escherichia coli DEC4F, Escherichia coli DEC4A, Escherichia coli DEC3F, Escherichia coli DEC4C,

Escherichia coli DEC4B, Helicobacter pylori Hp P-2, Clostridioides difficile 002-P50-2011, Clostridioides difficile 050-P50-2011, Helicobacter pylori GAM239Bi, Helicobacter pylori GAM244Ai, Helicobacter pylori GAM245Ai, Helicobacter pylori GAM246Ai, Helicobacter pylori GAM121Aii, Helicobacter pylori GAM201Ai, Helicobacter pylori GAM210Bi, Helicobacter pylori GAM231Ai, Helicobacter pylori GAM249T, Helicobacter pylori

GAM250AFi, Escherichia coli O104:H4 LB226692, Helicobacter pylori HR260BΪ,

Acinetobacter junii CIP 64.5, Helicobacter pylori NQ4161, Helicobacter pylori Hp A-l 1, Helicobacter pylori Hp A-20, Helicobacter pylori Hp A- 17, Lactobacillus delbrueckii lactis DSM 20072, Helicobacter pylori GAM 100 Ai, Campylobacter concisus UNSWCD,

Helicobacter pylori HP260ASii, Helicobacter pylori HP260BFii, Helicobacter pylori

HP250ASii, Helicobacter pylori HP250BFi, Helicobacter pylori HP250BFii, Helicobacter pylori HP250BFiii, Helicobacter pylori HP250BFiV, Helicobacter pylori HP250BSi,

Helicobacter pylori HP260AFi, Helicobacter pylori HP260AFii, Yokenella regensburgei ATCC 43003, Helicobacter pylori Hp P-2b, Helicobacter pylori GAM42Ai, Helicobacter pylori GAM71Ai, Helicobacter pylori GAM268Bii, Helicobacter pylori GAM270ASi, Helicobacter pylori GAM83T, Helicobacter pylori GAM93Bi, Helicobacter pylori GAM80Ai, Helicobacter pylori GAM83Bi, Helicobacter pylori GAM96Ai, Helicobacter pylori

GAMchJsl06B, Bifidobacterium breve 26M2, Listeria innocua ATCC 33091, Escherichia coli Nissle 1917, Helicobacter pylori Hp A-27, Helicobacter pylori Hp A-26, Clostridioides difficile 70-100-2010, Butyricicoccus pullicaecorum 1.2, Helicobacter pylori Hp H-45, Helicobacter pylori Hp H-44, Helicobacter pylori Hp A- 16, Helicobacter pylori Hp A-8, Helicobacter pylori Hp A-6, Klebsiella pneumoniae WGLW5, Klebsiella pneumoniae

WGLW3, Proteus mirabilis WGLW6, Bifidobacterium longum 1-6B, Clostridium perfringens WAL- 14572, Clostridiales sp. OBRC5-5, Helicobacter pylori Hp H-42, Helicobacter pylori Hp H-29, Helicobacter pylori Hp H-16, Lactobacillus saerimneri 30a, Helicobacter pylori Hp P- 26, Helicobacter pylori Hp P-25, Helicobacter pylori Hp P-23, Helicobacter pylori Hp P-16, Helicobacter pylori Hp P-15, Helicobacter pylori Hp P-13, Helicobacter pylori Hp P-11, Helicobacter pylori Hp P-30, Helicobacter pylori Hp H-l, Helicobacter pylori Hp H-3, Helicobacter pylori Hp H-4, Helicobacter pylori Hp H-6, Helicobacter pylori Hp H-9,

Helicobacter pylori Hp H-10, Helicobacter pylori Hp H-l 8, Helicobacter pylori Hp P-lb, Helicobacter pylori Hp P-3b, Helicobacter pylori Hp P-4d, Helicobacter pylori Hp P-4c, Helicobacter pylori Hp P-8b, Helicobacter pylori Hp P-13b, Helicobacter pylori Hp P-1 lb, Lactobacillus johnsonii pfOl GCA_000219475, Helicobacter pylori Hp A- 14, Helicobacter pylori Hp M5, Helicobacter pylori Hp M4, Helicobacter pylori Hp M3, Helicobacter pylori Hp M2, Helicobacter pylori Hp M9, Helicobacter pylori Hp M6, Fusobacterium nucleatum animalis ATCC 51191, Bifidobacterium animalis lactis Bi-07, Listeria monocytogenes SLCC 7179, Cedecea davisae 005, DSM 4568, Corynebacterium sp. HFH0082, Paenibacillus sp. HGH0039, Coprococcus sp. HPP0048, Coprococcus sp. HPP0074, Streptomyces sp.

HGB0020, Paenisporosarcina sp. HGH0030, Prevotella oralis HGA0225, Sutterella

wadsworthensis HGA0223, Veillonella sp. HPA0037, Actinomyces sp. HPA0247,

Acidaminococcus sp. HPA0509, Streptococcus sp. HPH0090, Bifidobacterium breve

HPH0326, Dermabacter sp. HFH0086, Propionibacterium sp. HGH0353, Streptomyces sp. HPH0547, Bifidobacterium bifidum ATCC 29521, Helicobacter pylori GAM117Ai,

Enterococcus faecalis V583, Clostridioides difficile CD196, Blautia sp. KLE 1732, Listeria monocytogenes L028, Listeria monocytogenes FSL Jl-208, Enterococcus faecalis V583, Enterococcus asini ATCC 700915, Megasphaera sp. NM10, Streptococcus sp. I-G2,

Streptococcus sp. I-P16, Listeria monocytogenes FSL Jl-208, Listeria monocytogenes FSL F2- 515, Coprobacillus cateniformis D6, Escherichia coli K-12 MG1655star, Campylobacter jejuni jejuni H22082, Listeria monocytogenes sv. 4b Scott A, Mycobacterium avium paratuberculosis CLH644, Kocuria rhizophila P7-4, Bifidobacterium breve DPC 6330, Metakosakonia massiliensis JC163, Alistipes senegalensis JC50, Anaerococcus senegalensis JC48,

Peptoniphilus senegalensis JC140, Brevibacterium senegalense JC43, Kurthia massiliensis JC30, Bacillus timonensis MM10403188, Paenibacillus senegalensis JC66, Dielma fastidiosa JC118, Senegalimassilia anaerobia JC110, Anaerococcus vaginalis PH9, Peptoniphilus grossensis ph5, Brevibacillus massiliensis phR, Enorma massiliensis phi, Alistipes obesi ph8, Cellulomonas massiliensis JC225, Timonella senegalensis JC301, Noviherbaspirillum massiliense JC206, Bacillus massiliosenegalensis JC6, Bacillus massilioanorexius AP8, Megasphaera massiliensis NP3, Peptoniphilus obesi phi, Edwardsiella tarda 080813, Edwardsiella tarda ATCC 15947, Mycobacterium avium paratuberculosis Ptl39, Mycobacterium avium paratuberculosis Ptl44, Mycobacterium avium paratuberculosis Ptl45, Mycobacterium avium paratuberculosis Ptl46, Mycobacterium avium paratuberculosis Ptl54, Mycobacterium avium paratuberculosis Ptl55, Mycobacterium avium paratuberculosis Ptl64, Mycobacterium avium paratuberculosis CLU623, Escherichia coli O104:H4 TY-2482, Mycobacterium avium paratuberculosis CLU361, Listeria monocytogenes FSL J2-003, Escherichia coli O104:H4 HI 12180541, Clostridioides difficile CD37, Oceanobacillus massiliensis Ndiop, Mycobacterium avium paratuberculosis 4B, Holdemania massiliensis AP2, Collinsella massiliensis GD3, Fusobacterium nucleatum subsp. animalis 4_8, Streptococcus anginosus 0051, Streptococcus anginosus C238, Fusobacterium nucleatum vincentii

3_1_36A2, Bacteroides sp. 3_2_5, Oscillibacter sp. 40911, Fusobacterium nucleatum 13_3C, Helicobacter macacae MIT 99-5501, Helicobacter canis NCTC 12740, Cetobacterium somerae ATCC BAA-474, Faecalitalea cylindroides ATCC 27803, Pseudomonas sp. HPB0071, Bifidobacterium breve S27, Listeria monocytogenes sv. 7 SLCC 2482, Vibrio cholerae V51, Vibrio cholerae sv. 037 V52, Listeria monocytogenes sv. l/2a F6900, Listeria monocytogenes sv. l/2a J2818, Subdoligranulum variabile DSM 15176, Clostridium sp M62/1, Marvinbryantia formatexigens 1-52, Citrobacter youngae ATCC 29220, Fusobacterium ulcerans 12- IB, Blautia hansenii VPI C7-24, DSM 20583, Providencia rettgeri DSM 1131, Ralstonia sp. 5_2_56FAA, Bacteroides finegoldii DSM 17565, Bifidobacterium angulatum DSM 20098, Bifidobacterium gallicum DSM 20093, LMG 11596, Bifidobacterium pseudocatenulatum DSM 20438, JCM 1200, LMG 10505, Collinsella intestinalis DSM 13280, Desulfovibrio sp. 3_l_syn3,

Enterobacter cancerogenus ATCC 35316, Escherichia sp. 4_1_40B, Fusobacterium nucleatum subsp. animalis 3_1_33, Fusobacterium ulcerans ATCC 49185, Helicobacter bilis ATCC 43879, Listeria grayi DSM 20601, Mitsuokella multacida DSM 20544, Roseburia intestinalis Ll-82, Ruminococcus sp. 5_1_39BFAA, Erysipelotrichaceae bacterium 5_2_54FAA,

Enterobacteriaceae bacterium 9_2_54FAA, Prevotella copri CB7, DSM 18205, Bacteroides sp. D2, Bilophila wadsworthia 3_1_6, Fusobacterium necrophorum D12, Lactobacillus plantarum ATCC 14917, Ruminococcaceae bacterium D16, Lachnospiraceae bacterium

3_1_57FAA_CT1, Lachnospiraceae bacterium 2_1_46FAA, Fusobacterium nucleatum vincentii 3_1_27, Escherichia coli O104:H4 GOS1, Escherichia coli O104:H4 GOS2,

Aneurinibacillus aneurinilyticus ATCC 12856, Fusobacterium nucleatum CTI-1,

Fusobacterium nucleatum CTI-2, Fusobacterium nucleatum CTI-3, Providencia alcalifaciens F90-2004, Providencia alcalifaciens 205/92, Providencia alcalifaciens R90-1475, Providencia alcalifaciens RIMD 1656011, Providencia alcalifaciens PAL-1, Providencia alcalifaciens PAL- 2, Providencia alcalifaciens PAL-3, Klebsiella oxytoca OK-1, Klebsiella oxytoca KA-2, Coprococcus sp. ART55/1, Eubacterium siraeum V10Sc8a, Fusobacterium nucleatum CTI-6, Fusobacterium nucleatum CTI-7, Fusobacterium nucleatum CTI-5, Bifidobacterium animalis RH, Enterococcus faecalis 918, Campylobacter jejuni jejuni HB93-13, Ruminococcus bicirculans 80/3, Klebsiella oxytoca KONIH1, Bifidobacterium longum BXY01, Escherichia coli Nissle 1917, Serratia marcescens marcescens Dbl l, Enorma timonensis GD5, Bacteroides salyersiae WAL 10018, DSM 18765, JCM 12988, Intestinibacter bartlettii DSM 16795, Eubacterium ruminantium ATCC 17233, Clostridioides difficile 630, Porphyromonas sp. 31_2, Klebsiella pneumoniae pneumoniae KPNIH28, Bifidobacterium breve JCM 7019, Escherichia coli B7A, Hafnia alvei Stuart 32011, ATCC 13337, Flavonifractor plautii 1_3_50AFAA, Collinsella sp. 4_8_47FAA, Shigella flexneri 2003036 (contamination screened), Shigella flexneri Shi06HN006 (contamination screened), Salmonella enterica enterica sv. typhimurium ATCC 13311, Vibrio cholerae sv. 01 bv. El Tor MAK676, Ruminococcus callidus ATCC 27760, Lachnoclostridium symbiosum ATCC 14940, Clostridium sp. ATCC BAA-442, Eubacterium ramulus VPI C6-27, ATCC 29099, Lactobacillus brevis ATCC 14869,

Marvinbryantia formatexigens 1-52, Clostridium sp. VPI C48-50, Oscillibacter sp. KLE 1728, Oscillibacter sp. KLE 1745, Clostridium sp. KLE 1755, Clostridium ihumii AP5,

Corynebacterium ihumii GD7, Escherichia albertii GTC 14781, Nesterenkonia massiliensis NP1, Clostridium bolteae WAL-14578 JGI Assembly and Annotation, Bacteroides neonati MS4, Bacteroides xylanisolvens SD CC lb, Gorillibacterium massiliense G5, Vibrio cholerae sv. 01 bv. El Tor 5/66, Helicobacter heilmannii ASB1.4, Peptoniphilus timonensis JC401, Akkermansia muciniphila ATCC BAA-835, Clostridium saudiense JCC, Clostridium jeddahense JCD, Eggerthella lenta 1_1_60AFAA, Parabacteroides sp. HGS0025,

Parabacteroides goldsteinii WAL 12034, Clostridiales sp. SM4/1, Corynebacterium

ammoniagenes DSM 20306 Genome sequencing, Bacteroides cellulosilyticus WH2,

Lactobacillus casei casei ATCC 393, Bacteroides sp. 3_1_13, Klebsiella oxytoca 09-7231, Parabacteroides sp. D26, Lachnoclostridium citroniae WAL-19142, Coprobacillus sp.

8_1_38FAA, Vibrio cholerae sv. 014 MZO-2, Aeromicrobium massiliense JCM, Dorea sp. D27, Klebsiella oxytoca 10-5244, Bacteroides eggerthii 1_2_48FAA, Escherichia coli

ST540an, Lachnoclostridium clostridioforme WAL-7855, Clostridium sp. 1_1_41A1FAA, Escherichia coli ST540a, Escherichia coli ST2747, Fusobacterium nucleatum animalis 7_1, Eubacterium sp. 3_1_31, Bifidobacterium bifidum NCIMB 41171, Fusobacterium nucleatum subsp. animalis 21_1A, Helicobacter pylori J99, Lactobacillus antri DSM 16041, Lactobacillus amylolyticus DSM 11664, Vibrio fluvialis NBRC 103150, Bacteroides ovatus ATCC 8483, Bacteroides clarus DSM 22519, Lactobacillus ultunensis DSM 16047, Bifidobacterium angulatum DSM 20098, Vibrio cholerae ZWU0020, Escherichia coli D9, Bifidobacterium breve DSM 20213, JCM 1192, Coprobacillus cateniformis JCM 10604, Ruminococcus faecis JCM 15917, Dielma fastidiosa DSM 26099, Blautia hansenii DSM 20583, Citrobacter portucalensis 30_2, Oxalobacter formigenes OXCC13, bacterium MS4, bacterium OL-1, Akkermansia sp. KLE1605, Clostridium sp. CL-2, Clostridium amazonitimonense LF2, Bacillus andreraoultii KW-12, Helicobacter pylori B508A-T4, Flavonifractor plautii DSM 6740, Pantoea septica LMG 5345, Helicobacter canadensis MIT 98-5491, ATCC 700968, Bacteroides caccae ATCC 43185, Fusobacterium mortiferum ATCC 9817, Fusobacterium gonidiaformans ATCC 25563, Fusobacterium ulcerans ATCC 49185, Blautia sp. Marseille- P2398, Ruminococcus albus, Clostridium sp. CAG:413, Lachnospiraceae str. KH1P17, Blautia sp. An81, Blautia wexlerae, Lachnospiraceae str. LC2019, Ruminococcus sp. CAG:488, Clostridium sp. L2-50, and Clostridium sp. CAG:253.

Acyl carrier protein (ACP)

Acyl carrier proteins (ACPs or T proteins) comprise a 4’-phosphopantetheine moiety and are capable of catalyzing the addition of a thioester to the phosphopantetheine moiety. The 4’-phosphopantetheine moiety may be post-translationally added to a serine residue on an ACP by a 4'-phosphopantetheinyl transferase (PPTase). Non-limiting examples of 4'- phosphopantetheinyl transferases include Sfp (e.g., UniProtKB - P39135 (SFP_BACSU)).

In some instances, an ACP comprises a sequence that has at least 50% (e.g., at least 55%, at least 60%, at least 65%, at least 70%, at least 71%, at least 72%, at least 73%, at least

74%, at least 75%, at least 76%, at least 77%, at least 78%, at least 79%, at least 80%, at least

81%, at least 82%, at least 83%, at least 84%, at least 85%, at least 86%, at least 87%, at least

88%, at least 89%, at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least

95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100%) sequence identity with an ACP sequence shown in Table 1.

A non-limiting example of a sequence logo for ACP proteins is shown in FIG. 12. Fatty acyl-ACP synthetase

As used herein, fatty acyl-ACP synthetases or A proteins are capable of conjugating a fatty acid substrate onto an acyl carrier protein (ACP). The fatty acid substrate may be endogenous or exogenous to a host cell or composition ( e.g ., cell lysate or cell culture broth).

A fatty acyl-ACP synthetase may use any type of fatty acid. For example, a fatty acyl-ACP synthetase may use a medium-chain fatty acid or a long-chain fatty acid as a substrate.

Fatty acyl-ACP synthetases use ATP to catalyze the production of fatty acyl-ACPs.

The activity of a fatty acyl-ACP synthetase activity may be measured using suitable method known in the art or described in the Examples section below. ATP consumption by a fatty acyl-ACP synthetase may be measured using an in vitro assay. For example, 2-amino-6- mercapto-7-methylpurine riboside (MESG) substrate, which is converted to ribose 1-phosphate and 2-amino-6-mercapto-7-methylpurine product in the presence of inorganic phosphate purine nucleoside phosphorylase (PNP) enzyme, can be used to determine the amount of ATP consumed by a fatty acyl-ACP synthetase in the presence of a fatty acid substrate.

In some instances, a fatty acyl-ACP synthetase comprises a sequence that has at least 50% (e.g., at least 55%, at least 60%, at least 65%, at least 70%, at least 71%, at least 72%, at least 73%, at least 74%, at least 75%, at least 76%, at least 77%, at least 78%, at least 79%, at least 80%, at least 81%, at least 82%, at least 83%, at least 84%, at least 85%, at least 86%, at least 87%, at least 88%, at least 89%, at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100%) sequence identity with a fatty acyl-ACP synthetase sequence shown in Table 1

A non-limiting example of a sequence logo for fatty acyl-ACP synthetases (A proteins) is shown in FIGS. 13A-13D.

Fatty acids

Fatty acids are carboxylic acids with aliphatic chains. They can be classified as either saturated or unsaturated. Saturated fatty acids do not have double bonds in their aliphatic chain and have the formula CH3(CH2)nCOOH, where C is carbon, H is hydrogen, O is oxygen, and n is any non-negative integer (e.g., n may be 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 40, 50, 60, 70, 80, 90, 100 or higher, including any values in between). In contrast, unsaturated fatty acids have at least one double bond in their aliphatic chain, resulting in either a cis (hydrogen atoms on the same side of the double bond) or trans hydrogen configuration. Fatty acids may also be classified by length. Short-chain fatty acids (SCFAs) have aliphatic chains of up to 5 carbons. Medium-chain fatty acids (MCFAs) have aliphatic chains of 6 to 12 carbons (e.g., 6 carbons, 7 carbons, 8 carbons, 9 carbons, 10 carbons, 11 carbons, 12 carbons, 6 to 8 carbons, 6 to 9 carbons, 6 to 10 carbons, 6 to 11 carbons, 7 to 9 carbons, 7 to 10 carbons, 7 to 11 carbons, 7 to 12 carbons, 8 to 10 carbons, 8 to 11 carbons, 8 and 12 carbons, 9 to 11 carbons, 9 to 12 carbons, or 10 to 12 carbons), inclusive. MCFAs include hexanoic acid, octanoic acid, decanoic acid, and dodecanoic acid. Long-chain fatty acids (LCFAs) have aliphatic chains ranging from 13 to 21 carbons (e.g., 13 carbons, 14 carbons, 15 carbons, 16 carbons, 17 carbons, 18 carbons, 19 carbons, 20 carbons, 21 carbons, or 15 to 20 carbons), inclusive. Non-limiting examples of LCFAs include oleic acid, palmitoleic acid, and nervonic acid. Very long chain fatty acids (VLCFAs) comprise aliphatic chains of 22 or more carbons (e.g., at least 23, at least 25, at least 30, at least 35, at least 40, at least 50, at least 60, at least 70, at least 80, at least 90, or at least 100, including all values in between). Non-limiting examples of VLCFAs include lignoceric acid and hexacosanoic acid. Most naturally-occurring fatty acids have unbranched aliphatic chains and have an even number of carbons.

In some embodiments, a fatty acid comprises a straight-chain alkyl chain. In some embodiments, the straight-chain alkyl chain comprises 10-12 carbons.

In some embodiments, a fatty acid comprises a cis-9 double bond.

A fatty acid may be a free fatty acid. A free fatty acid is a non-esterified fatty acid (NEFA).

Non-limiting examples of fatty acids include acetic acid, propionic acid, butyric acid, valeric acid, hexanoic acid, octanoic acid, capric acid, lauric acid, myristic acid, pentadecanoic acid, palmitic acid, stearic acid, arachidic acid, iso-pentadecanoic acid, palmitoleic acid, oleic acid, vaccenic acid, linoleic acid, g-linolenic acid, a-linolenic acid, dihomo-y-linolenic acid, arachidonic acid, eicosapentaenoic acid, docosapentaenoic acid, docosahexaenoic acid, 8- methyl-6-nonenoic acid, octynoic acid, myristic acid alkyne, and palmitic acid alkyne. See, e.g., Table 3.

A fatty acid may be exogenous or endogenous in reference to a host cell or a composition. As used herein an exogenous fatty acid refers to a fatty acid that is introduced. The exogenous fatty acid may be a fatty acid that is introduced to a host cell or added to a composition (e.g., a cell culture broth or cell lysate). In some instances, the host cell is a bacterium (e.g., Escherichia coli). An exogenous fatty acid may be derived from another cell or produced synthetically. As a non-limiting example, a host cell may naturally produce a particular fatty acid but the same fatty acid may be derived from an exogenous source and the exogenous fatty acid may be introduced to the cell. In some embodiments, an exogenous fatty acid is not naturally produced by a host cell.

In some embodiments, an exogenous fatty acid is acetic acid, propionic acid, butyric acid, valeric acid, hexanoic acid, octanoic acid, capric acid, lauric acid, myristic acid, pentadecanoic acid, palmitic acid, stearic acid, arachidic acid, iso-pentadecanoic acid, palmitoleic acid, oleic acid, vaccenic acid, linoleic acid, g-linolenic acid, a-linolenic acid, dihomo-y-linolenic acid, arachidonic acid, eicosapentaenoic acid, docosapentaenoic acid, docosahexaenoic acid, 8-methyl-6-nonenoic acid, octynoic acid, myristic acid alkyne, or palmitic acid alkyne. See, e.g., Table 3.

In contrast, an endogenous fatty acid refers to a fatty acid that is present in a host cell or composition (e.g. a cell lystate or a cell culture broth). In some embodiments, an endogenous fatty acid is acetic acid, propionic acid, butyric acid, valeric acid, hexanoic acid, octanoic acid, capric acid, lauric acid, myristic acid, pentadecanoic acid, palmitic acid, stearic acid, arachidic acid, iso-pentadecanoic acid, palmitoleic acid, oleic acid, vaccenic acid, linoleic acid, g- linolenic acid, a-linolenic acid, dihomo-y-linolenic acid, arachidonic acid, eicosapentaenoic acid, docosapentaenoic acid, docosahexaenoic acid, 8-methyl-6-nonenoic acid, octynoic acid, myristic acid alkyne, or palmitic acid alkyne. See, e.g., Table 3.

Fatty acyl-transferases

The fatty acyl-transferases of the present disclosure are capable of conjugating an amine to a thiotemplated fatty acid (e.g., a fatty acyl-ACP) to produce a fatty acid amide and may be referred to as C proteins. The amine may be endogenous or exogenous to a host cell or composition (e.g., cell lysate or cell culture broth).

In some instances, a fatty acyl-transferase comprises a sequence that has at least 50% (e.g., at least 55%, at least 60%, at least 65%, at least 70%, at least 71%, at least 72%, at least

73%, at least 74%, at least 75%, at least 76%, at least 77%, at least 78%, at least 79%, at least

80%, at least 81%, at least 82%, at least 83%, at least 84%, at least 85%, at least 86%, at least

87%, at least 88%, at least 89%, at least 90%, at least 91%, at least 92%, at least 93%, at least

94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100%) sequence identity with a fatty acyl-transferase sequence shown in Table 1.

FIGS. 14A-14D show a non-limiting example of a fatty acyl-transferase (C protein) sequence logo. Amines

Amines are derivatives of ammonia, wherein one or more of the hydrogen atoms have been replaced with a substituent group (R group). Non-limiting examples of amines include phenylalanine, tryptophan, tyrosine, histidine, lysine, glycine, alanine, valine, leucine, isoleucine, methionine, proline, serine, threonine, cysteine, asparagine, glutamine, aspartic acid, glutamic acid, arginine, ornithine, b-alanine, L-DOPA, creatine, citrulline,

phenylacetylglutamine, phenylethylamine, tryptamine, tyramine, histamine, serotonin, dopamine, epinephrine, norepinephrine, g-aminobutryic acid (GABA), aminovaleric acid, ethanolamine, cadaverine, putrescine, spermine, spermidine, agmatine, propylamine, butylamine, dimethylamine, pyrollidine, piperidine, homocysteine, cysteamine,

homocysteamine, taurine, hypotaurine, glutathione, octopamine, 3-iodothyronamine, melatonin, and vanillylamide. See, e.g., Table 3.

An amine may be endogenous or exogenous to a host cell or a composition. As used herein an exogenous amine refers to an amine that is introduced. The exogenous amine may be an amine that is introduced to a host cell or added to a composition (e.g., a cell culture broth or cell lysate). In some instances, the host cell is a bacterium (e.g., Escherichia coli). An exogenous amine may be derived from another cell or produced synthetically. As a non limiting example, a host cell may naturally produce a particular amine but the same amine may be derived from an exogenous source and the exogenous amine may be introduced to the cell. In some instances, an exogenous amine is not naturally produced by a particular host cell.

Variants

Variants of the sequences (e.g., biosynthetic enzymes), including nucleic acid or amino acid sequences described herein are also encompassed by the present disclosure. A variant may share at least 5%, at least 10%, at least 15%, at least 20%, at least 25%, at least 30%, at least 35%, at least 40%, at least 45%, at least 50%, at least 55%, at least 60%, at least 65%, at least 70%, at least 71%, at least 72%, at least 73%, at least 74%, at least 75%, at least 76%, at least 77%, at least 78%, at least 79%, at least 80%, at least 81%, at least 82%, at least 83%, at least 84%, at least 85%, at least 86%, at least 87%, at least 88%, at least 89%, at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100% sequence identity with a reference sequence, including all values in between. Sequence identity to a particular reference sequence may be determined by sequence alignment programs and parameters described herein and known to those skilled in the art.

The comparison of sequences and determination of percent identity between two sequences can be accomplished using a mathematical algorithm. Techniques for determining identity are codified in publicly available computer programs. Exemplary computer software to determine homology between two sequences include, but are not limited to, the GCG program package (Devereux, J. et al. Nucleic Acids Research, 12(1): 387, 1984), the BLAST suite (Altschul, S. F. et al. Nucleic Acids Res. 25: 3389, 1997), and FASTA (Altschul, S. F. et al. J. Molec. Biol. 215: 403, 1990). Other techniques include: the Smith- Waterman algorithm (Smith, T.F. et al. J. Mol. Biol. 147: 195, 1981); the Needleman-Wunsch algorithm

(Needleman, S.B. et al. J. Mol. Biol. 48: 443, 1970); and the Fast Optimal Global Sequence Alignment Algorithm (FOGSAA) (Chakraborty, A. et al. Sci Rep.3: 1746, 2013).

In some instances, a biosynthetic enzyme comprises a conservative amino acid substitution relative to a reference sequence disclosed herein. A conservative amino acid substitution is an amino acid substitution that does not alter the relative charge or size characteristics of the protein or peptide in which the amino acid substitution is made.

Conservative substitutions of amino acids include substitutions made amongst amino acids within the following groups: (a) M, I, L, V; (b) F, Y, W; (c) K, R, H; (d) A, G; (e) S, T; (f) Q, N; and (g) E, D.

Variants can be prepared according to methods for altering polypeptide sequences known to one of ordinary skill in the art such as those are found in references that compile such methods (see, e.g., Current Protocols in Molecular Biology, Ausubel, F.M., et al., New York: John Wiley & Sons, 2006; Molecular Cloning: A Laboratory Manual, Green, M.R. and Sambrook J., New York: Cold Spring Harbor Laboratory Press, 2012; Gibson, D.G., et al., Nature Methods 6(5):343-345 (2009), the teachings of which relating to polypeptide preparation and modifications are herein incorporated by reference).

As a non-limiting example, any of the biosynthetic enzymes may be from one of the homologous pathways listed in Table 4.

Table 4. Homologous pathways in non-redundant database

Vectors

Provided herein, in some embodiments, are vectors encoding human gut microbiome- derived bacterium genes involved in biosynthesis. In some embodiments, a vector encodes at least one human gut microbiome-derived bacterium gene selected from a fatty acyl transferase, an acyl carrier protein, and a fatty acyl-ACP synthetase. In some embodiments, a vector encodes at least two human gut microbiome-derived bacterium genes selected from a fatty acyl transferase, an acyl carrier protein, and a fatty acyl-ACP synthetase. In some embodiments, a vector encodes at least three human gut microbiome-derived bacterium genes selected from a fatty acyl transferase, an acyl carrier protein, and a fatty acyl-ACP synthetase. A vector may further encode a hydrolase, a lipid transfer protein, a glycosyltransferase, or any combination thereof.

In some embodiments, a vector comprises an open reading frame encoding at least one biosynthetic enzyme. An open reading frame is a nucleic acid sequence that is transcribed continuously into an mRNA molecule, and then translated continuously into an amino acid sequence, uninterrupted by stop codons. The translated open reading frame may be all or a portion of a gene encoding a protein or polypeptide.

The vectors comprise at least one regulatory element controlling expression of at least one biosynthetic gene from a human gut microbiome-derived bacterium. The regulatory element may be a promoter. In some instances, all biosynthetic genes in a vector are in a single operon and are regulated by one regulatory element. In some instances, at least 2, 3, 4, 5, 6,7, 8, 9, 10, or more biosynthetic genes in a vector are in a single operon and are regulated by one regulatory element. A vector may comprise 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, or more operons.

For example, a vector may comprise at least one promoter operably linked to a nucleic acid encoding a fatty acyl transferase, an acyl carrier protein, a fatty acyl-ACP synthetase, or any combination thereof. In some embodiments, a vector comprises one promoter operably linked to a nucleic acid encoding a human gut microbiome-derived bacterium fatty acyl transferase, a human gut microbiome-derived bacterium acyl carrier protein, and a human gut microbiome-derived bacterium fatty acyl-ACP synthetase. In some embodiments, nucleic acids encoding the fatty acyl transferase, the acyl carrier protein, and the fatty acyl-ACP synthetase are each operably linked to a separate promoter in a vector.

A vector may comprise a nucleic acid encoding a biosynthetic enzyme from a human gut microbiome-derived bacterium, and the nucleic acid may be codon optimized for heterologous expression in a host cell. For example, the nucleic acid may be codon optimized for expression in an E. coli host cell. A codon usage database may be used to improve expression of a heterologous sequence in a host cell. See, e.g., Gustafsson el al, Trends Biotechnol. 2004 Jul;22(7):346-53.

A promoter is a control region of a nucleic acid sequence through which initiation and rate of transcription of the remainder of a nucleic acid sequence are controlled. A promoter may also contain sub-regions at which regulatory proteins and molecules, such as RNA polymerase and other transcription factors, may bind. A promoter drives expression or drives transcription of the nucleic acid sequence that it regulates. A promoter is considered to be “operably linked” to a nucleotide sequence when it is in a correct functional location and orientation in relation to the nucleotide sequence to control (“drive”) transcriptional initiation and/or expression of that sequence. Promoters may be constitutive or inducible. An inducible promoter is a promoter that is regulated (e.g., activated or inactivated) by the presence or absence of a particular factor. Inducible promoters for use in accordance with the present disclosure include those that function in bacteria. An exemplary inducible promoter for use herein is an isopropyl b-D-l- thiogalactopyranoside (IPTG)-inducible promoter or a tetracycline-inducible promoter (e.g., when used with a reverse tetracycline-controlled transactivator (rtTA)). In some embodiments, the IPTG-inducible promoter is an inducible T7 promoter. The vector may further encode a lac repressor (Lad).

In some embodiments, a vector may encode a repressible promoter, which inactivates expression of an operably linked nucleic acid in the presence of a factor. For example, a tetracycline-sensitive promoter may be used with a tetracycline-controlled transactivator (tTA), such that in the presence of tetracycline, the tTA binds to the tetracycline- sensitive promoter and prevents expression of the operably linked nucleic acid.

In some embodiments, a vector further encodes a selection marker. Suitable selection markers include antibiotic resistance genes (e.g., genes encoding resistance to kanamaycin, spectinomycin, ampicillin, carbenicillin, bleomycin, chloramphenicol, coumermycin, gentamycin, tetracycline, or any combination thereof).

In some instances, a vector encodes a ribosomal binding site (RBS). Ribosomal binding sites promote ribosomal recruitment during initiation of protein translation and are located upstream of a start codon. In prokaryotes, the RBS may be referred to as a Shine- Dalgamo sequence. A RBS sequence may comprise AGGAGG. As a non-limiting example, a RBS for use in E. coli may comprise the sequence AGGAGGU.

Examples of vectors for expressing include, but are not limited to, plasmids, phagemids and bacterial artificial chromosomes (BACs). The vectors of the present disclosure may be generated using standard molecular cloning methods (see, e.g., Current Protocols in Molecular Biology, Ausubel, F.M., et ah, New York: John Wiley & Sons, 2006; Molecular Cloning: A Laboratory Manual, Green, M.R. and Sambrook J., New York: Cold Spring Harbor Laboratory Press, 2012; Gibson, D.G., et ah, Nature Methods 6(5):343-345 (2009), the teachings of which relating to molecular cloning are herein incorporated by reference).

Without further elaboration, it is believed that one skilled in the art can, based on the above description, utilize the present invention to its fullest extent. The following specific embodiments are, therefore, to be construed as merely illustrative, and not limitative of the remainder of the disclosure in any way whatsoever. All publications cited herein are incorporated by reference for the purposes or subject matter referenced herein. Engineered cells

Any of the vectors described herein may be introduced into a host cell to produce an engineered cell using routine methods known in the art. An engineered cell comprises at least one engineered nucleic acid or is otherwise structurally or functionally distinct from a wild- type counterpart. Thus, a host cell comprising a vector encoding at least one biosynthetic enzyme that is from a human gut microbiome-derived bacterium is considered an engineered cell. In some instances, an engineered cell comprises at least three human gut microbiome- derived bacterium biosynthetic enzymes, including a fatty-acyl transferase, an acyl carrier protein synthase, and a fatty acyl-ACP synthetase. In some instances, all three enzymes (a fatty-acyl transferase, an acyl carrier protein synthase, and a fatty acyl-ACP synthetase) are derived from the same species of human gut microbiome bacterium. In some embodiments, at least two of the three enzymes (a fatty-acyl transferase, an acyl carrier protein synthase, and/or a fatty acyl-ACP synthetase) are derived from the same species of human gut microbiome bacterium. In some embodiments, all three enzymes (a fatty-acyl transferase, an acyl carrier protein synthase, and a fatty acyl-ACP synthetase) are derived from distinct species of human gut microbiome bacteria. In some embodiments, one or more of the enzymes (a fatty-acyl transferase, an acyl carrier protein synthase, and/or a fatty acyl-ACP synthetase) are derived from the same species of human gut microbiome bacterium, and other one or more of the enzymes are derived from a different one or more species of human gut microbiome bacteria. In some instances, the human gut microbiome bacterium is from the Clostridia class.

The engineered cells of the present disclosure may further comprise other biosynthetic enzymes from human gut microbiome-derived bacterium. For example, the engineered cells may comprise a hydrolase, a lipid transfer protein, a glycosyltransferase, or any combination thereof. Lipid transfer proteins, which may also be referred to as sterol carrier proteins or sterol transfer proteins, are capable of transferring steroids between cellular membranes. See, e.g., Pfam Identifier: PF02036. The hydrolases may be from the alpha/beta hydrolase superfamily and comprise eight beta strands connected by 6 alpha helices. See, e.g., Pfam Identifier: PF12146. Glycosyltransferases are capable of catalyzing the trasfer of sugars from donor molecules to acceptor molecules. See, e.g., Lozupone el al, Proc Natl Acad Sci U S A. 2008 Sep 30; 105(39): 15076-81 ; Bhattacharya et al, PLoS One. 2015 Nov 6;10(l l):e0142038; Brockhausen, Front Immunol. 2014 Oct 20;5:492.

A non-limiting example of a glycosyltransferase is: EreG: (E. rectale pathway): MHN QT AILFFYKGKKMKLLS FAIPC YN S KD YMEHCIES ILPGGDD VEIIIVDDGS KDETAAIADRYAAEYPDIVKAVHQENGGHGEAVNTGLKNATGKYFKVVDSD DWVDLDSYKKILDKLREFEQENTQIDMLLANYVYEKEGAKRKKVMRQTGFPR NEIFT W S DIKHIYKGH YILMHS VIYRTELLRS C GLKLPKHTF Y VDNIY V YKPLP Y VRTMYYLDVDFYRYFIGRDDQSVNEQVMIRRIDQQIRVNKIMFDDVKLHEITN EMCRKYM Y S YLEIITTIS TILAIIS GTDENM AKKDELW A YMKEHDEET YKKLRH G VMGQLMNLPGKGGRKV AIG A YKLS QKV V GFN (SEQ ID NO: 145)

In some instances, at least two, at least three, at least four, at least five, at least six, at least seven, at least eight, at least nine, at least ten, or more of the biosynthetic enzymes in an engineered cell are from the same species of human gut microbiome-derived bacterium. In some instances, all biosynthetic enzymes from human gut microbiome-derived bacteria are from different species of human gut microbiome-derived bacteria. In some instances, all biosynthetic enzymes from human gut microbiome-derived bacteria are from the same species of human gut microbiome-derived bacteria.

The host cell may be, for example, Yersinia spp., Escherichia spp., Klebsiella spp., Acinetobacter spp., Bordetella spp., Neisseria spp., Aeromonas spp., Franciesella spp., Corynebacterium spp., Citrobacter spp., Chlamydia spp., Hemophilus spp., Brucella spp., Mycobacterium spp., Legionella spp., Rhodococcus spp., Pseudomonas spp., Helicobacter spp., Salmonella spp., Vibrio spp., Bacillus spp., Erysipelothrix spp., Salmonella spp.,

Streptomyces spp., Bacteroides spp., Prevotella spp., Clostridium spp., Bifidobacterium spp., or Lactobacillus spp.

In some instances, the host cell is a yeast cell.

The engineered cells of the present disclosure may be propagated under conditions well known in the art (e.g. temperature, culture broth and incubation times). In some embodiments, in which the engineered cells comprise nucleic acids operably linked to inducible promoters, the engineered cells are cultured in the presence of an effective amount inducing agent to induce expression from the inducible promoter. In some embodiments, the inducible promoter driving expression of a nucleic acid encoding a heterologous protein is a IPTG-inducible promoter, thus, the engineered cells are cultured in an effective amount of IPTG to induce expression from the tetracycline-inducible promoter. In some embodiments, a vector encoding T7 RNA polymerase may also be introduced into a cell with a IPTG-inducible promoter. In some embodiments, the inducible promoter driving expression of a nucleic acid encoding a heterologous protein is a tetracycline-inducible promoter, thus, the engineered cells are cultured in an effective amount of tetracycline to induce expression from the tetracycline- inducible promoter.

Methods of producing fatty acid amides

Some aspects of the present disclosure provide methods for producing a fatty acid amide using human gut microbiome-derived bacterium biosynthetic enzymes. Biosynthetic enzymes include fatty acyl-transferase, acyl carrier protein (ACP), fatty acyl-ACP synthetase, hydrolases, lipid transfer proteins, and glycosyltransferases.

The methods comprise contacting a composition comprising at least one fatty acid and at least one amine with one or more of the biosynthetic enzymes described herein. A set of biosynthetic enzymes (e.g., at least 2, at least 3, at least 4, at least 5, at least 6, at least 7, at least 8, at least 9, at least 10, or more) may be used to produce fatty acid amides. In some instances, at least 2, at least 3, at least 4, at least 5, at least 6, at least 7, at least 8, at least 9, at least 10, or more of the biosynthetic enzymes are from the same species of human gut microbiome-derived bacterium. In some instances, the set of biosynthetic enzymes may be all from the same species of human gut microbiome-derived bacterium. In some instances, the set of biosynthetic enzymes are all from different species of human gut microbiome-derived bacterium. In some embodiments, the set of biosynthetic enzymes comprises a fatty acyl- transferase, an acyl carrier protein (ACP), and a fatty acyl-ACP synthetase.

For example, at least 1, at least 2, at least 3, at least 4, at least 5, at least 6, at least 7, at least 8, at least 9, at least 10, or more of the microbiome-derived bacterium biosynthetic enzymes in a set may be from the Clostridia class. In sets with more than one Clostridia biosynthetic enzyme, at least 1, at least 2, at least 3, at least 4, at least 5, at least 6, at least 7, at least 8, at least 9, at least 10, or more of the biosynthetic enzymes may be from the same order, family, genus, species, or any combination thereof.

The fatty acid may be a free fatty acid. In some instances, at least 1, at least 2, at least 3, at least 4, at least 5, at least 6, at least 7, at least 8, at least 9, at least 10, at least 15, at least 16, at least 17, at least 18, at least 19, at least 20, at least 30, at least 40, at least 50, at least 60, at least 70, at least 80, at least 90, at least 100, or more fatty acids, including all values in between, may be used. A composition may comprise fatty acids that are all of the same type, fatty acids that are all different, or some fatty acids that are the same and some fatty acids that are different. In some instances, at least 1, at least 2, at least 3, at least 4, at least 5, at least 6, at least 7, at least 8, at least 9, at least 10, at least 15, at least 16, at least 17, at least 18, at least 19, at least 20, at least 30, at least 40, at least 50, at least 60, at least 70, at least 80, at least 90, at least 100, or more fatty acids, including all values in between, may be used.

A composition may comprise amines that are all of the same type, amines that are all different, or some amines that are the same and some amines that are different. In some instances, at least 1, at least 2, at least 3, at least 4, at least 5, at least 6, at least 7, at least 8, at least 9, at least 10, at least 15, at least 16, at least 17, at least 18, at least 19, at least 20, at least 30, at least 40, at least 50, at least 60, at least 70, at least 80, at least 90, at least 100, or more amines, including all values in between, may be used.

As one of ordinary skill in the art would appreciate, selection of a particular fatty acid and amine pair may be dependent upon the fatty acid amide of interest. As a non-limiting example, palmitoeyl putrescine is a fatty acid amide (FAA) that is formed by the conjugation of palmitoleic acid and putrescine. To make palmitoeyl putrescine, palmitoleic acid would be selected as the fatty acid and putrescine would be selected as the amine.

The methods may comprise using one or more more purified biosynthetic enzymes, a lysate comprising one or more biosynthetic enzymes, or culturing a cell comprising a vector encoding one or more biosynthetic enzymes.

Protein purification methods include, but are not limited, to size exclusion

chromatography, ammonium sulfate precipitation, ion exchange chromatography, immobilized metal chelate chromatography, thiophilic adsorption, melon gel chromatography and antibody ligand chromatography, any of which may be used as provided herein to recover a protein

The engineered cells of the present disclosure may be propagated under conditions well known in the art (e.g. temperature, culture media and incubation times). In some

embodiments, in which the engineered cell comprises nucleic acids operably linked to inducible promoters, the engineered cells are cultured in the presence of an effective amount inducing agent to induce expression from the inducible promoter. In some embodiments, the inducible promoter driving expression of a nucleic acid encoding a heterologous protein is a IPTG-inducible promoter, thus, the engineered cells are cultured in an effective amount of IPTG to induce expression from the IPTG inducible promoter.

Any of the compounds described herein, including fatty acids, amines, and fatty acid amides, disclosed herein may be identified and extracted using any method known in the art. Mass spectrometry (e.g., LC-MS, GC-MS) is a non-limiting example of a method for identification and may be used to extract a compound of interest. EXAMPLES

EXAMPLE 1

Clostridia NRPS-like Pathways Identified from Human Gut Sequencing Data

The human gut microbiome genomic datasets were systematically screening for the presence of metabolite-encoding biosynthetic pathways (FIG. 1A). Metadata from the Joint Genome Institute (JGI) was used to identify 1042 bacterial genomic datasets associated with the human gastrointestinal tract, which were retrieved from either JGI or the National Center for Biotechnology Information (NCBI) database. The software antiSMASH, ²² which employs a pattern recognition algorithm based on protein domains from known natural product pathways, was used to bioinformatically detect 3531 putative pathways from the genomic datasets. The pathways were further narrowed to those that are prevalent and actively transcribed in healthy human subjects, reasoning that they are more likely encode for small molecule modulators that are relevant and mechanistically conserved across different human gut strains. To screen by prevalence, the antiSMASH-detected pathways were analyzed for their representation in the shotgun metagenomic sequencing data collected from the Human Microbiome Project (HMP; 148 gut samples of healthy subjects). ²⁰ The resulting 2013 human gut-associated pathways were narrowed based on transcription by examining their presence in RNA sequencing datasets from the stool samples of eight healthy subjects from the HMP. ³² The 336 transcribed pathways were organized using the software BiG-SCAPE, ³³ which uses network- similarity algorithms to group the pathways into pathway families (FIG. 3). The network groups were then manually inspected to disregard the known families, such as polysaccharide A, ¹⁵ catecholate-type siderophores, ³⁴ and aldehyde-type protease inhibitors. ²⁴

Among the HMP-derived, transcribed, and uncharacterized pathways, a family of eight pathways specific to the Clostridia bacterial class was chosen for investigation (FIG. IB). Analysis with MetaQuery ³⁵ confirms that these pathways are prevalent in the human gut, with at least one pathway appearing in 99.6% of the >2000 publically available human gut metagenomes. Three biosynthetic genes that are conserved amongst all eight pathways encode for the signature C, T, and A domains of the NRPS minimal module. The system comprises three unique characteristics: 1) the presence of only one NRPS module; 2) a lack of termination domain for product release; and 3) the architecture of each NRPS domain encoded as a separate, standalone protein. Bioinformatics analysis thereby suggests that this family encodes for an as-yet uncharacterized chemical modulatory mechanism that is highly utilized across diverse gut commensal Clostridia. FAA Pathway Phytogeny and Taxonomical Phytogeny are Different

To investigate the representation and distribution of the eight HMP-derived pathways with respect to other members in a general database, homologous pathways were

computationally searched on the NCBI nonredundant (nr) database and a total of 148 pathways were identified. When these pathways were organized based on protein sequence phylogeny, the Coprococcus eutactus, Lachnoclostridium clostridioforme, and Eubacterium rectale pathways from HMP data were found to be particularly represented across multiple strains, with eight to nine instances each at >98% sequence identity (FIG. 4). While all of the pathways come from the Clostridia class, analysis by re-organizing the pathways based on taxonomical phylogeny reveals that they are encoded by diverse Clostridia species. Notably, the strains that harbor close homologs of C. eutactus form one distinct taxonomical clade, while each of the L. clostridiofrome, and E. rectale pathways form multiple taxonomical clades. Therefore, while there is some correlation between the pathway- and taxonomy-based phylogenic distributions (e.g., C. eutactus pathways), they are different in many cases (e.g., L. clostridiofrome, and E. rectale pathways). As a result, taxonomical classification of the bacterial strains is not sufficient in predicting the FAA pathway types and any health-related functional outputs that they encode for.

Fatty Acid Amide Isolated from Gut Clostridia Pathways

Considering that some of the pathways were identified using metagenomic data and the host organisms were never isolated, E. coli heterologous expression was first attempted as a characterization strategy. The eight pathways were synthesized from genome or metagenome sequence with E. coli optimized codons. Each gene in the pathway was placed under a strong T7 promoter and ribosomal binding site for robust and inducible expression. The resulting redesigned pathways were introduced into E. coli BAP1 host that expresses Sfp PPTase to ensure proper activity of the heterologous NRPS machinery. ²⁸ Various fermentation and extraction conditions were attempted for the eight pathways, but clone-specific compound production was detected with sufficient titers for only the E. rectale pathway (FIG. 1C). The major compound was found to be palmitoeyl putrescine, as elucidated by 1- and 2-D NMR and then confirmed by comparison with a synthetic sample (FIG. 5).

The three NRPS domain-containing biosynthetic genes are the only genes conserved across all eight characterized FAA pathways. However, some pathways also contained genes encoding a sterol transfer protein (Pfam: PF02036) and an alpha/beta hydrolase protein (Pfam: PF12146) (Table 1). The exclusion of these genes had no effect on FAA production in a heterologous expression system (FIG. 6A). They are thus predicted to be accessory genes, with the sterol transfer protein facilitating the transport of exogenous fatty acids from the human gut lumen to the bacterial cytoplasm, ³⁶ while the hydrolase acts to either cleave human endogenous FAAs to generate free substrates for the pathway, or to degrade the FAA pathway product as a negative regulatory mechanism in much the same way as human FAA hydrolases do. ³⁷

Mechanism Elucidated for Clostridia FAA Biosynthesis

Palmitoeyl putrescine is a fatty acid amide (FAA) that is formed by the conjugation of palmitoleic acid and putrescine. Much like peptide bond formation, it is biochemically viable for NRPS domains to join a fatty acid and an amine to generate a FAA. However, considering that putrescine is highly produced in E. coli, but deficient in Clostridia, ³⁸ it is questionable as to whether palmitoleic acid and putrescine are actually the preferred substrates of this pathway. To investigate the contribution of the biosynthetic enzymes, single-gene knockout pathways without one of the biosynthetic genes were constructed and tested for production in E. coli. Removing either the A, T, or Sfp PPTase gene demolished compound production, thereby confirming the thiotemplating activity of A onto T in FAA production. However, removing the C gene had no effect on production, indicating that palmitoeyl putrescine is a shunt product that formed as a result of a different metabolic pool in E. coli host (FIG. 1C). Moreover, feeding the culture broth of the E. coli host harboring the wildtype pathway with either octanoic acid or tryptamine led to the production of octanoyl putrescine or palmitoeyl tryptamine, respectively (FIG. 6B). The pathway is therefore capable of incorporating a different fatty acid or amine that is exogenously available in the environment.

To further investigate the biosynthetic mechanism of FAA production, each of the three biosynthetic genes in the palmitoeyl putrescine-producing E. rectale pathway were cloned in a separate construct, expressed for protein purification, and reconstituted in vitro , such that the A and C activity can be measured independently with defined substrates. The A activity was first measured based on the amount of inorganic pyrophosphate (PPi) that was formed from ATP consumption. Up to 30-fold increase in PPi level relative to negative control was observed with fatty acid substrates, while no difference was observed with amino acid substrates (FIG. ID). Therefore, as opposed to the conventional NRPS adenylation enzyme that tethers an amino acid, the A protein in this pathway tethers a fatty acid onto the T protein. This is consistent with the fact that the conserved aspartic acid residue that interacts with the a-amino group of an amino acid substrate in canonical NRPS A domain ³⁹ is lacking in this A protein, thereby being similar to other acyl-coenzyme A synthetase enzymes. ⁴⁰

On the other hand, although the result from E. coli heterologous production suggests that the C protein does not catalyze a reaction involving putrescine, it may instead recognize a different amine. Since the condensation enzyme does not consume ATP, the amount of the product FAA itself was used as the metric for enzyme activity. Because different compounds ionize at varying intensities, ion abundance as measured by LCMS cannot be directly compared between FAAs of different amine moieties. However, based on the observation that the amine substrate, such as putrescine, can non-enzymatically release the tethered fatty acid to generate FAA, the FAA level was normalized in the presence of the C enzyme relative to the level in its absence. Among the small set of biological amines that were tested, the C enzyme showed preference towards incorporation of tryptamine, with about 8-fold higher activity than putrescine (FIG. IE).

Based on in vivo gene knockout and in vitro reconstitution studies, a biosynthetic scheme for FAA formation can be constructed, where T is an acyl carrier protein, A is a fatty acyl-acyl carrier protein synthetase, and C is a /V- fatty acyltransferase (FIG. IF). Upon Sfp- dependent activation of the T protein with PPT arm, the A enzyme selects an exogenous fatty acid substrate and tethers it onto T. The C enzyme then selects an exogenous amine and conjugates it onto the thiotemplated fatty acid (fatty acyl-T) to form the FAA product.

Bioinformatics has at times permitted the prediction of compound structure from primary sequence data alone, ⁴¹ but structure prediction was misleading for these pathways. While a peptide-based structure can be predicted from the sequence based on the presence of NRPS domains, it instead produced FAA with a biosynthetic machinery that shows limited homology to previously known FAA pathways. ⁴² Surprisingly, as described herein, this NRPS condensation domain protein could be used to catalyze the incorporation of an untethered substrate, like tryptamine, that does not have a carboxylate moiety.

Preferred FAA Products Characterized by In Vitro Substrate Combinatorics

The human gut lumen consists of various fatty acid and amine substrates from different sources that the pathway can incorporate for FAA production. To characterize the preferred product, an in vitro substrate panel assay was set up to investigate substrate selectivity of the biosynthetic enzymes for each of the eight identified HMP-derived pathways (FIG. 2A). A and C enzyme activity was measured for defined fatty acid and amines, respectively, to determine the substrates that each pathway prefers to incorporate for FAA production (FIG. 2B-2C). A panel of 25 fatty acids and 53 biogenic amines derived from Clostridia, human cells, and human diet that are known to be abundant in the human gut was selected (FIG. 2D). ^{43,44,45,46,47} For consistency, FAA product level, instead of PPi level, was used as the metric of A activity for the panel assay by adding tryptamine for all query fatty acid substrates to measure the level of C-independent FAA formation based on the UV absorbance of the indole chromophore (FIG. 2B). For C activity, the fatty acid that generated the highest A activity was added for all of the query amine substrates, with the metric being the normalized FAA product level as described previously (FIG. 2C). To alleviate the bottleneck of LCMS sample runtime, the number of runs was reduced by subpooling the query substrates together and then separating the FAA products by LC to measure enzymatic activity for each substrate. In this manner, the enzymatic activity of the A and C proteins for each pathway was elucidated for each fatty acid and amine substrates in the panel, respectively (FIG. 2E).

The A protein had a higher enzymatic activity than the C protein, with around 40-60 fold difference in the A protein between the highest and lowest substrates, compared to 5-15 fold difference in the C protein for the C. eutactus, M. formatexigens, R. bacterium, E. rectale pathways. This can be partially attributed to the fact that the enzymatic activity of the C protein is competing with nonenzymatic release of the tethered substrate, a phenomenon that is known in multimodular assembly reaction ⁴⁸ However, even taking that into consideration, the observed enzymatic activity of the C protein for the L. clostridioforme, B. producta, C.

celatum, C. sp pathways was low, with 2 fold difference with homocysteine substrate being the highest. Because of the lack of precedence for this type of condensation enzyme, the possibility that this 2 fold difference in activity is relevant in the context of this in vitro system and that homocysteine is the preferred amine substrate for these four pathways cannot be discounted. However, considering that previously characterized product release domains in NRPS systems compete against nonenzymatic release with at least 2 fold higher activity, ^49,50 it is more likely that the amine panel does not contain the appropriate substrate for these four pathways to observe C enzymatic activity higher than the 2 fold-difference threshold.

On the other hand, the C protein show higher substrate selectivity than the A protein, with 25-100% higher activity between the highest and second highest substrates in the C protein, compared to 8-40% in the A protein. Barring for differences in rank order, the A proteins in all pathways generally preferred to incorporate either middle-chain fatty acid ( e.g . lauric acid) or long-chain fatty acid with cis-9 double bond (e.g. oleic acid), both of which fit in a hydrophilic pocket that selects for a straight-chain alkyl chain of 10-12 carbons in length from the amide bond. Meanwhile, the C proteins show diversity in substrate selectivity, from the straight-chain amine (e.g. aminovaleric acid) of the C. eutactus pathway to the arylamine (e.g. tryptamine) of the E.rectale pathway.

The fatty acid and amine substrates yielding the highest enzymatic activity were determined for each pathway, leading to the production of oleoyl aminovaleric acid for the C. eutactus pathway, oleoyl dopamine for the M. formatexigens pathway, oleoyl tyramine for the R. bacterium pathway, and lauroyl tryptamine for the E. rectale pathway as the major FAA (FIG. 2E and FIG. 7). The production of a-linolenoyl homocysteine yielded the highest enzymatic activity for the L clostridioforme, B. producta, C. celatum, and C. sp. pathways, but the low C enzymatic activity diminishes its likelihood as being the major FAA. The products that are generated by the incorporation of second highest fatty acid or amine are also listed as examples of minor FAAs (FIG. 2E). These may become relevant depending on the abundance of the fatty acid and amine substrates that the native strains are subjected to in the human gut environment.

C and T Protein Interaction is Pathway Specific

Organisms use acyl carrier proteins homologous to the FAA pathway's T protein to make endogenous fatty acids. ⁵¹ Therefore, in order for the FAA pathway to utilize exogenous fatty acids, the biosynthetic proteins may have a mechanism that prevents crosstalk with other homologous proteins. To investigate such pathway-specific interactions, the biosynthetic proteins from the two FAA pathways that make oleoyl aminovaleric acid and lauroyl tryptamine were swapped and their effect on enzyme activity was measured. While the C protein lost activity, the A protein retained activity after swapping the T protein. Swapping the A domain, while retaining the T and C protein pair from the same pathway, permitted the enzymatic production of hybrid FAA that incorporated the substrates from two different pathways (FIG. 8). This suggests that the T and C proteins interact in a pathway-specific manner, possibly by containing donor and acceptor communication-mediating domains found in known NRPS systems. ⁵² This has implications in pathway engineering, where the communication between the T and C protein from the same pathway needs to be retained when considering the combinatorial biosynthesis of these pathways for the production of FAA chemical library. Methods

Strains and plasmids

Escherichia coli DH10B (New England BioLabs) was used for routine cloning. E. coli BAP1 E. coli BAP1 containing T7 DNA polymerase and Sfp phosphopantetheinyl transferase was used for heterologous expression of re-designed pathways. ²⁸ E. coli BL21(DE3) (Sigma- Aldrich) was used for protein expression and pathway expression in the absence of Sfp.

Plasmid was introduced by electroporation, with electrocompetent cells prepared for the non- commercially available BAP1 strains using published protocol. ⁶² Vector pET28a (Novagen) was used as backbone for all pathway and protein expression constructs. Kanamycin was used as for selection at 30 pg/ml.

Pathways including the biosynthetic and accessory genes were re-designed such that each gene was codon optimized for E. coli without Type IIS restriction sites (GeneArt CodonOptimizer), placed under T7 promoter (with lacO operator) and RBS parts from pET28a (Novagen), and arranged in the same direction in the order that they appear natively (FIG.

10, Tables 1 and 2). The re-designed pathways were synthesized by either Invitrogen or Gen9 with flanking Bsal sites. The synthesized pathways, either as purified fragments or cloned in default vector, were cloned into pET28a using Golden Gate assembly and transformed into E. coli BAP1. In Table 1, genes with names that end with“C" at the end encode fatty acyl- transferases ( e.g ., ceuC, mfoC, etc.). In Table 1, genes with names that end with“A" at the end encode fatty acyl-ACP synthetases (e.g., ceuA, mfoA, etc.). In Table 1, genes with names that end with“T” at the end encode acyl carrier proteins (ACPs) (e.g., ereT, ceuT, mfoT).

Table 1. Non-limiting examples of biosynthetic enzymes from human gut microbiome -derived bacterium.

- Ill -

Table 2. Non-limiting examples of vector components.

Reagents and chemicals

Commercially available reagents and chemicals were used. Fatty acid and amine substrates for the panel assay are listed separately (Table 3). Fatty acid substrates are prepared as 50 mM stock solutions in ethanol, neutralized with sodium hydroxide solution, stored at - 80°C prior to use. Amine substrates are prepared as 50 mM stock solutions in water (with pH adjusted to increase solubility), stored at -80°C prior to use. IPTG was prepared as 1 M stock solution and stored at -20°C. Co A solution was freshly prepared from powder. PCR primers are ordered from Integrated DNA Technologies .

Table 3. Non-limiting examples of amines and fatty acids.

Computational detection of Clostridia NRPS pathways Genome datasets from the human gut was searched on the genome browser of the JGI- IMG database querying "Gastrointestinal tract" as the Sample Body Site and "Human" as a keyword. The metadata from search result was used to locate and download the sequencing data from either JGI GOLD or NCBI GenBank database. The resulting datasets were run on antiSMASH 3.0 using default parameters with ClusterFinder-based border prediction, but excluding putative pathways detected by ClusterFinder (low-confidence). ²² The antiSMASH- detected pathways were blastn searched on the metagenomic reads of 148 fecal samples from HMP that passed QC assessment, retaining pathways that had at least two hits with e-value < 1X10 ⁵ from the metagenomic reads spanning different stretches of the pathway. These were blastn searched, using the same cutoff filter as the previous step, on the mRNA reads from fecal samples of eight healthy subjects. The remaining pathways were run on BiG-SCAPE ³³ to generate a similarity network matrix file with a distance cutoff of 0.75 and visualized using Cytoscape. For analyzing prevalence, all three biosynthetic proteins from the eight NRPS pathways were combined as a single query and submitted on MetaQuery with default parameters, except with a minimum percent identity of 80.

Detection of homologous pathways in non-redundant sequence database

E. rectale condensation protein and adenylation protein was each used as a query for blastp search with default parameters. Upon removing hits that were different in protein size (>800 or <200 amino acids) from the top 1000 hits, the hit table from each was cross-examined using Python script for their co-occurrence in the same pathway based on proximity in NCBI accession number.

Phylogenetic tree construction

For pathway phylogeny, MUSCLE alignment was performed on the adenylation protein sequences using MacVector version 16. The tree was constructed from the alignment using Paraeggerthella hongkongensis adenylation protein (accession: WP_123191107) as an outgroup for rooting (Method: Neighbor Joining Method, Distance: Uncorrected, Best Tree Mode). For taxonomical phylogeny, NCBI taxon ID extracted from the GenBank annotation of each pathway was collected, which was submitted to phyloT for tree construction

(https://phylot.biobyte.de/). The trees are visualized using iTOL. ⁶³ E. coli in vivo heterologous expression

Overnight culture of pathway-harboring E coli in LB broth at 37°C overnight (230 rpm) was diluted to fresh media the next morning. Multiple fermentation and extraction conditions were attempted. For IPTG induction, the culture was induced at either early, middle, or late log phase (OD600 = 0.2, 0.5, 0.8) and at final concentration of 0.2 or 0.5 mM IPTG. After induction, culture media was grown in LB or M9 minimal media. The induced culture was grown at 25 or 30 °C and extracted after 8, 16, or 40 hr. The culture was extracted with equal volume of methanol, mixed on the vortex, spun down to remove particulates, and subjected to analytical LCMS run. The condition with highest clone- specific compound production relative to background was defined as the best condition for subsequent in vivo production studies (OD600 = 0.8; 0.2 mM IPTG; LB after induction; 25 °C; 16 hr).

Analytical LCMS run

Analytical LCMS was conducted using an Agilent 6130 quadrupole MS on a C 18 reverse phase column (Phenomenex Luna 5 mhi Cl 8(2), 100 x 4.6 mm) with 1.0 mL/min flow rate and a gradient system of 90%:10% to 0%:100% water: acetonitrile with 0.1% formic acid for 15 min, followed by 2 min isocratic run at 100% acetonitrile (wash) and then 3 min at 90%: 10% water: acetonitrile (re-equilibration).

Single-gene knockout pathway characterization

PCR fragment with flanking Bsal site was made using primer pair ereCko-F/ereCko-R, ereAko-F/ ereAko-R, or ereTko-F/ereTko-R, which was then digested and ligated to generate a E. rectale pathway construct with the condensation, adenylation, or thiolation gene knockout, respectively. For pathway consisting of just the three biosynthetic genes, the primer pair ereC- F/ereA-R was used to amplify the appropriate fragment that was then cloned into pET28a.

Each construct was transformed, cultured with the best condition (OD600 = 0.8; 0.2 mM IPTG; LB after induction; 25 °C; 16 hr), and analyzed like the wildtype pathway.

E. coli in vivo substrate feeding assay

Native E. rectale pathway-harboring E. coli was cultured with the best condition and as previously, except either octanoic acid (neutralized with sodium hydroxide) or tryptamine was added into the culture at a final concentration of 5 mM during IPTG induction. The extracts are analyzed as previously. E. coli in vivo compound characterization

Large-scale fermentation of pathway-harboring E. coli (16 L) was conducted using the best condition. Culture was extracted with equal volume of ethyl acetate, and the organic layer was dried in vacuo. The extract was run on preparatory C18 reverse phase column

(Phenomenex Luna 5 mhi 08(2), 250 x 21.2 mm) with a gradient system of 90%:10% to 0%:100% water: acetonitrile with 0.1% acetic acid in 20 min at 10 mL/min. Major compound eluted with the 20%:80% water: acetonitrile fraction. Compound was then purified with a semi- prep C18 column (Phenomenex Luna 5 mhi Cl 8(2), 250 x 10 mm) with a gradient system of 70%:30% to 0%:100% water: methanol with 0.1% acetic acid in 30 min at 5 mL/min. The major compound eluted in the 45%:65% watenmethanol fraction as a white powder (0.1 mg). HRMS acquired on Agilent 6530 Q-TOF and 1-D and 2-D NMR spectra collected on Bruker Avance II 600 were used to determine chemical structure (FIG. 5). The compound identify was further confirmed by having an authentic standard chemically synthesized by KareBay

Biochem and was found to be spectroscopically identical.

Protein expression plasmid construction

Individual biosynthetic gene was PCR amplified from the synthesized construct using the primer set containing the gene name (nomenclature of "pathway source strain name abbreviation, followed by "NRPS domain type" (C, T, A).The amplified gene was cloned into pET28a and transformed into E. coli BL21(DE3). The exception was C. eutactus condensation protein, whose construct was sent to ABclonal Technology for custom protein expression and purification.

Protein purification

Except the C. eutactus condensation protein construct, the construct-harboring E. coli was grown in LB at 25°C (230 rpm) to OD = 0.5, induced with IPTG at 0.5 mM final concentration, and grown at 16°C (230 rpm) for 18 hr. The cells were pelleted at 4000 g for 20 min, resuspended in 10 mL Lysis Buffer (300 mM NaCl, 10 mM Imidazole, 50 mM Tris, pH 8.0) with EDTA-free Protease Inhibitors, and lysed using Q500 Qsonica sonicator with 1/8" stepped microtip probe (45% amplitude, 5 min continuous). Upon centrifuging the lysed cells at 14,000 g for 20 min, the supernatant was added to pre-equilibrated Ni-NTA resin and rotated at 4°C for 16 hr. The beads were spun down at 1000 g for 1 min, resuspended in 10 mL Wash Buffer (300 mM NaCl, 20 mM Imidazole, 50 mM Tris, pH 8.0). Upon transferring to a new tube, the beads were washed three times with 10 mL Wash Buffer. In a new tube, the beads were eluted with 2 mL of Elution Buffer (300 mM NaCl, 20 mM Imidazole, 250 mM Tris, pH 8.0) and dialyzed using Slide-A-Lyzer (3.5K MWCO) against Dialysis Buffer (50 mM NaCl, 1 mM TCEP, 10% (v/v) glycerol). Protein purity was confirmed by 16% Tricine protein gel with SimplyBlue SafeS tain.

In vitro reconstitution

Reaction mixture was set up in 100 mE volume with 100 mM Tris, 10 mM MgCh, 1 mM TCEP, 0.1 mM Sfp Synthase (New England Biolabs), 0.1 mM CoA, 5 mM ATP, 1 mM of each substrate (fatty acid, amine), and 1 mM of each biosynthetic protein (condensation, thiolation, adenylation) and proceeded at 23 °C.

Pyrophosphate measurement for adenylation activity

Enzchek Pyrophospate Assay Kit (Molecular Probes) was used for pyrophosphate detection as similarly done if previous studies. ⁶⁴ After a 2 hr reaction with adenylation and thiolation protein (no condensation), 10 mE of reaction sample was mixed with the kit solution containing MESG, phosphorylase, and pyrophosphatase in a 100 mE mixture following manufacturer protocol, incubated for 60 min at room temperature, spun down at 4000 g for 1 min to remove particulates, and transferred into Half Area 96 well UV microplate (Corning) for measurement of absorbance at 360 nm using BioTek Epoch Microplate Spectrophotometer, normalized to no adenylation enzyme control.

MS-based measurement for adenylation and condensation activity

For adenylation activity, adenylation and thiolation proteins (no condensation) were added with the query fatty acid substrate(s) and tryptamine as a UV-active chromophore. For condensation activity, a pair of reaction with all three biosynthetic proteins and all but without the condensation protein was set up with the query amine substrate(s) and the fatty acid that yielded the highest adenylation activity. The reaction was mixed with equal volume of methanol, spun down at 21,000 g for 1 min, and analyzed with the same analytical LCMS run as described previously. The product appeared as the major peak on the extracted-ion chromatogram (EIC) of the corresponding m/z in either positive or negative mode, without appearing on both the no substrate control and no enzyme control. The peak, either from absorbance at 280 nm for adenylation activity or MS EIC for condensation activity, was integrated using the software Mestrelab MNova 10.0. For condensation activity, the reaction sample with and without condensation protein was run one after the other, and the ratio of the two peak areas was calculated.

Substrate panel assay

For the panel assays (Table 3), the fatty acid substrates were subpooled into five:

Subpool Fa = FI, 6, 10, 13, 17, 20, 21; Fb = F2, 9, 11, 16, 22, 24; Fc = F3, 5, 7, 12, 14, 18; FD = Fd, 8, 15, 19, 23, 25. The amine substrates were subpooled into six: Subpool Aa = A9, 27,

28, 29, 30, 31, 32, 33, 34; Ab = Al, 2, 3, 4, 12, 16, 17, 20, 23; Ac = 5, 11, 14, 21, 35, 37, 39, 42, 43; Ad = A6, 7, 8, 10, 13, 15, 18, 19, 38; Ae = 22, 24, 25, 26, 40, 41, 47, 51, 53; Af = 36, 44, 45, 46, 48, 49, 50, 52. First, reaction sample with C. eutactus proteins was analyzed over the course of ten different time points (FIG. 11). The reaction time that provided the largest difference between the highest activity and the rest was defined as the best time point (60 min for fatty acid and 120 min for amine panel) for collecting data in the determination of preferred product. Upon data collection with the same subpools, the experiments were conducted with a separate subpool consisting of the highest activity substrate from each subpool to confirm that the rank order of the substrate remained the same. The identities of the preferred products were further confirmed by having authentic standards chemically synthesized by KareBay Biochem (with the exception of oleoyl dopamine, which was purchased from Cayman Chemical) and conducting side-by-side analytical FCMS runs (FIG. 6).

References Section 1

1 Arafat, E. S., Trimble, J. W., Andersen, R. N., Dass, C. & Desiderio, D. M. Identification of fatty acid amides in human plasma. Life sciences 45, 1679-1687 (1989).

2 Ezzili, C., Otmbova, K. & Boger, D. L. Fatty acid amide signaling molecules. Bioorganic & medicinal chemistry letters 20, 5959-5968, doi:10.1016/j.bmcl.2010.08.048 (2010).

3 Devane, W. A. et al. Isolation and structure of a brain constituent that binds to the cannabinoid receptor.

Science 258, 1946-1949 (1992).

4 Cravatt, B. F. et al. Molecular characterization of an enzyme that degrades neuromodulatory fatty-acid amides. Nature 384, 83-87, doi:10.1038/384083a0 (1996).

5 Eisenstein, T. K., Meissler, J. J., Wilson, Q., Gaughan, J. P. & Adler, M. W. Anandamide and Delta9- tetrahydrocannabinol directly inhibit cells of the immune system via CB2 receptors. Journal of neuroimmunology 189, 17-22, doi:10.1016/j.jneuroim.2007.06.001 (2007). 6 Osei-Hyiaman, D. et al. Endocannabinoid activation at hepatic CB1 receptors stimulates fatty acid synthesis and contributes to diet-induced obesity. The Journal of clinical investigation 115, 1298-1305, doi: 10.1172/JCI23057 (2005).

7 Chu, C. J. et al. N-oleoyldopamine, a novel endogenous capsaicin-like lipid that produces hyperalgesia.

The Journal of biological chemistry 278, 13633-13639, doi:10.1074/jbc.M211231200 (2003).

8 Ross, H. R., Gilmore, A. J. & Connor, M. Inhibition of human recombinant T-type calcium channels by the endocannabinoid N-arachidonoyl dopamine. British journal of pharmacology 156, 740-750, doi: 10.1111/j .1476-5381 2008.00072.x (2009).

9 Chu, Z. L. et al. N-oleoyldopamine enhances glucose homeostasis through the activation of GPR119.

Molecular endocrinology 24, 161-170, doi: 10.1210/me.2009-0239 (2010).

10 Sergeeva, O. A. et al. N-oleoyldopamine modulates activity of midbrain dopaminergic neurons through multiple mechanisms. Neuropharmacology 119, 111-122, doi:10.1016/j.neuropharm.2017.04.011 (2017).

11 Przegalinski, E., Filip, M., Zajac, D. & Pokorski, M. N-oleoyl-dopamine increases locomotor activity in the rat. International journal of immunopathology and pharmacology 19, 897-904,

doi: 10.1177/039463200601900419 (2006).

12 Wang, Y. et al. Docosahexaenoyl serotonin emerges as most potent inhibitor of IL-17 and CCL-20 released by blood mononuclear cells from a series of N-acyl serotonins identified in human intestinal tissue. Biochimica et biophysica acta. Molecular and cell biology of lipids 1862, 823-831,

doi:10.1016/j.bbalip.2017.05.008 (2017).

13 Huang, S. M. et al. Identification of a new class of molecules, the arachidonyl amino acids, and

characterization of one member that inhibits pain. The Journal of biological chemistry 276, 42639-42644, doi: 10.1074/jbc.M107351200 (2001).

14 Milman, G. et al. N-arachidonoyl L-serine, an endocannabinoid-like brain constituent with vasodilatory properties. Proceedings of the National Academy of Sciences of the United States of America 103, 2428- 2433, doi: 10.1073/pnas.0510676103 (2006).

15 Mazmanian, S. K., Round, J. L. & Kasper, D. L. A microbial symbiosis factor prevents intestinal

inflammatory disease. Nature 453, 620-625, doi: 10.1038/nature07008 (2008).

16 Schulz, S., Green, C. K., Yuen, P. S. & Garbers, D. L. Guanylyl cyclase is a heat-stable enterotoxin receptor. Cell 63, 941-948 (1990).

17 Cuevas-Ramos, G. et al. Escherichia coli induces DNA damage in vivo and triggers genomic instability in mammalian cells. Proceedings of the National Academy of Sciences of the United States of America 107, 11537-11542, doi:10.1073/pnas. l001261107 (2010).

18 Nougayrede, J. P. et al. Escherichia coli induces DNA double-strand breaks in eukaryotic cells. Science 313, 848-851, doi: 10.1126/science.1127059 (2006).

19 Schneditz, G. et al. Enterotoxicity of a nonribosomal peptide causes antibiotic-associated colitis.

Proceedings of the National Academy of Sciences of the United States of America 111, 13181-13186, doi: 10.1073/pnas.1403274111 (2014).

20 Human Microbiome Project, C. Structure, function and diversity of the healthy human microbiome.

Nature 486, 207-214, doi: 10.1038/naturel l234 (2012). 21 Cohen, L. J. et al. Commensal bacteria make GPCR ligands that mimic human signalling molecules. Nature 549, 48-53, doi: 10.1038/nature23874 (2017).

22 Weber, T. et al. antiSMASH 3.0-a comprehensive resource for the genome mining of biosynthetic gene clusters. Nucleic acids research 43, W237-243, doi:10.1093/nar/gkv437 (2015).

23 Donia, M. S. et al. A systematic analysis of biosynthetic gene clusters in the human microbiome reveals a common family of antibiotics. Cell 158, 1402-1414, doi:10.1016/j.cell.2014.08.032 (2014).

24 Guo, C. J. et al. Discovery of Reactive Microbiota-Derived Metabolites that Inhibit Host Proteases. Cell 168, 517-526 e518, doi: 10.1016/j.cell.2016.12.021 (2017).

25 Macdonald, I. A., Bokkenheuser, V. D., Winter, J., McLernon, A. M. & Mosbach, E. H. Degradation of steroids in the human gut. Journal of lipid research 24, 675-700 (1983).

26 Brotherton, C. A. & Balskus, E. P. A prodrug resistance mechanism is involved in colibactin

biosynthesis and cytotoxicity. Journal of the American Chemical Society 135, 3359-3362,

doi: 10.1021/ja312154m (2013).

27 Fischbach, M. A. & Walsh, C. T. Assembly-line enzymology for polyketide and nonribosomal Peptide antibiotics: logic, machinery, and mechanisms. Chemical reviews 106, 3468-3496, doi:10.1021/cr0503097 (2006).

28 Pfeifer, B. A., Admiraal, S. J., Gramajo, H., Cane, D. E. & Khosla, C. Biosynthesis of complex

polyketides in a metabolically engineered strain of E. coli. Science 291, 1790-1792,

doi: 10.1126/science.1058092 (2001 ).

29 Roche, E. D. & Walsh, C. T. Dissection of the EntF condensation domain boundary and active site residues in nonribosomal peptide synthesis. Biochemistry 42, 1334-1344, doi: 10.1021/bi026867m (2003).

30 Mori, S. et al. Activation and Loading of the Starter Unit during Thiocoraline Biosynthesis. Biochemistry 56, 4457-4467, doi:10.1021/acs.biochem.7b00661 (2017).

31 Stachelhaus, T., Mootz, H. D., Bergendahl, V. & Marahiel, M. A. Peptide bond formation in

nonribosomal peptide biosynthesis. Catalytic role of the condensation domain. The Journal of biological chemistry 213, 22773-22781 (1998).

32 Franzosa, E. A. et al. Relating the metatranscriptome and metagenome of the human gut. Proceedings of the National Academy of Sciences of the United States of America 111, E2329-2338,

doi: 10.1073/pnas.1319284111 (2014).

33 Navarro-Munoz, J. et al. A computational framework for systematic exploration of biosynthetic diversity from large-scale genomic data. bioRxiv (2018).

34 Raymond, K. N., Dertz, E. A. & Kim, S. S. Enterobactin: an archetype for microbial iron transport.

Proceedings of the National Academy of Sciences of the United States of America 100, 3584-3588, doi: 10.1073/pnas.0630018100 (2003).

35 Nayfach, S., Fischbach, M. A. & Pollard, K. S. MetaQuery: a web server for rapid annotation and

quantitative analysis of specific genes in the human gut microbiome. Bioinformatics 31, 3368-3370, doi:10.1093/bioinformatics/btv382 (2015).

36 Frolov, A., Cho, T. H., Billheimer, J. T. & Schroeder, F. Sterol carrier protein-2, a new fatty acyl

coenzyme A-binding protein. The Journal of biological chemistry 271, 31878-31884 (1996). 37 McKinney, M. K. & Cravatt, B. F. Structure and function of fatty acid amide hydrolase. Annual review of biochemistry 74, 411-432, doi: 10.1146/annurev.biochem.74.082803.133450 (2005).

38 Pugin, B. et al. A wide diversity of bacteria from the human gut produces and degrades biogenic amines.

Microbial ecology in health and disease 28, 1353881, doi:10.1080/16512235.2017.1353881 (2017).

39 Stachelhaus, T., Mootz, H. D. & Marahiel, M. A. The specificity-conferring code of adenylation domains in nonribosomal peptide synthetases. Chemistry & biology 6, 493-505, doi:10.1016/S1074-5521(99)80082-9 (1999).

40 Gulick, A. M. Conformational dynamics in the Acyl-CoA synthetases, adenylation domains of non ribosomal peptide synthetases, and firefly luciferase. ACS chemical biology 4, 811-827,

doi:10.1021/cb900156h (2009).

41 Chu, J. et al. Discovery of MRS A active antibiotics using primary sequence from the human

microbiome. Nature chemical biology 12, 1004-1006, doi: 10.1038/nchembio.2207 (2016).

42 Farrell, E. K. & Merkler, D. J. Biosynthesis, degradation and pharmacological importance of the fatty acid amides. Drug discovery today 13, 558-568, doi:10.1016/j.drudis.2008.02.006 (2008).

43 Hill, M. J. Microbial Metabolism In The Digestive Tract. (CRC Press, 2018).

44 Abdelmagid, S. A. et al. Comprehensive profiling of plasma fatty acid concentrations in young healthy Canadian adults. PloS one 10, eOl 16195, doi:10.1371/journal.pone.0116195 (2015).

45 Chan, M., Himes, R. H. & Akagi, J. M. Fatty acid composition of thermophilic, mesophilic, and

psychrophilic Clostridia. Journal of bacteriology 106, 876-881 (1971).

46 Vernocchi, P., Del Chierico, F. & Putignani, L. Gut Microbiota Profiling: Metabolomics Based

Approach to Unravel Compounds Affecting Human Health. Frontiers in microbiology 7, 1144, doi:10.3389/fmicb.2016.01144 (2016).

47 Lyte, M. & Freestone, P. P. E. Microbial Endocrinology: Interkingdom Signaling in Infectious Disease and Health. (Springer New York, 2010).

48 Schwarzer, D., Mootz, H. D. & Marahiel, M. A. Exploring the impact of different thioesterase domains for the design of hybrid peptide synthetases. Chemistry & biology 8, 997-1010 (2001).

49 Yeh, E., Lin, H., Clugston, S. L., Kohli, R. M. & Walsh, C. T. Enhanced macrocyclizing activity of the thioesterase from tyrocidine synthetase in presence of nonionic detergent. Chemistry & biology 11, 1573- 1582, doi: 10.1016/j.chembiol.2004.09.003 (2004).

50 Keating, T. A., Miller, D. A. & Walsh, C. T. Expression, purification, and characterization of HMWP2, a 229 kDa, six domain protein subunit of Yersiniabactin synthetase. Biochemistry 39, 4729-4739 (2000).

51 Campbell, J. W. & Cronan, J. E., Jr. Bacterial fatty acid biosynthesis: targets for antibacterial drug

discovery. Annual review of microbiology 55, 305-332, doi: 10.1146/annurev. micro.55.1.305 (2001).

52 Hahn, M. & Stachelhaus, T. Selective interaction between nonribosomal peptide synthetases is facilitated by short communication-mediating domains. Proceedings of the National Academy of Sciences of the United States of America 101, 15585-15590, doi: 10.1073/pnas.0404932101 (2004).

53 Ziemert, N., Alanjary, M. & Weber, T. The evolution of genome mining in microbes - a review. Natural product reports 33, 988-1005, doi:10.1039/c6np00025h (2016).

54 Zhang, H., Wang, Y. & Pfeifer, B. A. Bacterial hosts for natural product production. Molecular

pharmaceutics 5, 212-225, doi:10.1021/mp7001329 (2008). 55 Dixit, S. & Das, M. Fatty acid composition including trans-fatty acids in edible oils and fats: probable intake in Indian population. Journal of food science 77, T188-199, doi: 10.1111/j.1750-3841.2012.02875.x (2012).

56 Kokatnur, M. G., Oalmann, M. C., Johnson, W. D., Malcom, G. T. & Strong, J. P. Fatty acid

composition of human adipose tissue from two anatomical sites in a biracial community. The American journal of clinical nutrition 32, 2198-2205, doi: 10.1093/ajcn/32.11.2198 (1979).

57 Donia, M. S. & Fischbach, M. A. F1UMAN MICROBIOTA. Small molecules from the human

microbiota. Science 349, 1254766, doi: 10.1126/science.1254766 (2015).

58 McCutcheon, J. P. & von Dohlen, C. D. An interdependent metabolic patchwork in the nested symbiosis of mealybugs. Current biology : CB 21, 1366-1372, doi: 10.1016/j.cub.2011.06.051 (2011).

59 Ridaura, V. & Belkaid, Y. Gut microbiota: the link to your second brain. Cell 161, 193-194,

doi: 10.1016/j. cell.2015.03.033 (2015).

60 Mittal, R. et al. Neurotransmitters: The Critical Modulators Regulating Gut-Brain Axis. Journal of

cellular physiology 232, 2359-2372, doi:10.1002/jcp.25518 (2017).

61 Pacher, P., Batkai, S. & Kunos, G. The endocannabinoid system as an emerging target of

pharmacotherapy. Pharmacological reviews 58, 389-462, doi:10.1124/pr.58.3.2 (2006).

62 Sharma, R. C. & Schimke, R. T. Preparation of electrocompetent E. coli using salt-free growth medium.

BioTechniques 20, 42-44, doi:10.2144/96201bm08 (1996).

63 Letunic, I. & Bork, P. Interactive tree of life (iTOL) v3: an online tool for the display and annotation of phylogenetic and other trees. Nucleic acids research 44, W242-245, doi:10.1093/nar/gkw290 (2016).

64 Duckworth, B. P., Wilson, D. J. & Aldrich, C. C. Measurement of Nonribosomal Peptide Synthetase Adenylation Domain Activity Using a Continuous Flydroxylamine Release Assay. Methods in molecular biology 1401, 53-61, doi:10.1007/978-l-4939-3375-4_3 (2016).

EXAMPLE 2

Clostridia NRPS-like Pathways Identified from Human Gut Sequencing Data

This work began with a broad survey of biosynthetic pathways present in human gut microbiome genomic datasets (Fig. la). Metadata from the Joint Genome Institute (JGI) was used to identify 1042 bacterial genomes associated with the human gastrointestinal tract. From these genomes, 3531 putative biosynthetic pathways were identified using antiSMASH, which recognizes protein domains from known natural product pathways (Fig. 15a) ^15,45 . This set was further reduced to those that are present and actively transcribed in healthy human subjects ⁴⁶ . This reduced the number of pathways to 2013 that were found in at least one metagenomic sample. This was further narrowed by examining their presence in RNA sequencing datasets from the stool samples of eight healthy subjects from the HMP (Fig. 15c) ⁴⁷ .

The 336 transcribed pathways were organized using the software BiG-SCAPE which uses network- similarity algorithms to group the pathways into families ⁴⁸ . The groups were then manually inspected to disregard the known families ^8,40,49 . Eight pathways of particular interest, specific to Clostridia that are prevalent in the human gut, were identified (Fig. 15a) ⁵⁰ . A MetaQuery analysis showed that at least one pathway appears in 1730 of the 2271 publicly available human gut metagenomes ⁵¹ . When the search was performed with the NCBI nonredundant (nr) database, not requiring presence in human gut samples, 148 unique homologous pathways spanning diverse Clostridia were identified.

The Clostridia-derived gene clusters encode three conserved biosynthetic genes (Fig. 15d). Intriguingly, each shows sequence similarity to the C, T, and A domains of a NRPS minimal module (Figures 16a and 17-19). Neither of the proteins corresponding to the C and A domains show sequence homology to biosynthetic enzymes known to make FAAs. The closest homologue to the A protein is a fatty acyl-CoA ligase from B. subtilis (UniProt#: 007610). Meanwhile, the closest homologue to the C protein is a condensation domain of surfactin NRPS from B. subtilis (UniProt#: P27206). This condensation subtype forms its own clade in the phylogenetic tree of known NRPS condensation domains, and is known to conjugate long chain fatty acyl-CoA with thiotemplated a-amino acid ⁵² . The tandem action of C and A domains are both necessary for and exclusive to NRPS production and thus their co localization is used as a search strategy to identify NRPS pathways. In rare occasions, the C, T, A functions appear as independent proteins. However, to the inventors’ knowledge, this is the first example where all three functions are co-located as independent proteins without a termination domain. This suggests that although the most closely related known homologs of these pathways are NRPSs, they are likely not canonical NRPSs.

These three biosynthetic genes are the only genes conserved across all Clostridia- derived pathways. However, some pathways also contained genes encoding a sterol transfer protein, an alpha/beta-hydrolase protein, or a PPTase (Fig. 15a and Table 1). In addition, some pathways consisted of a TetR-like transcriptional regulator (Fig. 20). These generally constitute the genes that make up the pathways with boundaries predicted by antiSMASH. Other genes close to the predicted gene cluster boundaries were manually inspected and found to be putative housekeeping genes ( e.g cell wall glycosyltransferase) that are not conserved across any of the other pathways.

Mechanism Elucidated for Clostridia FAA Biosynthesis

Some of the strains in which the pathways are found have not been isolated or are not available from a strain bank. Therefore, the pathways were reconstituted in E. coli containing a genome- encoded IPTG-inducible T7 RNA polymerase (RNAP). Initially focusing on the eight human gut Clostridia-derived pathways, the C, T, and A genes were codon optimized and placed under the control of strong T7 promoters and ribosome binding sites (RBSs) (Fig. 21, Tables 1 and 2). Sfp, a promiscuous PPTase from Bacillus subtilis, was placed under inducible control in the genome to ensure PPTation of T domain for proper NRPS activity. When the genes were expressed in E. coli, only a compound for the E. rectale pathway was able to be obtained, despite numerous attempts to optimize the fermentation and extraction conditions. The major compound was determined to be palmitoeyl putrescine by 1- and 2-D NMR, formed by palmitoleic acid and putrescene (Figs. 5 and 16b). Considering that putrescine is highly produced in E. coli compared to Clostridia, it was questioned whether these are the actual substrates of this pathway.

To investigate the mechanism, variants of the E. rectale pathway were constructed where each gene was individually removed and the impact on pamitoeyl putrescine production was tested (Figs. 6a and 16b). Removing either the A, T, or Sfp PPTase gene eliminated production, thereby confirming the thiotemplated reaction of A and T in FAA production. In contrast, deleting the C gene had no effect, thus indicating it is not participating in palmitoeyl putrescine production. Moreover, feeding the culture broth with either octanoic acid or tryptamine led to the production of octanoyl putrescine or palmitoleoyl tryptamine, respectively (Fig. 6b). The pathway is therefore capable of incorporating an exogenous fatty acid or amine.

Each enzyme from the E. rectale pathway was then purified and reconstituted in vitro (Fig. 21b). The reaction is based on previously- studied NRPS in vitro systems, which include Sfp and Coenzyme A (CoA), TCEP as reducing agent, and ATP. First, the loading of the substrate onto T was measured by monitoring the ATP consumption using a phosphate assay based on a dye that quantifies PPi. PPi was measured in the presence and absence of the protein A, and this ratio was used to identify substrates (Fig. 16c). These data demonstrate that fatty acids are strongly preferred, with no observed incorporation of putrescine or any amino acid substrates that may further be processed to form a putrescine moiety. This was further confirmed by the detection of fatty acid thiotemplated T protein fragment using LC-MS/MS (Fig. 22). This is consistent with the protein sequence alignment showing that the Clostridia- derived A domain lacks the corresponding aspartic acid residue of a canonical NRPS A domain that interacts with the a-amino group (Fig. 23). Therefore, the A protein in this pathway is not a NRPS adenylation protein, but a fatty acyl-CoA ligase that tethers a fatty acid (instead of a- amino acid) onto the T protein.

Deleting the C gene had no effect on the production of palmitoeyl putrescine in E. coli. This could be due to palmitoeyl putrescine being a shunt product that forms because E. coli lacks the appropriate substrate. To determine the substrate, a LC-MS assay was developed to measure the relative FAA product concentration formed by the in vitro reaction because condensation domains that utilize thiotemplated substrates catalyze ligation without ATP consumption. Different FAAs ionize at varying intensities, so ion abundance cannot be used for side-by-side comparisons of different amines. Instead, the fact that there is some background incorporation of amines (e.g., putrescine) was used, and the ratio of the LC-MS peaks in the presence and absence of the C protein was reported. Different amines are preferred, with tryptamine having 8-fold higher enzymatic incorporation (Fig. 16d). Catalytic incorporation was confirmed by mutating the histidine residue essential for the NRPS condensation reaction, which eliminated activity (Figs. 16d and 24).

A mechanism can be inferred from the in vivo heterologous expression data and in vitro reconstruction. The T protein serves as an acyl carrier protein, the A protein is a fatty acyl-acyl carrier protein synthetase (fatty acyl-CoA ligase), and the C protein is a /V- fatty acyltransferase (Fig. 16a). Upon Sfp-dependent activation of T with the PPT arm, A selects an exogenous fatty acid substrate and tethers it onto T. C then selects an exogenous amine and conjugates it onto the thiotemplated fatty acid (fatty acyl-T) to form the FAA product. The chemical mechanism differs from canonical NRPS systems in that the amine substrate is not selected by the A enzyme, and the C enzyme joins activated fatty acid with a free amine, thereby resembling N- acyltransferase systems characterized to generate FAAs in humans and bacteria. However, where canonical - acyl transferase systems rely on endogenously available fatty acyl-CoA as substrates, this pathway has a mechanism to scavenge fatty acid that is available in the environment.

In addition to the three biosynthetic genes, most pathways also contained some combination of three additional conserved genes (genes labeled“Saccharide” in Fig. 15c). Excluding these genes has no effect on FAA production in Sfp-harboring E. coli (Fig. 6a). One gene has predicted homology with sterol transporters; thus, it is likely facilitating the transport of exogenous fatty acids from the human gut lumen to the bacterial cytoplasm in order to increase in vivo substrate availability of the A protein. The second gene encodes for serine aminopeptidase-like alpha/beta-hydrolase that could either serve to cleave human endogenous FAAs to generate free substrates for the Clostridia pathway or degrade the FAA pathway product as a negative regulatory mechanism much like human FAA hydrolases. The third gene is homologous to Sfp-like PPTase involved in the functionalization of the T protein with a PPT arm. This pathway-specific PPTase gene is likely necessary for FAA production in the native strain, but it is dispensable in the E. coli system because the E. coli strain harbors the B. subtilis- derived Sfp.

Preferred FAA Products Characterized by In Vitro Substrate Screening

The human gut lumen consists of fatty acids and amines from different sources that could be used by the enzymes to build FAAs. The in vitro system was expanded to determine the substrate specificities of the pathways. Rather than test all combinations, two assays were developed to test fatty acid and amine incorporation separately, where the other substrate is held constant. For the fatty acid screening to test adenylation activity, tryptamine is added to all the fatty acid reactions along with the purified T and A proteins and FAA formation is measured based on the UV absorbance (280 nm) of its indole chromophore. The reaction contained no C protein, thereby measuring FAA product as it is non-enzymatically released from the fatty acyl-T produced by the T and A proteins. The incorporation is presented as the ratio of the UV absorbance peak of the fatty acyl-tryptamine product with T and A proteins relative to that obtained from the control containing only T (Fig. 25a). To screen for amine specificity, the same assay is used as for Fig. 16d, and for each pathway, the fatty acid found to be the most active in the fatty acid screening was chosen to be held constant (Fig. 25b).

A panel of 25 fatty acids and 53 amines was collected, representative of biogenic sources present in the human gut, including bacteria, human cells, and diet (Fig. 25c, Table 3). This list includes all of the amines and fatty acids that identified from literature that are abundant in the human body and, for fatty acids, where the amine moiety is exposed for amide bond formation (e.g., excluding choline). The fatty acids and amines that constitute known mammalian FAAs are included in this panel. Some esoteric fatty acids were not included that are low in abundance or not commercially available (e.g., trans fatty acids, cyclopropane fatty acids).

The enzymes associated with each of eighteen pathways (eight from HMP gut metagenome plus ten from the nr database) were screened against the fatty acid and amide panels. The A protein showed a broad range of substrate specificity that is similar amongst the homologues (Figs. 25d and 26). The fatty acid specificity profile of Clostridia A domains resembles human fatty acid binding proteins (FABPs), which also bind a broad range of fatty acids. The correlation is particularly strong with the binding profile of human brain-FABP measured by isothermal titration calorimetry (ITC) (Fig. 27).

In contrast, considering amine incorporation, the C protein has a much narrower substrate specificity (Figs. 25d and 26). Since the A protein is incapable of loading decarboxylated amines and the pathway lacks a separate release domain, the C protein has taken the place to select for specific amines and compete with the nonenzymatic release of the tethered substrate. The substrates of the C proteins show diversity in structure, ranging from straight-chain amines to arylamines. All amine substrates that are present in the major products are human neurotransmitters made by decarboxylation of proteinogenic a- amino acids, including dopamine, tyramine, tryptamine, and aminovaleric acid (GABA analog). For 6 out of the 18 pathways, no major amine substrate was detected (Fig. 28a). An expanded panel of 24 additional amine substrates, including D-amino acids, were screened for these pathways, but none provided enzymatic product formation (Fig. 28b and Table 3).

The major products of each pathway can be deduced from the in vitro screens by pairing the most active fatty acid with the most active amine (Figs. 25d and 26). A cutoff was applied to the two panels to identify minor products, which may be physiologically relevant depending on substrate availability. After being deduced based on the panel screen, each of these products was confirmed by adding the pair of substrates and confirming that the pathway produces the FAA product (Fig. 29). Intriguingly, oleoyl dopamine is known human GPCR- targeting FAA and this is the major product for the M. formatexigens, B. sp. Marseille-P2398, B. wexlerae, L. str. LC2019, C. sp. L2-50, and C. sp. CAG:253 pathways. The other major products are oleoyl tyramine, oleoyl aminovaleric acid, a-linolenoyl phenylethylamine, caproyl tryptamine, and lauroyl tryptamine for the R. bacterium and B. sp. An81 pathways, C. eutactus pathway, R. albus pathway, R. sp. CAG:488 pathway, and E. rectale pathway, respectively (Fig. 19d).

FAA targeting of human GPCRs

FAAs are known to modulate diverse biological functions by interacting with human GPCRs. The products of the FAA screen are either known GPCR ligands or appear to structurally mimic these molecules (Fig. 30). The FAAs were analyzed for activity against a panel of known and orphan human GPCRs using the DiscoverX platform. The major compound from the four human gut metagenome-derived pathways (oleoyl dopamine, oleoyl tyramine, oleoyl aminovaleric acid, and lauroyl tryptamine) were chemically synthesized and screened at 10 uM on a panel of 168 known GPCRs covering 60 distinct receptors in both agonist and antagonist mode and 73 orphan GPCRs in agonist mode. For the four FAAs tested, several GPCR hits were activated or inhibited at above an empirical threshold value provided by DiscoverX, indicating that the interaction is potentially significant. The results of the screen for lauryl tryptamine are shown in Fig. 31a and the other compounds are shown in Figs. 32, 33, and 34.

The four Clostridia FAAs were found to interact with a subset of GPCRs, with some overlapping activities. The glucose homeostasis receptor GPR119 and orphan GPR132 were activated by all of the FAAs tested. While the activity of GPR132 is high for the FAAs tested, GPR119 activity is notably higher for the three with the oleic acid moiety. GPR119 and GPR132 agonist activity have previously been reported for known FAAs, including gut commensal-derived commendamide and oleoyl serinol ¹³ . This suggests that GPR119 and GPR132 are promiscuous for diverse human and bacterial FAAs.

Oleoyl dopamine was found to interact with the most GPCRs, including the previously reported receptors (CNR1, DRD2s, GPR119) and those associated with inflammatory bowel disease (IBD) and inflammation. Oleoyl tyramine differs from oleoyl dopamine by only a single hydroxyl group and it shares activity on NPSRlb and PRLHR, which have been associated with IBD and colorectal cancer. Lauroyl tryptamine was also found to activate known monoamine receptors, specifically serotonin receptors (HTRs). Specific to this compound was the inhibition of EBI2/GPR183, associated with IBD, and P2RY4. Finally, oleoyl aminovaleric acid is structurally similar to the known human FAAs, arachidonyl and oleoyl GABA. Inhibition was observed for PTGER4, associated with IBD, and has been previously reported to be a target for a gut commensal FAA, acyloxyacyl glutamine ¹³ .

Lauroyl tryptamine as EBI2 inhibitor

Lauroyl tryptamine was selected to further validate its GPCR targets. The concentration- response curves were determined for the top three hits with activity/inhibition surpassing the empirical threshold: EBI2 antagonist (against 7a,25-diHC), P2RY4 antagonist (against UTP), and GPR132 agonist (Fig. 31b, c, and d). The concentration-response curve of lauric acid and tryptamine on these GPCRs were also obtained to determine whether the effect can be achieved from the fatty acid or amine substituents alone. Titrations were performed with lauroyl tryptamine concentrations from 0.005 to 100 mM. The 50% inhibitory concentration (IC50) against EBI2 is 0.98 pM in the presence of 0.232 uM 7a,25-diHC (Fig. 31b). Neither lauric acid nor tryptamine showed inhibition on EBI2. The activity is slightly more potent than the previously reported human microbiome-derived fatty acid amide GPCR modulators, including commendamine (GPR132 agonist at 11.8 mM) and oleoyl serinol (GPR119 agonist at 7 pM) ¹² ’ ¹³ .

In contrast to EBI2, lauroyl tryptamine shows modest concentration-dependent inhibition on P2RY4 at the highest tested concentration. Although lauroyl tryptamine exhibited activity on GPR132 with and EC50 of 1.45 pM, free lauric acid also showed activity on GPR132 with EC50 of 25.2 pM. Therefore, the hits against P2RY4 and GPR132 may have been false positives.

FAA production by Eubacterium rectale

Experiments were performed to determine whether the FAAs identified through the in vitro screen are produced by the gut commensal strain containing the native gene cluster. Doing so is not always possible because the pathways are obtained from metagenomic data and the original species can be unobtainable. However, two native Clostridia strains were identified in a public repository (Leibniz Institute DSMZ), M. formatexigens (DSM 14469) and E. rectale (DSM 17629), and tested these for the production of the FAAs identified in vitro. These strains were cultivated under different laboratory conditions, initially without fed substrates. LC-MS analysis of extract from E. rectale grown in RCM media revealed the presence of lauroyl tryptamine (titer of 0.04 mg/L), the major compound identified from the in vitro expression of the E. rectale- derived pathway (Fig. 31e and 35). Although extracellular lauric acid is toxic to certain Clostridia strains, some Clostridia has been reported to endogenously produce lauric acid ⁶⁶ . None of the minor compounds from the in vitro study were detected from the E. rectale culture extract; however, supplementation of the media with alternative substrates led to the minor products being produced (Fig. 3 If). The titer increased to 0.2 mg/L when the substrates were added to the media. FAA production was not detected for M. formatexigens under the conditions tested.

The culture conditions may not be reflective of the environmental niche occupied by the bacteria in the human gut. Directly measuring the production of a FAA in this environment would be difficult. Instead, we analyzed published metatranscriptomic data obtained from human isolates ⁴⁷ . The reads from RNA-sequencing datasets from the stool samples of eight healthy subjects were searched with the E. rectale pathway as the query sequence. Transcriptional reads across the FAA pathway were detected in three out of the eight human samples (subX319146421, subX311245214, subX316701492). Transcription was further analyzed for subject subX316701492 by mapping the reads to a 80 kb region of the E. rectale genome centered on the FAA pathway (Fig. 3 lg). The mean coverage of the 80 kb segment was 0.8 (SD = 3.3), while the mean coverage of the FAA pathway by itself was 1.5 (SD = 1.1). The pathway expression level is comparable to that of surrounding housekeeping genes ( e.g . sugar biosynthesis, cell wall biogenesis), thus confirming transcription of the FAA pathway in the host.

Strains, plasmids, and media.

E. coli DH10B (C3019, New England Biolabs, Ipswich, MA) was used for routine cloning. E. coli BAP1 containing T7 DNA polymerase and Sfp phosphopantetheinyl transferase was used for heterologous expression of engineered pathways ⁵⁵ . E. coli BL21(DE3) (CMC0016, Millipore Sigma, St. Louis, MO) was used for protein expression and pathway expression in the absence of Sfp. Vector pET28a (69864, Millipore Sigma) was used as backbone for all pathway and protein expression constructs (Fig. 21). LB media (L3152, Millipore Sigma) was used for routine cloning and in vivo compound production. M9 minimal media, consisting of M9 Minimal Salts (M6030, Millipore Sigma), 2 mM MgS0 ₄ (230391, Millipore Sigma), 100 uM CaCl ₂ (97062, VWR, Radnor, PA), 0.4% glucose (BDH9230, VWR), was used as an alternative media for in vivo compound production. LB + 1.5% agar (214010, BD, Franklin Lakes, NJ) was used for growth on solid media. Kanamycin (K4000, Millipore Sigma) was used for selection at 30 ug/ml. IPTG (12481, Gold Biotechnology, St. Louis, MO) was used at final concentrations of 0.1 - 1.0 mM IPTG for induction of T7-driven expression. Marvinbryantia formatexigens DSM 14469 and Eubacterium rectale DSM 17629 were acquired from Leibniz Institute DSMZ (Braunschweig, Germany).

Pathway design and gene synthesis.

Pathways including the biosynthetic and accessory genes were re-designed such that each gene was codon optimized for E. coli without Type IIS restriction sites (Gene Art CodonOptimizer, Thermo Fisher Scientific, Waltham, MA), placed under T7 promoter (with lacO operator) and RBS parts from pET28a, and arranged in the same direction in the order that they appear natively (Fig. 21 and Tables 1-2). The re-designed pathways were synthesized by either Thermo Fisher Scientific or Gen9 (Cambridge, MA) with flanking Bsal (R733, New England BioLabs) sites. The synthesized pathways, either as purified fragment or cloned in default vector, were cloned into pET28a using Golden Gate assembly and transformed into E. coli BAP1.

Reagents and chemicals.

Fatty acid and amine substrates for the panel assay are summarized in Table 3. Fatty acid substrates were prepared as 50 mM stock solutions in ethanol neutralized with sodium hydroxide solution, and stored at -80°C prior to use. Amine substrates were prepared as 50 mM stock solutions in water, and stored at -80°C prior to use.

Computational detection of Clostridia NRPS pathways.

Genome datasets from the human gut were searched on the genome browser of the JGI-IMG database querying "Gastrointestinal tract" as the Sample Body Site and "Human" as a keyword. The metadata from search result was used to locate and download the sequencing data from either JGI GOLD or NCBI GenBank database. The resulting datasets were run on antiSMASH 3.0 using default parameters with ClusterFinder-based border prediction, but excluding putative pathways detected by ClusterFinder (low-confidence). The antiS MAS H-detected pathways were blastn searched on the metagenomic reads of 148 fecal samples from HMP that passed QC assessment, retaining pathways that had at least two hits with e-value < 1X10 ⁵ from the metagenomic reads spanning different stretches of the pathway. These were blastn searched, using the same cutoff filter as the previous step, on the mRNA reads from fecal samples of eight healthy subjects. The remaining pathways were run on BiG-SCAPE to generate a similarity network matrix file with a distance cutoff of 0.75 and visualized using Cytoscape ⁴⁸ . For 16S rRNA-based phylogenic tree, 16S rRNA sequences of the strains were collected from NCBI database. The tree was constructed from the sequences using MacVector version 16 (Method: MUSCLE alignment, Neighbor Joining Method, Distance: Uncorrected, Best Tree Mode).

MetaQuery search.

For analyzing prevalence, the condensation domain protein sequence for each of the pathways was used as the query for MetaQuery search with default parameters, except with a minimum percent identity of 98 ⁵¹ .

Pathway homology search.

E. rectale condensation protein (NCBI: WP_015516887) and adenylation protein (NCBI: WP_015516889) were each used as a query for blastp search with default parameters. Upon removing hits that were different in protein size (>800 or <200 amino acids) from the top 1,000 hits, the hit table from each was cross-examined using Python script for their co-occurrence in the same pathway based on proximity in NCBI accession number.

Biosynthetic enzyme homology search.

The E. rectale biosynthetic protein sequences EreC and EreA were individually blastp searched on UniProt (EMBL-EBI, United Kingdom) using default parameters and “UniProtKB/Swiss-Prot” of characterized proteins as the target database.

Heterologous expression of clusters in E. coli.

To test for in vivo compound production under a comprehensive set of fermentation conditions, a total of 24 cultures were set up that each make up the possible combinations of the following four parameters: timing of IPTG induction, IPTG concentration, culture media, and pathway expression temperature. Overnight cultures of pathway-harboring E. coli in LB broth at 230 rpm and 37°C (MS012NF, Multitron Standard, INFORS HT, Bottmingen, Switzerland) were diluted 200-fold the next morning to 4 mL fresh LB media in partially unscrewed 50 mL Conical tube (352098, Thermo Fisher Scientific). Upon growth to one of three IPTG induction times (corresponding to ODeoo = 0.2, 0.5, and 0.8, respectively), the culture was pelleted at 4,000 g and 4 °C for 3 min (75004537, Multifuge X3 FR, Thermo Fisher Scientific). The pellet was resuspended in 4 mL of one of two fresh culture media (LB or M9) at one of two IPTG concentrations (0.2 or 0.5 mM) for induction. The culture was grown with shaking (230 rpm) at one of two temperatures (25 or 30 °C). To monitor compound production over time, 1 mL sample was removed from each of 24 cultures after three different timepoints (8, 16, or 40 hr). Each sample was immediately extracted with 1 mL of methanol (BJLC230, VWR), mixed with the vortexer at maximum speed for 30 sec (SI-0236, Vortex-Genie 2, Scientific Industries, Bohemia, NY), and pelleted at 20,000 g at room temperature for 1 min (022620401, Centrifuge 5424, Eppendorf, Hamburg, Germany). A 100 uL aliquot of the supernatant was injected for analytical LC-MS run (below). Clone- specific compound production was identified by visual inspection of clone- specific peaks in the chromatograph of electrospray ionization (ESI)+ total ion current (TIC), ESI- TIC, and diode array detector (DAD) total wavelength using software ChemStation (version 1.9, Agilent) and MestReNova (version 10, Mestrelab Research, Compostela, Spain). For side-by-side compound production comparison of E. rectale pathway with gene- knockout variants or with fed substrate, each LCMS analytical sample was prepared following the same protocol, but only with a single fermentation condition and at a single extraction time point. In brief, the freshly inoculated 4 mL culture from overnight was grown to ODeoo = 0.8. The pelleted culture was resuspended in fresh LB containing 0.2 mM IPTG. For substrate feeding assay, the IPTG-containing LB also included 5 mM octanoic acid (neutralized with sodium hydroxide) or tryptamine. The culture was grown with shaking at 25 °C and extracted after 16 hr. The presence of expected product ( i.e . palmitoleoyl putrescine, octanoyl putrescine, palmitoleoyl tryptamine) was identified by inspection of a clone- specific peak in the ESI+ extracted ion chromatograph (EIC) corresponding to the mass-to-charge ratio (m/z, 0.3 Da tolerance) of the protonated ion.

Analytical LC-MS.

Analytical LCMS was conducted using an Agilent (Agilent, Santa Clara, CA) 1260 Infinity system with 6130 quadrupole MS, binary pump (G1312B), and DAD (G1315D). The sample was run on a C18 reverse phase column [Luna 5 um Cl 8(2), 100 x 4.6 mm] (00D-4252-E0, Phenomenex, Torrance, CA) with 1.0 mL/min flow rate and a gradient system of 90%: 10% to 0%: 100% water (WX0001, Millipore Sigma) : acetonitrile (BJLC015, VWR) with 0.1% formic acid (5330020050, Millipore Sigma) for 15 min, followed by 2 min isocratic run at 100% acetonitrile (wash) and then 3 min at 90%: 10% water: acetonitrile (re-equilibration).

E. coll in vivo compound characterization.

Large-scale fermentation of pathway-harboring E. coli (16 L) was conducted. Overnight culture was diluted 200-fold to 1 L fresh LB in 4 L Erlenmeyer flasks (10545-845, VWR), and grown at 200 rpm and 30°C (Multitron Standard). At ODeoo of around 0.4, the temperature was lowered to 25°C. The culture at ODeoo = 0.8 and 25°C was then added with IPTG at a final concentration of 0.2 mM and grown with shaking at 25 °C for 16 hr. The 16 L culture was extracted with equal volume of ethyl acetate (JT9282, VWR), 1 L at a time in separatory funnel (4301-2000, Thermo Fisher Scientific). The organic layer was concentrated in vacuo using rotary evaporator (11100C2102, Rotavapor R-100, Buchi, Flawil, Switzerland), transferred to glass vial (66030-678, VWR), and dried using SpeedVac concentrator (SPD2010-230, Savant SpeedVac, Thermo Fisher Scientific). Preparatory LCMS was conducted using an Agilent (Agilent, Santa Clara, CA) 1260 Infinity system with quaternary pump (G1311B), DAD (G1315D), and fraction collector (G1364B). The extract resuspended in 5 mL acetonitrile was run on preparatory C18 reverse phase column [Luna 5 um Cl 8(2), 250 x 21.2 mm] (00G-4252-P0-AX, Phenomenex) and a gradient system of 90%: 10% to 0%: 100% water: acetonitrile with 0.1% acetic acid (A11350, Thermo Fisher Scientific) in 20 min at 10 mL/min. Major compound eluted with the 20%:80% water: acetonitrile fractions, which were pooled and dried in vacuo. Upon resuspension in 2 mL acetonitrile, the compound was then purified over the course of ten injections (200 uL at a time) on a semi-prep C18 column [Luna 5 um Cl 8(2), 250 x 10 mm] (00G-4252-N0, Phenomenex) and a gradient system of 70%:30% to 0%: 100% watenmethanol with 0.1% acetic acid in 30 min at 5 mL/min. The major compound eluted in the 45%:65% watenmethanol fraction as a white powder (0.1 mg). HRMS acquired on 6530 Q-TOF (Agilent) and 1-D and 2-D NMR spectra in chloroform-*:/ (151823, Millipore Sigma) collected on Avance II 600 (Bruker, Billerica, MA) were used to determine chemical structure (Fig. 5). The compound identity was further confirmed by having a standard chemically synthesized by KareBay Biochem (Monmouth Junction, NJ) and was found to be spectroscopically identical.

Protein expression and purification.

The individual genes were cloned into the pET28a plasmid system shown in Fig. 21 and Tables 1-2. The plasmids were transformed into E. coli BL21(DE3). Overnight E. coli cultures containing the plasmids were diluted 200-fold to 1 L fresh LB in 4 L Erlenmeyber flask (10545-845, VWR) and grown at 200 rpm and 25°C (Multitron Standard). At ODeoo = 0.35, the temperature was lowered to 16°C. The culture at ODeoo = 0.5 and 16°C was then added with IPTG at a final concentration of 0.5 mM and grown with shaking at 16 °C for 18 hr. The culture was transferred to centrifuge bottles (75007300, Thermo Fisher Scientific) and pelleted at 4,000 g and 4 °C for 20 min (75004537, Multifuge X3 FR, Thermo Fisher Scientific). As a wash step, the pellet was resuspended in 200 mL Lysis Buffer, consisting of 200 mM NaCl (AB01915, American Bioanalytical, Canton, MA), 10 mM imidazole (15513, Millipore Sigma), 50 mM Tris (AB02000, American Bioanalytical) pH 8.0, and EDTA-free protease inhibitor cocktail (5056489001, Millipore Sigma). Upon pelleting again at 4,000 g 4 °C for 20 min, it was resuspended in 10 mL Lysis Buffer, transferred to 50 mL Conical tube (352098, Thermo Fisher Scientific), and lysed using a sonicator (Q125, Qsonica, Newtown CT) with 1/8" stepped microtip probe (4422, Qsonica) at 45% amplitude for 5 min continuously on ice.

Upon centrifuging the lysed cells at 14,000 g for 20 min (75004520, Sorvall Legend XTR Thermo Fisher Scientific), the supernatent was transferred to a new 50 mL Conical tube, mixed with 1 mL of pre-equilibrated Ni-NTA resin (88221, Thermo Fisher Scientific), and rotated using Tube Rotator at 20 rpm and 4°C for 16 hr (1205R81, Scilogex, Rocky Kill, CT). The bead slurry was spun down at 1,000 g and 4°C for 1 min (Multifuge X3 FR), and resuspended in 10 mL Wash Buffer, containing 200 mM NaCl, 20 mM Imidazole, and 50 mM Tris, pH 8.0. Upon transferring to a new 15 mL Conical tube (352097, Millipore Sigma), the bead slurry was washed twice with 10 mL Wash Buffer by spinning down at 1,000 g and 4°C for 1 min. The bead slurry was then resuspended in 2 mL Wash Buffer, transferred to new 2 mL microcentrifuge tubes (022600044, Eppendorf), and spun down at 5000 g and 4°C for 1 min (Centrifuge 5424R, Eppendorf). The bead slurry was added with 2 mL of Elution Buffer, consisting of 200 mM NaCl, 250 mM imidazole, and 50 mM Tris, ph 8.0, and spun down at 20,000 g and 4°C for 1 min. The supernatant containing the purified protein was dialyzed using Slide-A-Lyzer, 3.5K MWCO (66333, Thermo Fisher), in 500 uL aliquot per cassette. Each cassette was submerged at 4°C for 6 hr in 14 mL of Dialysis Buffer, consisting of 50 mM NaCl, 1 mM TCEP (TCEP, Gold Biotechnology), and 10% glycerol (AB00751, American Bioanalytical). The buffer was discarded, changed to a fresh Dialysis Buffer, and left at 4°C for another 16 hr. The dialyzed protein sample was pooled from the cassettes, distributed in 20 uL aliquots for single freeze-cycle use, and frozen for storage at -20 °C. Protein purity was confirmed by the presence of a single band corresponding to the expected size on Novex 16% Tricine protein gel (EC66952BOX, Thermo Fisher) in Tricine running buffer (LC1675, Thermo Scientific) stained with SimplyBlue SafeStain (LC6060, Thermo Fisher) alongside with Precision Plus Protein ladder (1610374, Bio-Rad Laboratories, Hercules, CA) (Supplementary Fig. 20). The protein sample was quantified by Bradford Assay following manufacturer protocol using Coomassie Plus (23236, Thermo Fisher). In brief, protein standards were prepared with bovine serine albumin at concentrations of 0; 25; 125; 250; 500; 750; 1,000; 1,500; 2,000 ug/mL. A 3.3 uL sample of each standard or protein sample (in triplicates) was pipetted into microplate wells (CLS3894, Millipore), added with 100 uL of the Coomassie reagent, and mixed in plate shaker at 900 rpm for 30 sec (I10103P, Multitron Pro, INFORS HT). After incubation at room temperature for 10 min, the plate was spun down at 4,000 g for 1 min (Multifuge X3 FR) to remove particulates. Upon transferring the supernatant into half-area microplate wells (CLS3679, Millipore Sigma), absorbance of each well at 595 nm was read using Synergy HI Microplate Spectrophotometer (8041000, BioTek, Winooski, VT). The mean of the triplicate absorbance of the samples/standards was subtracted by the mean triplicate absorbance of blank water. The mean value of BSA standards was plotted using Prism (version 7, GraphPad, San Diego, CA) to prepare a standard curve and used to determine protein concentration from mean absorbance value. The exception to the above purification strategy was for the condensation protein from C. eutacus, where a presence of stronger and broader Coomassie band was detected on the protein gel at > 250 kDa, suggesting aggregation. The pCeuC construct was thus sent to ABclonal Technology (Wobum, MA) for custom protein expression, purification, and quantification.

ATP consumption (PPi) assay.

The EnzChek Pyrophospate Assay Kit (E6646, Thermo Fisher) was used for pyrophosphate detection to measure adenylation activity. A pair of 100 pL in vitro reaction samples were prepared, one with E. rectale adenylation and thiolation proteins (no condensation) and the other with thiolation protein only, each protein at 1 mM final concentration. The protein(s) were added with query substrate at 1 mM final concentration (palmitoleic acid, palmitic acid, putrescine, ornithine, arginine, agmatine, cadaverine, or lysine) in an in vitro reaction mix, consisting of 100 mM Tris, 10 mM MgCk (JT2448, VWR), 1 mM TCEP, 0.1 uM Sfp Synthase (P9302, New England Biolabs), 0.1 mM Coenzyme A (951-50, Lee Biosolutions, Maryland Heights, MO), and 5 mM ATP (A-081, Gold Biotechnology). After 60 min incubation at 23°C, 10 uL of each reaction sample was mixed with kit solution containing 0.2 mM MESG dye, 1 U mL ¹ purine nucleoside phosphorylase, 0.03 U mL ¹ inorganic pyrophosphatase, and 1 x reaction buffer in 100 pL total volume. Upon transferring into microplate wells (CLS3894, Millipore Sigma), the sample was mixed in plate shaker at 900 rpm for 30 sec (I10103P, Multitron Pro, INFORS HT). After incubation at room temperature for 60 min, the plate was spun down at 4,000 g for 1 min (Multifuge X3 FR) to remove particulates. Upon transferring the supernatant into half-area microplate wells (CLS3679, Millipore Sigma), absorbance of each well at 360 nm was read using Synergy HI Microplate Spectrophotometer (8041000, BioTek, Winooski, VT). Absorbance from each sample was subtracted by the mean triplicate absorbance of blank water. The absorbance from reaction with adenylation and thiolation proteins was normalized to thiolation protein only control. The data was taken in triplicate per day, and the average of the mean triplicate normalized absorbance conducted on three different days was reported.

LC-MS detection of T protein intermediate peptide. A pair of 1000 pL in vitro reaction samples were prepared with 1 mM final concentration of the E. rectale adenylation protein, 1 pM of the thiolation protein, and 1 mM lauric acid in an in vitro reaction mix consisting of 100 mM Tris, 10 mM MgCh, 1 mM TCEP, 0.1 uM Sfp, and 0.1 mM Coenzyme A. ATP at 5 mM final concentration was added to only one of the sample. After 60 min incubation at 23 °C, the proteins in the reaction mixtures were denatured and digested with trypsin (V511A, Promega, Madison, WI) following manufacturer’s protocol, but at pH 7.0 to prevent base-catalyzed hydrolysis of the thiotemplated substrate. In brief, the mixtures were incubated at 90°C for 20 min in 6M guanidine HCL (G3272, Sigma-Aldrich) and 1M DTT (D9779, Sigma-Aldrich) in Tris-HCl (50 mM pH 7.0) before cooling to room temperature. Upon 6-fold dilution of the solution with Tris-HCl (50 mM pH 7.0) and CalCh (1 mM) down to 0.96 M of guanidine HC1, the protein solutions were added with trypsin to a final protease:protein ratio of 1:50 (w/w) for 24 hr at 37°C. The resulting mixtures were cleaned using a C-18 Sep-Pak column (WAT0519, Waters, Milford, MA). For peptide analysis, the samples were injected on Agilent 6530 qTOF MS in ESI+ mode coupled to Agilent 1290 Infinity UHPFC system. The samples were run on AdvanceBio Peptide Mapping column [Cl 8, 2.7 um, 120 A, 150 x 2.1 mm] (653750-902, Agilent) with a 0.2 mF/min flow rate and a gradient system of 97%:3% to 35%:65% water: acetonitrile with 0.1% formic acid for 75 min. Sequences for peptides were identified using Agilent MassHunter BioConfirm (version 8) with a mass tolerance of 0.25 Da, MS and MS ^E mass match tolerance of 30.0 ppm and trypsin digest with 0 missed cleavages.

MS-based measurement for condensation activity.

A pair of 100 pF in vitro reaction samples were prepared, one with E. rectale condensation, adenylation, and thiolation proteins and the other with adenylation and thiolation proteins only, each protein at 1 mM final concentration. The proteins were added with query amine at 1 mM final concentration (putrescine, ornithine, phenylalanine, tryptophan, tyrosine, phenylethylamine, tryptamine, or tyramine) and 1 mM of fatty acid (palmitoleic acid) in an in vitro reaction mix (above). After 90 min incubation at 23°C, each sample was extracted with 1 mL of methanol, mixed with the vortexter at maximum speed for 30 sec (SI-0236, Vortex- Genie 2), pelleted at 20,000 g and room temperature for 1 min. A 100 ul of the supernatant was injected for analytical FC-MS run (above). Using MestReNova (version 10), the presence of FAA product was identified by inspection of a clone- specific peak in the ESI+ and EST extracted ion chromatograph (EIC) corresponding to the mass-to-charge ratio (m/z, 0.3 Da tolerance) of the protonated or deprotonated ion. Automated trace baseline correction, peak detection, and peak integration were applied by MestReNova. The peak area (ESI+ or EST depending on the amine substrate) from reaction with condensation, adenylation and thiolation proteins was normalized to adenylation and thiolation proteins only control. The data was taken in triplicate per day, and the average of the mean triplicate normalized peak area conducted on three different days was reported.

Substrate panel assay.

Adenylation and condensation activities from the eight HMP-derived pathways were measured on a panel of 25 fatty acids and 53 amines, respectively. Fatty acids were mixed with equimolar amounts to prepare five subpools, such that no two compounds sharing the same mass are in the same subpool: subpool Fa = FI, 6, 10, 13, 17, 20, 21; Fb = F2, 9, 11, 16, 22, 24; Fc = F3, 5, 7, 12, 14, 18; FD = Fd, 8, 15, 19, 23, 25 (Fig. 25 and Table 3). Amines were similarly mixed to prepare six subpools: subpool Aa = A9, 27, 28, 29, 30, 31, 32, 33, 34; Ab = Al, 2, 3, 4, 12, 16, 17, 20, 23; Ac = 5, 11, 14, 21, 35, 37, 39, 42, 43; Ad = A6, 7, 8, 10, 13, 15, 18, 19, 38; Ae = 22, 24, 25, 26, 40, 41, 47, 51, 53; Af = 36, 44, 45, 46, 48, 49, 50, 52 (Fig. 25 and Table 3). For the additional panel of amines that were tested on the L. clostridioforme, B. producta, C. sp. 1_7_47FAA, C. celatum, C. sp. CAG.413, and L. str. KH1P17 pathways, they were prepared as the following subpools: subpool Ag = 54, 55, 56, 57, 58, 59, 60, 62; Ah = 61, 63, 64, 65, 66, 67, 68, 70; Ai = 69, 71, 72, 73, 74, 75, 76, 77 (Fig. 28 and Table 3).

For MS-based measurement of adenylation activity, a pair of 100 pF in vitro reaction samples were prepared, one with adenylation and thiolation proteins (no condensation) and the other with thiolation protein only, each protein at 1 mM final concentration. The protein(s) were added with one of the fatty acid subpool (Fa, Fb, Fc, or Fd) at a final concentration of 1 mM total fatty acid and tryptamine at a final concentration of 10 mM in an in vitro reaction mix (above). After 60 min incubation at 23 °C, each sample was extracted with 1 mL of methanol, mixed with the vortexter at maximum speed for 30 sec (SI-0236, Vortex-Genie 2), and pelleted at 20,000 g and room temperature for 1 min. A 100 uF of the supernatant was injected for analytical FC-MS run (above). Using MestReNova (version 10), the presence of fatty acyl tryptamine product from each fatty acid substrate was identified by inspection of a clone- specific peak in the ESI+ extracted ion chromatograph (EIC) corresponding to the mass- to-charge ratio (m/z, 0.3 Da tolerance) of the protonated ion. The corresponding peak at the same retention time was identified on the DAD chromatogram at 280 nm (0.5 nm tolerance). Automated trace baseline correction, peak detection, and peak integration were applied by MestReNova. The DAD chromatogram peak area from reaction with adenylation and thiolation proteins was normalized to thiolation protein only control. The data was taken in triplicate per day, and the average of the mean triplicate normalized peak area conducted on three different days was reported.

The MS-based measurement of condensation activity follows similar steps as described (above). A pair of 100 pL in vitro reactions were prepared, one with condensation, adenylation, and thiolation proteins and the other with adenylation and thiolation protein only, each protein at 1 mM final concentration. The proteins were added with one of the amine subpool (Aa, Ab, Ac, Ad, Ae, Af, Ag, Ah, or Ai) at a final concentration of 1 mM total amine and fatty acid that provided highest adenylation activity (oleic acid for C. eutactus, M. formatexigens, R. bacterium, B. sp. Marseille-P2398, B. wexlerae, L. str. LC2019, C. sp. L2-50, C. sp. CAG.253, and B. sp. An81 pathways; a-linolenic acid for R. albus, L. clostridioforme, B. producta, C. celatum, and C. sp. 1_7_47AA pathways; lauric acid for C. eutactus, C. sp. CAG.413, and L. str. KH1P17 pathways; capric acid for R. sp. CAG:488 pathway) at a final concentration of 1 mM in an in vitro reaction mix (above). After 90 min incubation at 23°C, each sample was extracted with methanol and prepped for 100 uL analytical LC-MS run as described (above). Using MestReNova (version 10), the presence of FAA product for each amine substrate was identified by inspection of a clone- specific peak in the ESI+ and ESI- extracted ion chromatograph (EIC) corresponding to the mass-to-charge ratio (m/z, 0.3 Da tolerance) of the protonated or deprotonated ion. Automated trace baseline correction, peak detection, and peak integration were applied by MestReNova. The peak area (ESI+ or ESI- depending on the amine substrate) from reaction with condensation, adenylation and thiolation proteins was normalized to adenylation and thiolation proteins only control. The data was taken in triplicate per day, and the average of the mean triplicate normalized peak area conducted on three different days was reported.

With 1 mM each of lauric acid and tryptamine as substrates, and 1 mM each of the biosynthetic proteins from the E. rectale system, the reaction reached steady state at 18% conversion of lauric acid and 15% conversion of tryptamine after 8 hours at 23°C.

The identities of the major product determined from the in vitro assay (oleoyl dopamine, oleoyl tyramine, lauroyl tryptamine, oleoyl aminovaleric acid, a-linolenoyl phenylethylamine, and caproyl tryptamine) and a-linolenoyl homocysteine were confirmed for each by setting up an in vitro reaction with the specific pair of substrates. The resultant FAA product was analyzed by injecting the sample for analytical LC-MS alongside with a structurally verified standard, obtained either by purchasing from Cayman Chemical (Ann Arbor, MI; oleoyl dopamine, 10115), or having them chemically synthesized by KareBay Biochem (Monmouth Junction, NJ).

The reaction time used in the adenylation and condensation activity assays was based on the time course of ten different time points. In brief, the MS -based measurements of adenylation and condensation activity were conducted as described above with C. euctacus proteins. However, a pair of 1,200 pL in vitro reactions were prepared, and instead of a single timepoint, a 100 pL aliquot was collected after 10, 20, 30, 45, 60, 90, 120, 240, 480, and 960 min of incubation at 23°C, extracted, and injected for analytical LC-MS run. The data was taken in triplicate per day. The time course graph was plotted for the average of the mean triplicate normalized peak area run on three different days and shows that the 60 min and 90 min timepoint for adenylation and condensation activity assay, respectively, resulted in the largest difference between the highest activity and the rest (Fig. 11).

Native clostridia strain compound characterization.

Marvinbryantia formatexigens DSM 14469 and Eubacterium rectale DSM 17629 were grown in an anaerobic chamber (1200001, Coy Laboratory Products, Grass Lack, MI) with an incubator (6100000, Coy Laboratory Products). The strains were inoculated in one of the following 10 pre -reduced liquid media: RCM (BD218081, Becton Dickinson, Franklin Lakes, NJ), BHI (M210, HiMedia, Mumbai, India), GAM (M1801, HiMedia), TSB (BD211825, Becton Dickinson), Casman (M766, HiMedia), WCABB (M863, HiMedia), Columbia (BD294420, Becton Dickinson), ABB (M1636, HiMedia), YCFAC (AS-680, Anaerobe Systems, Morgan Hill, CA), PYEG (AG24H, Hardy Diagnostics, Santa Maria, CA). Media without reducing agent L-cysteine in the ingredients were supplemented with L-cysteine at a final concentration of 0.05% (w/v). Each culture was grown as 2 mL volume in 14 mL Falcon culture tube (352059, Corning) at 37 °C for 2 days. The culture was extracted in the chamber with 2 mL of methanol and mixed by vigorous shaking (BJLC230, VWR). Upon taking the sample out from the chamber, the mixture was pelleted at 20,000 g for 1 min (022620401, Centrifuge 5424). A 100 uL aliquot of the supernatant was injected for analytical LC-MS run, as previously described. The presence of lauroyl tryptamine in E. rectale and oleoyl dopamine in M. formatexigens was checked by running alongside the synthesized standard and inspecting the ESI+ EIC corresponding to the m/z (0.3 Da tolerance) of the protonated ion. For E. rectale feeding experiment, 20 pL of the 2-day E. rectale culture in RCM was inoculated in 2 mL of fresh RCM with 0.2 mM of tryptamine and 0.1 mM of the fatty acid substrate (lauric acid, linoleic acid, oleic acid, docosahexaenoic acid, or a-linolenic acid), and grown for an additional 2 days at 37°C. The culture was extracted and analyzed by LC-MS in the same way as above, with methanol extraction and injection of a 100 uL aliquot. Standards for linoleoyl, oleoyl, docosahexaenoyl, and a-linolenoyl tryptamine were made by adding the appropriate substrates in the in vitro system with purified E. rectale enzymes, and were run alongside to confirm presence in the E. rectale extracts (ESI+ EIC).

For product yield measurements, a standard curve for fatty acyl tryptamine was constructed based on the peak area of the DAD chromatogram at 280 nm (0.5 nm tolerance) with 100 pg, 10 pg, 1 pg, 100 ng, 10 ng, 1 ng, 100 pg, 10 pg, and 1 pg injections (in triplicates) of the chemically synthesized lauroyl tryptamine standard. Automated trace baseline correction, peak detection, and peak integration were applied by MestReNova.

Metatranscriptomics.

The E. rectale FAA pathway was first used as a query sequence for blastn search with default parameters on NCBI SRA datasets. A 80 kb region of the E. rectale genome centered at the FAA pathway was obtained from the NCBI nr database (accession: FP929042, region 2065556-2145556). The metatranscriptomic reads file (fastq) from subject subX316701492 was obtained from the NCBI Sequence Read Archive (SRA) database (accession: SRX247340). Upon loading the files to Geneious Prime (version 11), the reads were mapped onto the 80 kb region as the reference sequence (Geneious Mapper, Medium-Low Sensitivity/Fast).

Chemical synthesis.

The FAA products lauroyl tryptamine, oleoyl tyramine, oleoyl aminovaleric acid, a-linolenoyl phenylethylamine, caproyl tryptamine, a-linolenoyl homocysteine, and palmitoeyl putrescine were chemically synthesized by KareBay Biochem (Monmouth Junction, NJ). Oleoyl dopamine was acquired from Cayman Chemical (Ann Arbor, MI).

GPCR screening.

The four human gut metagenome-derived major products (oleoyl dopamine, oleoyl tyramine, lauroyl tryptamine, oleoyl aminovaleric acid) were obtained from commercial source or chemical synthesis and confirmed to be identical to in vitro products. They were sent to DiscoverX (Fremont, CA) for a cell-based assay on a panel of 168 GPCRs with known ligands (gpcrMAX) in both agonist and antagonist mode, as well as 73 orphan GPCRs (orphanMAX) in agonist mode. Chemiluminescence indicating b-arrestin recruitment was used as output to measure agonist and antagonist activity of each compound on each GPCR at 10 micromolar concentration. Agonist mode measures % activity relative to the baseline value (0% activity) and maximum value activated by a known ligand (100% activation). For orphan GPCR, twofold increase in value over baseline is set as 100% activation. Antagonist mode measures % inhibition of target GPCR by the compound in the presence of a known ligand, relative to the value at the EC80 of the known ligand (0% inhibition) and basal value (100% inhibition). DiscoverX provides an empirical threshold value of 30%, 35%, or 50% for GPCR agonist, GPCR antagonist, or orphan GPCR agonist, respectively, where activity/inhibition higher than the threshold indicate that the interaction is potentially significant.

Concentration-response curve.

Lauroyl tryptamine, tryptamine, and lauric acid were sent to DiscoverX for a cell-based assay on EBI2 antagonist mode, P2RY4 antagonist mode, and GPR132 agonist mode. The experimental setup and output are the same as the panel assay, except % inhibition/activity was measured on ten concentration points (100, 33, 11, 3.7, 1.2, 0.41, 0.14, 0.046, 0.015, 0.0051 mM) to plot the concentration-response curve and determine IC50 or EC50.

References Section 2

(Below references refer to reference numbers included only in Example 2)

1 Wilson, M. R., Zha, L. & Balskus, E. P. Natural product discovery from the human microbiome. The Journal of biological chemistry 292, 8546-8552, doi: 10.1074/jbc.Rl 16.762906 (2017).

2 Lee, W. J. & Hase, K. Gut microbiota-generated metabolites in animal health and disease. Nature

chemical biology 10, 416-424, doi: 10.1038/nchembio. l535 (2014).

3 Donia, M. S. & Fischbach, M. A. HUMAN MICROBIOTA. Small molecules from the human

microbiota. Science 349, 1254766, doi: 10.1126/science.1254766 (2015).

4 Sharon, G. et al. Specialized metabolites from the microbiome in health and disease. Cell metabolism 20,

719-730, doi: 10.1016/j.cmet.2014.10.016 (2014).

5 Chen, H. et al. A Forward Chemical Genetic Screen Reveals Gut Microbiota Metabolites That Modulate Host Physiology. Cell, doi: 10.1016/j.cell.2019.03.036 (2019).

6 Wang, Z. et al. Gut flora metabolism of phosphatidylcholine promotes cardiovascular disease. Nature

472, 57-63, doi: 10.1038/nature09922 (2011). 7 Ridlon, J. M., Kang, D. J. & Hylemon, P. B. Bile salt biotransformations by human intestinal bacteria. Journal of lipid research 47, 241-259, doi: 10.1194/jlr.R500013-JLR200 (2006).

8 Guo, C. J. et al. Discovery of Reactive Microbiota-Derived Metabolites that Inhibit Host Proteases. Cell

168, 517-526 e518, doi: 10.1016/j.cell.2016.12.021 (2017).

9 Wieland Brown, L. C. et al. Production of alpha-galactosylceramide by a prominent member of the human gut microbiota. PLoS biology 11, el001610, doi:10.1371/journal.pbio.1001610 (2013).

10 Ozaki, H. et al. Molecular structure of the toxin domain of heat-stable enterotoxin produced by a

pathogenic strain of Escherichia coli. A putative binding site for a binding protein on rat intestinal epithelial cell membranes. The Journal of biological chemistry 266, 5934-5941 (1991).

11 Round, J. L. et al. The Toll-like receptor 2 pathway establishes colonization by a commensal of the human microbiota. Science 332, 974-977, doi: 10.1126/science.1206095 (2011).

12 Cohen, L. J. et al. Functional metagenomic discovery of bacterial effectors in the human microbiome and isolation of commendamide, a GPCR G2A/132 agonist. Proceedings of the National Academy of Sciences of the United States of America 112, E4825-4834, doi:10.1073/pnas.1508737112 (2015).

13 Cohen, L. J. et al. Commensal bacteria make GPCR ligands that mimic human signalling molecules.

Nature 549, 48-53, doi: 10.1038/nature23874 (2017).

14 Donia, M. S. et al. A systematic analysis of biosynthetic gene clusters in the human microbiome reveals a common family of antibiotics. Cell 158, 1402-1414, doi:10.1016/j.cell.2014.08.032 (2014).

15 Blin, K. et al. antiSMASH 4.0-improvements in chemistry prediction and gene cluster boundary

identification. Nucleic acids research 45, W36-W41, doi:10.1093/nar/gkx319 (2017).

16 Magnusdottir, S. & Thiele, I. Modeling metabolism of the human gut microbiome. Current opinion in biotechnology 51, 90-96, doi:10.1016/j.copbio.2017.12.005 (2018).

17 Sonnenburg, J. L. & Backhed, F. Diet-microbiota interactions as moderators of human metabolism.

Nature 535, 56-64, doi: 10.1038/naturel8846 (2016).

18 Brotherton, C. A. & Balskus, E. P. A prodrug resistance mechanism is involved in colibactin

biosynthesis and cytotoxicity. Journal of the American Chemical Society 135, 3359-3362,

doi: 10.1021/ja312154m (2013).

19 Arafat, E. S., Trimble, J. W., Andersen, R. N., Dass, C. & Desiderio, D. M. Identification of fatty acid amides in human plasma. Life sciences 45, 1679-1687 (1989).

20 Ezzili, C., Otrubova, K. & Boger, D. L. Fatty acid amide signaling molecules. Bioorganic & medicinal chemistry letters 20, 5959-5968, doi:10.1016/j.bmcl.2010.08.048 (2010).

21 Devane, W. A. et al. Isolation and structure of a brain constituent that binds to the cannabinoid receptor.

Science 258, 1946-1949 (1992).

22 Eisenstein, T. K., Meissler, J. J., Wilson, Q., Gaughan, J. P. & Adler, M. W. Anandamide and Delta9- tetrahydrocannabinol directly inhibit cells of the immune system via CB2 receptors. Journal of neuroimmunology 189, 17-22, doi:10.1016/j.jneuroim.2007.06.001 (2007).

23 Osei-Hyiaman, D. et al. Endocannabinoid activation at hepatic CB1 receptors stimulates fatty acid synthesis and contributes to diet-induced obesity. The Journal of clinical investigation 115, 1298-1305, doklO.l 172/JCI23057 (2005).

24 Caterina, M. J. et al. The capsaicin receptor: a heat-activated ion channel in the pain pathway. Nature 389, 816-824, doi: 10.1038/39807 (1997).

25 Chu, C. J. et al. N-oleoyldopamine, a novel endogenous capsaicin-like lipid that produces hyperalgesia.

The Journal of biological chemistry 278, 13633-13639, doi:10.1074/jbc.M211231200 (2003). 26 Przegalinski, E., Filip, M., Zajac, D. & Pokorski, M. N-oleoyl-dopamine increases locomotor activity in the rat. International journal of immunopathology and pharmacology 19, 897-904,

doi : 10.1177/039463200601900419 (2006) .

27 Chu, Z. L. et al. N-oleoyldopamine enhances glucose homeostasis through the activation of GPR119.

Molecular endocrinology 24, 161-170, doi: 10.1210/me.2009-0239 (2010).

28 Ross, H. R., Gilmore, A. J. & Connor, M. Inhibition of human recombinant T-type calcium channels by the endocannabinoid N-arachidonoyl dopamine. British journal of pharmacology 156, 740-750, doi: 10.1111/j .1476-5381 2008.00072.x (2009).

29 Sergeeva, O. A. et al. N-oleoyldopamine modulates activity of midbrain dopaminergic neurons through multiple mechanisms. Neuropharmacology 119, 111-122, doi:10.1016/j.neuropharm.2017.04.011 (2017).

30 Raboune, S. et al. Novel endogenous N-acyl amides activate TRPV1-4 receptors, BV-2 microglia, and are regulated in brain in an acute model of inflammation. Frontiers in cellular neuroscience 8, 195, doi: 10.3389/fncel.2014.00195 (2014).

31 Huang, S. M. et al. Identification of a new class of molecules, the arachidonyl amino acids, and

characterization of one member that inhibits pain. The Journal of biological chemistry 276, 42639-42644, doi: 10.1074/jbc.M107351200 (2001).

32 Oh, D. Y. et al. Identification of farnesyl pyrophosphate and N-arachidonylglycine as endogenous ligands for GPR92. The Journal of biological chemistry 283, 21054-21064, doi: 10.1074/jbc.M708908200 (2008).

33 Sasso, O. et al. Endogenous N-acyl taurines regulate skin wound healing. Proceedings of the National Academy of Sciences of the United States of America 113, E4397-4406, doi: 10.1073/pnas.1605578113 (2016).

34 Hannedouche, S. & Roy, M. Ligand for G-protein coupled receptor GPR72 and uses thereof. United States patent (2008).

35 Milman, G. et al. N-arachidonoyl L-serine, an endocannabinoid-like brain constituent with vasodilatory properties. Proceedings of the National Academy of Sciences of the United States of America 103, 2428- 2433, doi: 10.1073/pnas.0510676103 (2006).

36 Zhang, X., Maor, Y., Wang, J. F., Kunos, G. & Groopman, J. E. Endocannabinoid-like N-arachidonoyl serine is a novel pro-angiogenic mediator. British journal of pharmacology 160, 1583-1594,

doklO.l 111/j.1476-5381.2010.00841.x (2010).

37 Camilleri, M. Review article: tegaserod. Aliment Pharmacol Ther 15, 277-289, doi:10.1046/j.l365- 2036.2001.00925.x (2001).

38 Farrell, E. K. & Merkler, D. J. Biosynthesis, degradation and pharmacological importance of the fatty acid amides. Drug discovery today 13, 558-568, doi: 10.1016/j.drudis.2008.02.006 (2008).

39 Sussmuth, R. D. & Mainz, A. Nonribosomal Peptide Synthesis-Principles and Prospects. Angewandte Chemie 56, 3770-3821, doi: 10.1002/anie.201609079 (2017).

40 Raymond, K. N., Dertz, E. A. & Kim, S. S. Enterobactin: an archetype for microbial iron transport.

Proceedings of the National Academy of Sciences of the United States of America 100, 3584-3588, doi: 10.1073/pnas.0630018100 (2003).

41 Schneditz, G. et al. Enterotoxicity of a nonribosomal peptide causes antibiotic-associated colitis.

Proceedings of the National Academy of Sciences of the United States of America 111, 13181-13186, doi: 10.1073/pnas.1403274111 (2014).

42 Fischbach, M. A. & Walsh, C. T. Assembly-line enzymology for polyketide and nonribosomal Peptide antibiotics: logic, machinery, and mechanisms. Chemical reviews 106, 3468-3496, doi: 10.1021/cr0503097 (2006). 43 Roche, E. D. & Walsh, C. T. Dissection of the EntF condensation domain boundary and active site residues in nonribosomal peptide synthesis. Biochemistry 42, 1334-1344, doi: 10.1021/bi026867m (2003).

44 Mori, S. et al. Activation and Loading of the Starter Unit during Thiocoraline Biosynthesis. Biochemistry

56, 4457-4467, doi:10.1021/acs.biochem.7b00661 (2017).

45 Weber, T. et al. antiSMASH 3.0-a comprehensive resource for the genome mining of biosynthetic gene clusters. Nucleic acids research 43, W237-243, doi:10.1093/nar/gkv437 (2015).

46 Human Microbiome Project, C. Structure, function and diversity of the healthy human microbiome.

Nature 486, 207-214, doi: 10.1038/naturel l234 (2012).

47 Franzosa, E. A. et al. Relating the metatranscriptome and metagenome of the human gut. Proceedings of the National Academy of Sciences of the United States of America 111, E2329-2338,

doi:10.1073/pnas.1319284111 (2014).

48 Navarro-Munoz, J. et al. A computational framework for systematic exploration of biosynthetic diversity from large-scale genomic data. bioRxiv (2018).

49 Mazmanian, S. K., Round, J. L. & Kasper, D. L. A microbial symbiosis factor prevents intestinal

inflammatory disease. Nature 453, 620-625, doi: 10.1038/nature07008 (2008).

50 Almeida, A. et al. A new genomic blueprint of the human gut microbiota. Nature 568, 499-504,

doi:10.1038/s41586-019-0965-l (2019).

51 Nayfach, S., Fischbach, M. A. & Pollard, K. S. MetaQuery: a web server for rapid annotation and

quantitative analysis of specific genes in the human gut microbiome. Bioinformatics 31, 3368-3370, doi: 10.1093/bioinformatics/btv382 (2015).

52 Rausch, C., Hoof, L, Weber, T., Wohlleben, W. & Huson, D. H. Phylogenetic analysis of condensation domains in NRPS sheds light on their functional evolution. BMC Evol Biol 7, 78, doklO.l 186/1471-2148-7 - 78 (2007).

53 Medema, M. H. et al. antiSMASH: rapid identification, annotation and analysis of secondary metabolite biosynthesis gene clusters in bacterial and fungal genome sequences. Nucleic acids research 39, W339-346, doi: 10.1093/nar/gkr466 (2011).

54 Keating, T. A., Marshall, C. G. & Walsh, C. T. Vibriobactin biosynthesis in Vibrio cholerae: VibH is an amide synthase homologous to nonribosomal peptide synthetase condensation domains. Biochemistry 39, 15513-15521 (2000).

55 Pfeifer, B. A., Admiraal, S. J., Gramajo, H., Cane, D. E. & Khosla, C. Biosynthesis of complex

polyketides in a metabolically engineered strain of E. coli. Science 291, 1790-1792,

doi: 10.1126/science.1058092 (2001 ).

56 Pugin, B. et al. A wide diversity of bacteria from the human gut produces and degrades biogenic amines.

Microbial ecology in health and disease 28, 1353881, doi:10.1080/16512235.2017.1353881 (2017).

57 Hansen, D. B., Bumpus, S. B., Aron, Z. D., Kelleher, N. L. & Walsh, C. T. The loading module of mycosubtilin: an adenylation domain with fatty acid selectivity. Journal of the American Chemical Society 129, 6366-6367, doi:10.1021/ja070890j (2007).

58 Stachelhaus, T., Mootz, H. D. & Marahiel, M. A. The specificity-conferring code of adenylation domains in nonribosomal peptide synthetases. Chemistry & biology 6, 493-505, doi:10.1016/S1074-5521(99)80082-9 (1999).

59 Stachelhaus, T., Mootz, H. D., Bergendahl, V. & Marahiel, M. A. Peptide bond formation in

nonribosomal peptide biosynthesis. Catalytic role of the condensation domain. The Journal of biological chemistry 273, 22773-22781 (1998). 60 Brady, S. F. & Clardy, J. Palmitoylputrescine, an antibiotic isolated from the heterologous expression of DNA extracted from bromeliad tank water. Journal of natural products 67, 1283-1286,

doi : 10.1021 /np0499766 (2004).

61 Frolov, A., Cho, T. H., Billheimer, J. T. & Schroeder, F. Sterol carrier protein-2, a new fatty acyl

coenzyme A-binding protein. The Journal of biological chemistry 271, 31878-31884 (1996).

62 McKinney, M. K. & Cravatt, B. F. Structure and function of fatty acid amide hydrolase. Annual review of biochemistry 74, 411-432, doi: 10.1146/annurev.biochem.74.082803.133450 (2005).

63 Quadri, L. E. et al. Characterization of Sfp, a Bacillus subtilis phosphopantetheinyl transferase for peptidyl carrier protein domains in peptide synthetases. Biochemistry 37, 1585-1595, doi:10.1021/bi9719861 (1998).

64 Hill, M. J. Microbial Metabolism In The Digestive Tract. (CRC Press, 2018).

65 Abdelmagid, S. A. et al. Comprehensive profiling of plasma fatty acid concentrations in young healthy Canadian adults. PloS one 10, eOl 16195, doi:10.1371/journal.pone.0116195 (2015).

66 Chan, M., Himes, R. H. & Akagi, J. M. Fatty acid composition of thermophilic, mesophilic, and

psychrophilic Clostridia. Journal of bacteriology 106, 876-881 (1971).

67 Vernocchi, P., Del Chierico, F. & Putignani, L. Gut Microbiota Profiling: Metabolomics Based

Approach to Unravel Compounds Affecting Human Health. Frontiers in microbiology 7, 1144, doi:10.3389/fmicb.2016.01144 (2016).

68 Lyte, M. & Freestone, P. P. E. Microbial Endocrinology: Interkingdom Signaling in Infectious Disease and Health. (Springer New York, 2010).

OTHER EMBODIMENTS

All of the features disclosed in this specification may be combined in any combination. Each feature disclosed in this specification may be replaced by an alternative feature serving the same, equivalent, or similar purpose. Thus, unless expressly stated otherwise, each feature disclosed is only an example of a generic series of equivalent or similar features.

From the above description, one skilled in the art can easily ascertain the essential characteristics of the present invention, and without departing from the spirit and scope thereof, can make various changes and modifications of the invention to adapt it to various usages and conditions. Thus, other embodiments are also within the claims.

EQUIVALENTS

While several inventive embodiments have been described and illustrated herein, those of ordinary skill in the art will readily envision a variety of other means and/or structures for performing the function and/or obtaining the results and/or one or more of the advantages described herein, and each of such variations and/or modifications is deemed to be within the scope of the inventive embodiments described herein. More generally, those skilled in the art will readily appreciate that all parameters, dimensions, materials, and configurations described herein are meant to be exemplary and that the actual parameters, dimensions, materials, and/or configurations will depend upon the specific application or applications for which the inventive teachings is/are used. Those skilled in the art will recognize, or be able to ascertain using no more than routine experimentation, many equivalents to the specific inventive embodiments described herein. It is, therefore, to be understood that the foregoing embodiments are presented by way of example only and that, within the scope of the appended claims and equivalents thereto, inventive embodiments may be practiced otherwise than as specifically described and claimed. Inventive embodiments of the present disclosure are directed to each individual feature, system, article, material, kit, and/or method described herein. In addition, any combination of two or more such features, systems, articles, materials, kits, and/or methods, if such features, systems, articles, materials, kits, and/or methods are not mutually inconsistent, is included within the inventive scope of the present disclosure.

All definitions, as defined and used herein, should be understood to control over dictionary definitions, definitions in documents incorporated by reference, and/or ordinary meanings of the defined terms.

All references, patents and patent applications disclosed herein are incorporated by reference with respect to the subject matter for which each is cited, which in some cases may encompass the entirety of the document.

The indefinite articles“a” and“an,” as used herein in the specification and in the claims, unless clearly indicated to the contrary, should be understood to mean“at least one.”

The phrase“and/or,” as used herein in the specification and in the claims, should be understood to mean“either or both” of the elements so conjoined, i.e., elements that are conjunctively present in some cases and disjunctively present in other cases. Multiple elements listed with“and/or” should be construed in the same fashion, i.e.,“one or more” of the elements so conjoined. Other elements may optionally be present other than the elements specifically identified by the“and/or” clause, whether related or unrelated to those elements specifically identified. Thus, as a non-limiting example, a reference to“A and/or B”, when used in conjunction with open-ended language such as“comprising” can refer, in one embodiment, to A only (optionally including elements other than B); in another embodiment, to B only (optionally including elements other than A); in yet another embodiment, to both A and B (optionally including other elements); etc.

As used herein in the specification and in the claims,“or” should be understood to have the same meaning as“and/or” as defined above. For example, when separating items in a list, “or” or“and/or” shall be interpreted as being inclusive, i.e., the inclusion of at least one, but also including more than one, of a number or list of elements, and, optionally, additional unlisted items. Only terms clearly indicated to the contrary, such as“only one of’ or“exactly one of,” or, when used in the claims,“consisting of,” will refer to the inclusion of exactly one element of a number or list of elements. In general, the term“or” as used herein shall only be interpreted as indicating exclusive alternatives (i.e.“one or the other but not both”) when preceded by terms of exclusivity, such as“either,”“one of,”“only one of,” or“exactly one of.” “Consisting essentially of,” when used in the claims, shall have its ordinary meaning as used in the field of patent law.

As used herein in the specification and in the claims, the phrase“at least one,” in reference to a list of one or more elements, should be understood to mean at least one element selected from any one or more of the elements in the list of elements, but not necessarily including at least one of each and every element specifically listed within the list of elements and not excluding any combinations of elements in the list of elements. This definition also allows that elements may optionally be present other than the elements specifically identified within the list of elements to which the phrase“at least one” refers, whether related or unrelated to those elements specifically identified. Thus, as a non-limiting example,“at least one of A and B” (or, equivalently,“at least one of A or B,” or, equivalently“at least one of A and/or B”) can refer, in one embodiment, to at least one, optionally including more than one,

A, with no B present (and optionally including elements other than B); in another embodiment, to at least one, optionally including more than one, B, with no A present (and optionally including elements other than A); in yet another embodiment, to at least one, optionally including more than one, A, and at least one, optionally including more than one, B (and optionally including other elements); etc.

It should also be understood that, unless clearly indicated to the contrary, in any methods claimed herein that include more than one step or act, the order of the steps or acts of the method is not necessarily limited to the order in which the steps or acts of the method are recited.