Methods for making a genetically engineered organism exclusively communicate to or rely on another engineered organism

Abstract

The present invention provides a method of enabling growth, survival, or a biological function of a recombinant utilizer organism. The growth, survival or biological activity of the recombinant utilizer organism depends on a non-standard amino acid (nsAA). 5 The method comprises: (a) providing a system comprising the recombinant utilizer organism and a recombinant producer organism, wherein the system lacks the nsAA and is not supplemented externally with the nsAA; (b) producing the nsAA by the recombinant producer organism into the system; and (c) incubating the recombinant utilizer organism in the system after step (b).

Description

[0001] METHODS FOR MAKING A GENETICALLY ENGINEERED ORGANISM EXCLUSIVELY COMMUNICATE TO OR RELY ON ANOTHER ENGINEERED ORGANISM

[0002] CROSS-REFERENCE TO RELATED APPLICATION

[0003] This application claims priority to U.S. Provisional Application No. 63 / 607,425, filed December 7, 2023, and U.S. Provisional Application No. 63 / 664,241, filed lune 26, 2024, the contents of each of which are incorporated herein by reference in their entireties for all purposes.

[0004] REFERENCE TO U.S. GOVERNMENT SUPPORT

[0005] This invention was made with government support under grant number FF-NIA21- 0000000060 awarded by the Foundation for Food and Agriculture Research (FFAR) and grant number D24AC00011-00 awarded by the Defense Advanced Research Projects Agency (DARPA). The United States has certain rights in the invention.

[0006] FIELD OF THE INVENTION

[0007] The invention relates to the design of an exclusive reliance - also referred to as an orthogonal and obligate commensalism - through pairing the biosynthesis of an orthogonal non-standard amino acid (nsAA) by a producer organism and synthetic auxotrophy in a utilizer organism within a system and uses thereof.

[0008] BACKGROUND OF THE INVENTION

[0009] Synthetic biology offers the promise of addressing issues in agriculture, environmental health, and public health by direct release of genetically engineered microbes into target environments. However, a grand challenge that has limited the use of engineered microbes in the environment is the design of effective safeguards that provide strict spatiotemporal control over the survival of engineered organisms. Because most naturally occurring auxotrophy can be overcome through microbial cross-feeding, more sophisticated biological containment strategies are required to be effective in open systems. The last decade has featured the demonstration of several innovative intrinsic biocontainment strategies, including genetic circuits that trigger cell death in response to specific molecules or environmental conditions as well as synthetic auxotrophy, where the functions of essential proteins are dependent upon availability of a synthetic nutrient. Synthetic auxotrophy can be created by engineering the translation and function of an essential protein to depend upon the incorporation of a non-standard amino acid (nsAA) within its sequence. It has shown promise as an intrinsic biocontainment strategy given the low rates of escape from biocontainment observed, which includes bacterial strains that have not exhibited detectable escape even after months of continuous culturing. These escape frequencies are well below the NIH standard of 10'8escapees per colony forming unit (escapees / CFU), though environmental release may necessitate a different standard.

[0010] While the demonstration that synthetic auxotrophs remain biocontained in monoculture is promising, further innovation is required for the implementation of biocontainment in the environment. Given that many applications require survival of engineered cells for timescales longer than hours, target environments must be designed such that they are permissive for growth of a synthetic auxotroph, preferably without repeated human interv ention (Fig. 1 A).

[0011] There remains a need for a method to enable survival, growth, or a biological function of a synthetic auxotroph in a target environment.

[0012] SUMMARY OF THE INVENTION

[0013] The present invention relates to enablement of grow th, survival or a biological function of a utilizer recombinant organism, which growth, survival or biological function depend on a non-standard amino acid (nsAA). The inventors have surprisingly discovered a system of comprising the recombinant utilizer organism and a recombinant producer organism producing the nsAA without supplementing the system externally with the nsAA.

[0014] The present invention provides a method of enabling growth of a recombinant utilizer organism. The growth of the recombinant utilizer organism depends on a nonstandard amino acid (nsAA). The method comprises: (a) providing a system comprising the recombinant utilizer organism and a recombinant producer organism, wherein the system lacks the nsAA and is not supplemented externally with the nsAA; (b) producing the nsAA by the recombinant producer organism into the system; and (c) incubating the recombinant utilizer organism in the system after step (b). whereby the recombinant utilizer organism grows in the system. The nsAA may be essential to survival of the recombinant utilizer organism.

[0015] The present invention also provides a method of enabling survival of a recombinant utilizer organism. The survival of the recombinant utilizer organism depends on a non- standard amino acid (nsAA). The method comprises: (a) providing a system comprising the recombinant utilizer organism and a recombinant producer organism, wherein the system lacks the nsAA and is not supplemented externally with the nsAA; (b) producing the nsAA by the recombinant producer organism into the system: and (c) incubating the recombinant utilizer organism in the system after step (b), whereby the recombinant utilizer organism survives in the system. The recombinant utilizer organism may grow in the system.

[0016] The present invention further comprises a method of enabling a biological function of the recombinant utilizer organism. The biological function of the recombinant utilizer organism depends on a non-standard amino acid (nsAA). The method comprises: (a) providing a system comprising the recombinant utilizer organism and a recombinant producer organism, wherein the system lacks the nsAA and is not supplemented externally with the nsAA; (b) producing the nsAA by the recombinant producer organism into the system; and (c) incubating the recombinant utilizer organism in the system after step (b), whereby the recombinant utilizer organism generates the biological function in the system. The recombinant utilizer organism may grow in the system. The nsAA may be essential to survival of the recombinant utilizer organism.

[0017] Each method of the present invention may further comprise incorporating the nsAA produced by the recombinant producer organism into a target protein in the recombinant utilizer organism. The target protein may be an enzyme. The target protein may be essential to a biological activity’ of the recombinant utilizer organism.

[0018] Each method of the present invention may further comprise adjusting the production of the nsAA by the recombinant producer organism, whereby the growth of the recombinant utilizer organism is controlled.

[0019] The recombinant producer organism and the recombinant utilizer organism may be in direct contact.

[0020] The nsAA may be present in the system at a concentration from 1 nM to 100 mM.

[0021] The nsAA may be O-methyl-tyrosine (OMeTyr).

[0022] The recombinant producer organism may be a bacterium.

[0023] The recombinant producer organism may be a plant. The plant may b Arabidopsis thaliana.

[0024] The recombinant utilizer organism may be a rhizobacterium. The rhizobacterium may be Bacillus subtilis. The recombinant utilizer organism may be a microorganism.

[0025] Each method of the present invention may further comprise one or more sets of microbes within a target microbiome selected from the group consisting of soil microbiomes, gut microbiomes, oral microbiomes, vaginal microbiomes, and skin microbiomes.

[0026] The system may further comprise an additional recombinant utilizer organism dependent on the recombinant producer organism.

[0027] BRIEF DESCRIPTION OF THE DRAWINGS

[0028] Figs. 1 A-F show characterization of the individual components envisioned to support an obligate commensal based on the non-standard amino acid OMeTyr. A, Illustration of the concept of an organism engineered to be an orthogonal and obligate commensal (blue microbe) that strictly relies on the inclusion of another engineered organism (yellow microbe) within a community. B, Schematic for one proposed strategy to create a form of commensalism mediated by the biosynthesis of a non-standard amino acid (nsAA) by one organism followed by the utilization of the nsAA by another organism that requires it. Included are design criteria for suitable nsAAs. C, Depiction of the biosynthesis of the nsAA chosen for this study, (9-methyl-L-tyrosine (OMeTyr), and the associated genetic construct that expresses mfnG and aroG* genes within a synthetic operon. D, Titers of OMeTyr as a function of time produced by an engineered E. coll strain that co-expresses mfnG and aroG* genes under a constitutive promoter in LB media or M9 minimal media with glucose. E, Depiction of the generalized orthogonal translation machinery used to achieve site-specific incorporation of OMeTyr in E. coir. An engineered aminoacyl tRNA synthetase-tRNA pair (o-AARS / o-tRNA) derived fromM j annas chii coupled with a fluorescent reporter (GFP_TAG) that requires suppression of an in-frame amber stop codon for full-length translation. F, Fluorescence (ex: 485 nm, em: 525 nm) normalized by ODeoo resulting from the incorporation of increasing concentrations of OMeTyr supplied to cultures of the genomically recoded RR3 (DE3) E. coli strain that harbors the NapARS synthetase in an aromatic amino acid dropout version of MOPS EZ rich media with glucose 9 h after inoculation. Sample sizes are n=3 using biological replicates (samples originated from the same glycerol stock). Data shown are mean ± standard deviation.

[0029] Figs. 2A-C show incorporation of biosynthesized OMeTyr during a co-culture across two distinct E. coli strains. A, Predicted schema of fluorescent protein production from co-cultures that contain either an OMeTyr producer strain (P) and an OMeTyr utilizer strain (U) or an OMeTyr non-producer strain (NP) and an OMeTyr utilizer strain. The producer strain constitutively expresses mfriG and aroG*. whereas the non-producer strain does not contain OMeTyr biosynthesis genes. The utilizer strain harbors the OTS and reporting system resulting in GFP production when supplied with its nsAA. B, Fluorescence from co-cultures between the RR3 (DE3) strain harboring an operon containing the OTS, including the NapARS synthetase, and GFP reporter and E. coli strains both with and without the ability to biosynthesize OMeTyr. Strains were inoculated at varied ODeoo ratios in MOPS EZ Rich media lacking aromatic amino acids. C, Mass spectra from intact protein MS of Ub-GFP with OMeTyr incorporation from metabolic synthesis in RR3 (DE3) containing an operon with the OTS, including the NapARS synthetase, and GFP reporter. Sample sizes are n=3 using biological replicates and data shown are mean ± standard deviation except for mass spectra results w ich represents n=l. Fluorescent outputs collected at (ex: 485 nm, em: 525 nm).

[0030] Figs. 3A-F show- generation and characterization of synthetic auxotrophs dependent on OMeTyr. A, Depiction of the previously reported B-adk.d6 synthetic auxotroph. B, Kinetic growth dynamics of B-adk.d6 in LB media without addition of an nsAA, with 0.1 mM BipA, or with 0.5 mM OMeTyr for 48 h at 34 °C. C, Escape frequencies of B-adk.d6 after 24 and 48 h collected by counting the colonies on each strains’ permissive and non- permissive media. Escapees / CFU were calculated as colonies observed on non-permissive media per average viable CFU plated for each strain. D, Depiction of the N-adk.d6 synthetic auxotroph. E, Kinetic grow4h dynamics of N-adk.d6 synthetic auxotroph grown in LB media with and without OMeTyr for 48 h at 34 °C. F, Escape frequencies of N-adk.d6, N- adk.d7, and N.dl-adk.d6 over the course of 14 days collected by counting colonies on permissive and non-permissive media. N-adk.d6 escape frequency only documented through Day 2 due to an inability to distinguish colonies from increased growth. Sample sizes are n=3 using biological replicates and data shown are mean ± standard deviation.

[0031] Figs. 4A-H demonstrate obligate commensalism mediated by OMeTyr. A, Predicted outcome of N.dl-adk.d6 survival based on various culture conditions. B, Co-culture results depicting n=3 replicates for each condition listed. 1strow : N.dl-adk.d6 monoculture grown in non-permissive media. 2ndrow-: N.dl-adk.d6 co-cultured with anon-producing RR3 (DE3) strain of E. coli in non-permissive media. 3rdrow: N.dl-adk.d6 co-cultured with the OMeTyr producing RR3 (DE3) strain of E. coli in non-permissive media. Images shown represent colonies 72 h after plating on solid media selective for N.dl-adk.d6. C, CFU / mL values of N-adk.d6 and N.dl-adk.d6 calculated from 10-fold serial dilutions of the samples grown on solid, permissive media for 96 h and 48 h respectively following 24 h co-cultures. Values determined through colony counting from 10 pL plated media samples. D, Experimental CFU / mL values calculated from co-cultures of each listed producer: utilizer ratio plated on N-adk.d6 permissive media. Values determined through colony counting from 10 pL plated media samples. Sample sizes are n=3 using biological replicates and data shown are mean ± standard deviation. E, Depiction of the DEP synthetic auxotroph, dependent on BipA (star), with the OMeTyr production machinery. F, Co-culture results depicting n=3 replicates for each condition listed. 1strow: N-adk.d6 monoculture in media lacking nsAAs. 2ndrow: N-adk.d6 monoculture in media containing 5 pM BipA externally supplied. 3rdrow: N-adk.d6 co-cultured with DEP capable of producing OMeTyr with 5 pM BipA externally supplied. Images shown represent colonies 72 h after plating on solid media selective for N-adk.d6. G, Depiction of mutualism between the histidine auxotroph, dependent on histidine (triangle), with OMeTyr production machinery and N.dl-adk.d6. H. Kinetic growth dynamics of the mutualism co-cultures with N.dl-adk.d6 and the histidine auxotroph OMeTyr (non-)producer in M9 minimal media without external addition of OMeTyr or histidine for 48 h at 34 °C. Sample sizes are n=3 using biological replicates and data shown are mean ± standard deviation, except for images in B and F which show individual replicates.

[0032] Figs. 5A-D demonstrate the orthogonality of the obligate commensalism within a microbial consortium consisting of soil isolates. A, Schema of desired N.dl-adk.d6 growth outcomes when grown in the absence and presence of OMeTyr producing E. coli when select members of a model soil consortium are also present. B. OMeTyr concentrations after 24 h growth of the soil consortium without externally supplied OMeTyr (left), with 0.5 mM externally supplied OMeTyr (center), and with a 1 :1 inoculation ratio of the OMeTyr producing E. coli strain at the start of the culture (right). C, Growth results of N.dl-adk.d6- strep on solid, permissive media following 24 h growth in non-permissive LB in a consortium containing N.dl-adk.d6-strep, S. mahophiha. C. pusilium, H. frsingense, P. putida, and a non-producing strain of E. coli (top) or an OMeTyr producing strain of E. coli (bottom). Images shown represent colonies 72 h after plating on solid media selective for N.dl-adk.d6-strep. Sample sizes are n=3 using biological replicates each plated separately. D, Values of CFU / mL for N.dl-adk.d6-strep 24 h after plating cultures that were grown in non-permissive liquid media for 24 h in a consortium including the soil microbes and a nonproducing (left) or OMeTyr producing (right) E. coli strain. Sample sizes are n=3 using biological replicates and data shown are mean ± standard deviation.

[0033] Figs. 6A-C show' genetic construct design and results of OMY incorporation assay in liquid media conditions. (A) Illustration of the two cassettes for heterologous expression that were genomically integrated at the lacA and amyE loci. These enable expression of an orthogonal translation system and a green fluorescent reporter whose full-length translation should depend on provision of OMY. (B) Measurements of fluorescence normalized to optical density at 600 nm (arbitrary units) as a function of OMY concentration and liquid media condition for the engineered system in the domesticated PY79 host. S750 is a defined minimal media used for B. subtilis culturing, and 1 / 2X MS + glut is a slight modification of a defined media used for plant culturing. Cultures w ere grown at 25 °C with shaking incubation. (C) Measurements of fluorescence normalized to optical density at 600 nm (arbitrary units) as a function of OMY concentration and liquid media condition for the engineered system in the undomesticated UDI 022 host.

[0034] Figs. 7A-E show evaluation of OMY incorporation within sterilized soil. (A) Workflow to allow incubation of engineered B. subtilis strains with or without OMY in sterilized soil, followed by extraction and filtration to recover cells from soil for analysis byflow cytometry. For each replicate, we used 1 gram of soil and added 1 mL of PBS. In the cases in which we added OMY, it corresponds to 500 pM OMY (in PBS) or 500 nmol / g OMY / soil. (B) Flow' cytometric output, specifically forward and side scatter, when analyzing pure cultures of B. subtilis (in this case, PY79) in PBS media (without having seen soil). This positive control serves to identify where to set gates (in red) for analysis primarily of cells rather than soil debris. (C) Flow cytometric output observed when analyzing extracted and filtered cultures of B. subtilis that had been incubating in soil. (D) Histograms of extracted engineered B. subtilis UDI 022 cell populations that had been incubating in the soil for three days, either with or without 500 pM OMY. The top distribution (in green) corresponds to the positive control strain population which harbors an mNeonGreen reporter that contains no TAG codons and is thus reflective of wild-type reporter protein expression level. The middle distribution (in purple) corresponds to the population of cells that harbor the mNeonGreen reporter that contains 1 TAG codon and that was not exposed to OMY. The bottom distribution (in magenta) corresponds to the population of cells that harbor the same 1 TAG codon reporter and that was incubated with OMY. (E) The median fluorescence observed in each population across three replicates and at two different timepoints: 1 day and 3 days after inoculation.

[0035] Fig. 8 shows genetic construct design for candidate single marker synthetic auxotrophs.

[0036] Fig. 9 shows the reduction in viable colony forming units of our engineered synthetic auxotrophic B. subtilis PY79 that contains two nsAA-dependent markers (dxs and metS), after removal of the nsAA OMY by washing, followed by plating under solid-media assay testing conditions of 25 °C and plant minimal media (l / 2x MS supplemented with 1.5% sucrose and 10 mM L-glutamate).

[0037] Fig. 10 visualizes stably transformed Arabidopsis expressing MfnG-sfGFP fusion protein under the control of A) the constitutive UBQ10 promoter and B) the root cap specific BEARSKIN 1 promoter (insets show magnified root tips). Plants were first imaged with a Epson scanner at 600 dpi (left). The same plants were then imaged under a GFP filter at 500 ms exposure (right).

[0038] Figs. 11 A-B show LCMS quantification of OMY production by MfhG-expressing Arabidopsis. A) We sowed seeds onto plastic meshes in individual wells of a 48-well plate for hydroponic growth in 250 pL growth media (IX Murashige-Skoog basal salts, 1% sucrose). Plants were grown at normal growth conditions for 7 days. We pooled media from 10-25 individual seedlings, lyophilized overnight, and extracted with 80% methanol for LCMS analysis (Agilent 6520, Cl 8 column, positive ion mode). B) Titers for exudate- conditioned media for each Arabidopsis line tested for OMY production.

[0039] Fig. 12 shows schematic of the assay we developed to measure exclusive communication from plant to microbe based on nsAA-producing seedlings and different strains of bacteria that are engineered to express the reporter enzyme luciferase. Zone of luminescence signal indicates the zone of influence for the nsAA-producing plant.

[0040] Fig. 13 shows plant-microbe communication results for three different lines of Arabidopsis thaliana (the wild-type Col-0, the ubiquitous expression of MfnG-GFP, and the root cap-specific expression of MfnG-GFP) and three different versions of B. subtilis PY79 (the wild-type PY79 that lacks a gene encoding luciferase, an engineered strain that expresses an nsAA-independent or constitutive luciferase as a positive control for the distribution of microbe, and the engineered strain of interest that expresses an nsAA- dependent luciferase that requires plant-microbe communication for luminescence. Top panels show an overlay of brightfield and chemiluminescence channels, while bottom panels show the chemiluminescence channels alone. All red color shown is false color in place of luminescence signal. Luminescence signal is adjusted to max contrast per image. For this assay, we added 5 mL of luciferase substrate-agarose to each plate and took 10s exposure images. Seedlings were 7 days post-stratification when transferred to the bacteria- agar matrix.

[0041] Figs. 14A-B show plant-microbe communication results where instead of testing PY79 and related derivatives we test UD1022 and related derivatives. No luminescence is observed when using the wild-type UDI 022 (shown in the top-most plate on the right), whereas nsAA-dependent luminescence is seen and particularly strong when the microbes are exposed to the Arabidopsis lines that feature ubiquitous expression. Top panels show an overlay of brightfield and chemiluminescence channels, while bottom panels show the chemiluminescence channels alone. All red color shown is false color in place of luminescence signal. Luminescence signal is adjusted to max contrast per image. Signal intensity is compared across A) different plant lines for the same strain and B) different strains for the same plant line. For this assay, we added 10 mL of luciferase substrate- agarose to each plate and took 2s exposure images. Seedlings were 8 days post-stratification when transferred to the bacteria-agar matrix.

[0042] DETAILED DESCRIPTION OF THE INVENTION

[0043] The present invention relates to an orthogonal and obligate commensalism through pairing the biosynthesis of an orthogonal non-standard amino acid (nsAA) by a producer organism and synthetic auxotrophy in a utilizer organism within a system, for example, a specific environment, and uses thereof. This system may be used to enable survival, growth, and / or a biological function of the utilizer organism in a target environment, for example, a non-sterile environment. The invention is based on the inventors’ surprising discover}' of genetically engineered recombinant nsAA producer organisms and the corresponding genetically engineered recombinant nsAA utilizer organisms. The inventors have established exclusive dependence of a utilizer microbe on a producer microbe even w ithin a model soil consortium as well as the exclusive reliance of nsAA utilizer strains of the rhizobacterium Bacillus subtilis on nsAA producer lines of the model plant species Arabidopsis thaliana.

[0044] The inventors have created an exclusive means of communication between two genetically engineered organisms, specifically to make the biological functions, growth or survival of one or more engineered organisms rely exclusively on another genetically engineered organism. Doing so affords an ability to control the spatial distribution of function and / or survival of an engineered organism when the position of one organism is fixed. The inventors have also created an ability to precisely control the relative abundance of engineered organisms within a broader community. These exclusive reliances are referred to as orthogonal symbioses.

[0045] An exemplary scenario is in the area of biological containment, where a microbe could be engineered for release in the environment, to perform functions such as removing pollutants or protecting a desired crop from stresses. After release without the system of the present invention, an engineered microbe could ordinarily proliferate into non-target environments or associate with weedy plants, with no clear way to remove the microbe from the environment and no targeting of beneficial functions to a particular crop. The present invention provides an avenue by which a microbe engineered to rely exclusively upon another organism could be introduced to an environment alongside another engineered organism whose existence is easier to detect and control, such as a sterile plant. In this manner, the engineered microbe only survives in the zone near the plant. Additionally, when the desired function is complete, the plant can be removed and the engineered microbe would be eliminated from the environment.

[0046] Another exemplary scenario relates to encapsulating one engineered microbe using physical containment strategies while releasing another microbe that is engineered to rely exclusively on the contained microbe. In such a system, which could be applied to the rhizosphere of a target plant without requiring engineering of the plant, a biologically contained microbe would be able to perform desired functions such as root association so long as the particles that contain encapsulated microbe are nearby. Notably, a microbe subject to physical containment alone would not have been able to associate with the root, and a microbe subject to biological containment alone without a producer organism would lack an ability to survive over time without regular human intervention to supply necessary molecules. Finally, yet another set of scenarios where the present technology7provides valuable control is for the engineering of artificial endosymbionts, where one microbe is engineered to reside within the cell of another organism. Such systems could be useful for bestowing upon the host organism a unique ability , such as photosynthesis and CO2 fixation if, for example, the endosymbiont is a cyanobacterium.

[0047] The present invention provides organisms in various community contexts, meaning that more than one organism is grown together. The initial proof of concepts of orthogonal symbiosis presented herein demonstrates exclusive reliance betw een pairs of microbes in microbial co-culture as well as plant-microbe co-culture systems in which there is exclusive communication between plant and rhizobacterium. In each case, the pairs of organisms consist of a producer of a non-standard amino acid (nsAA) and a utilizer of the nsAA in a growth medium. The nsAA producer is a genetically engineered organism that biosynthesizes an nsAA. The nsAA utilizer is a genetically engineered organism that can either rely on the nsAA for the full-length translation of an enzyme that results in an nsAA- dependent function, or the nsAA utilizer is engineered to rely on the nsAA for survival through the intrinsic biological containment technique of synthetic auxotrophy.

[0048] In all cases, the grow th medium is not supplemented externally with the nsAA. The nsAA may be O-methyl-tyrosine (OMeTyr) or any other nsAA that can be site-specifically incorporated within protein sequences and that is not abundant naturally in the target environment or conditions chosen for application of the system.

[0049] The term ‘"orthogonal symbiosis” as used herein refers to exclusive reliance of survival, growth or a biological function of one genetically engineered organism on another genetically engineered organism.

[0050] The terms “recombinant” and “genetically engineered” are used herein interchangeably and refer to an organism that is not naturally occurring.

[0051] The term “organism” as used herein refers to a living individual. The individual may be an animal, plant, or microorganism. A microorganism may be a bacterium, virus, or fungus.

[0052] The term “growth rate” as used herein refers to a percentage of change in the number of cells in or of an organism over a specific time period, for example, about 1, 2, 3. 4, 5, 6, 12, 24, 36, 48, 72 or 96 hours, 1, 2, 3, 4, 5, 6, 7, 14, 21 or 28 days, or 1, 2, 3, 6, 9 or 12 months.

[0053] The term "survival rate’7as used herein refers to a percentage of cells in or of an organism that remain alive after a specific time period, for example, about 1 , 2, 3, 4, 5, 6, 12, 24, 36, 48, 72 or 96 hours, 1, 2, 3, 4, 5, 6, 7, 14, 21 or 28 days, or 1, 2, 3, 6, 9 or 12 months.

[0054] The term "biological function” as used herein refers to an activity of an organism. Examples of biological functions include nitrogen fixation, phosphate solubilization, plant growth promotion, environmental remediation, plastic deconstruction, chemical sensing, and pest control.

[0055] The term “‘microbiome” as used herein refers to a set of two or more microorganisms, for example, bacteria, viruses, fungi, and protozoa, that live in a specific environment.

[0056] The present invention provides a method of enabling growth of a recombinant utilizer organism. The growth of the recombinant utilizer organism depends on a nonstandard amino acid (nsAA). The growth enabling method comprises providing a system. The system comprises the recombinant utilizer organism and a recombinant producer organism. The system lacks the nsAA and is not supplemented externally with the nsAA. The growth enabling method also comprises producing the nsAA by the recombinant producer organism into the system. The growth enabling method further comprises incubating the recombinant utilizer organism in the system after the production of nsAA. As a result, the recombinant utilizer organism grows in the system.

[0057] Where growth of a recombinant utilizer organism depends on an nsAA, the recombinant utilizer organism may have a growth rate of less than about 10%, 5%, 1%, 0. 1% or 0.01% in the absence of the nsAA. or the growth rate of the recombinant utilizer organism may be reduced by at least about 90%, 95%, 99%, 99.9% or 99.99% within a predetermined time, for example, about 0.5, 1, 2, 3, 4, 5, 6, 12, 24, 48, 72 or 96 hours after the nsAA is removed from the recombinant utilizer organism.

[0058] According to the growth enabling method, the nsAA may be essential to survival of the recombinant utilizer organism. The recombinant utilizer organism may have a survival rate of less than about 10%, 5%, 1%, 0.1% or 0.01% in the absence of the nsAA, or the survival rate of the recombinant utilizer organism may be reduced by at least about 90%, 95%, 99%, 99.9% or 99.99% within a predetermined time, for example, about 0.5, 1, 2, 3, 4, 5, 6, 12, 24, 48, 72 or 96 hours after the nsAA is removed from the recombinant utilizer organism.

[0059] The growth enabling method may further comprise incorporating the nsAA produced by the recombinant producer organism into a target protein in the recombinant utilizer organism. The target protein may be an enzyme. Exemplary enzy mes include luciferase, nitrogenase, phytase, acid phosphatase, surfactin synthetase, phosphopantetheinyl transferase, nitroreductase, polyethylene terephthalate esterase (PETase), cutinase, and allosteric transcription factors. Alternatively, the target protein may be a non-enzymatic protein, such as delta endotoxin (insecticide), green fluorescent protein, other fluorescent proteins, or other reporter proteins. The target protein may be essential to a biological activity of the recombinant utilizer organism. The activity of the biological function of the recombinant utilizer organism may be reduced by at least about 90%, 95%, 99%, 99.9% or 99.99% within a predetermined time, for example, about 0.5, 1, 2, 3, 4, 5, 6, 12, 24, 48, 72 or 96 hours after the nsAA is removed from the recombinant utilizer organism, or the increased by at least about 10. 100, 1.000 or 10,000 folds after the nsAA is incorporated into the target protein provided to the recombinant utilizer organism.

[0060] The grow th enabling method may further comprise adjusting the production of the nsAA by the recombinant producer organism such that the growth of the recombinant utilizer organism is controlled.

[0061] According to the growth enabling method, the recombinant producer organism and the recombinant utilizer organism may be in direct contact. The recombinant producer organism and the recombinant utilizer organism may be in close proximity to each other, for example, within a distance of about 0, 0.00001, 0.0001, 0.001, 0.01, 0.1, 1, 10 or 100 cm, or about 0-100, 0.00001-100, 0.0001-100. 0.001-100, 0.01-100, 0.1-100. 1-100, 10-100. 0.00001 -10, 0.0001 -10, 0.001-10, 0.01-10, 0.1 -10, 1-10, 0.00001-1, 0.0001 -1 , 0.001-1 , 0.01-1, 0.1-1, 0.00001-0.1, 0.0001-0.1, 0.001-0.1, 0.01-0.1, 0.00001-0.01, 0.0001-0.01, or 0.00001-0.0001 cm.

[0062] According to the growth enabling method, the nsAA may be present in the system at a desirable concentration, which may vary depending on what is required by the nsAA utilizer organism and / or conceivable to attain via biosynthesis. The nsAA concentration in the system may be the range of about 0.01-1,000,000 nM, 0.1-100,000 nM, 1-100,000 nM, 10-10.000 nM or 100-1,000 nM.

[0063] According to the grow th enabling method, the nsAA may be any of the nsAAs known for genetic code expansion using orthogonal translation systems, such as analogs of tyrosine / phenylalanine, tryptophan, leucine, lysine, histidine, and others. The nsAA may be O-methyl-tyrosine (OMeTyr), p-iodo-L-phenylalanine, p-bromo-L-phenylalanine, p-chloro- L-phenylalanine, p-fluoro-L-phenylalanine, p-propargyloxyphenylalaninc. p-propargyl- phenylalanine, L-3-(2-naphthyl)alanine, 3-methyl-phenylalanine, p-methyl-phenylalanine, p-azido-L-phenylalanine. p-acetyl-L-phenylalanine, p-carboxy-L-phenylalanine, p-formyl- phenylalanine, p-hydroxymethyl-L-phenylalanine, (2S)-2-amino-3-[4-(2- hydroxyethoxycarbonyl)phenyl]propanoic acid, 3,4-dihydroxyphenylalanine, sulfotyrosine, p-amino-L-phenylalanine, p-nitro-L-phenylalanine, p-ethylthiocarbonyl-L-phenylalanine, p- (3-oxobutanoyl)-L-phenylalanine, 1,5-dansyl-alanine, 7-amino-coumarin-alanine. 7- hydroxy-coumarin-alanine, o-nitrobenzyl-serine, O-(2-nitrobenzyl)-L-tyrosine, p- carboxymethyl-L-phenylalanine, m-cyano-L-phenylalanine, p-cyano-L-phenylalanine, biphenylalanine, 3-amino-L-tyrosine, bipyridyalanine, p-(2-amino-l-hydroxyethyl)-L- phenylalanine, p-isopropylthiocarbonyl-L-phenylalanine, N-acetyl-L-lysine, methyl-L- histidine, 5 -hydroxy -L-try ptophan. 7-chloro-try ptophan. and 7-bromo-tryptophan. For example, the nsAA may be O-methyl-tyrosine (OMeTyr).

[0064] According to the growth enabling method, the recombinant producer organism may be a bacterium or plant. The bacterium may be gram negative or gram positive. The bacterium may be Escherichia coli or Bacillus subtilis. The plant may be Arabidopsis thaliana. Where the recombinant producer organism is a plant, for example. Arabidopsis thaliana, the recombinant utilizer organism may be a rhizobacterium, for example, Bacillus subtilis. In one embodiment, the recombinant producer organism is Arabidopsis thaliana and the recombinant utilizer organism is Bacillus subtilis.

[0065] According to the growth enabling method, the recombinant utilizer organism may be a microorganism. In one embodiment, the recombinant utilizer organism and the recombinant producer organism are both microorganisms. The recombinant utilizer organism and the recombinant producer organism may be derived from the same microorganism, for example, Escherichia coli or Bacillus subtilis. The growth enabling method may further comprise one or more sets of microbes within a target microbiome. The microbiome may be selected from the group consisting of soil microbiomes, gut microbiomes, oral microbiomes, vaginal microbiomes, and skin microbiomes.

[0066] The growth enabling method may further comprise an additional recombinant utilizer organism dependent on the recombinant producer organism. Growth, survival, or a biological activity' of the additional recombinant utilizer organism may depend on the recombinant producer organism. The additional recombinant utilizer organism may be dependent on the nsAA produced by the recombinant producer organism.

[0067] The present invention also provides a method of enabling survival of a recombinant utilizer organism. The survival of the recombinant utilizer organism depends on a nonstandard amino acid (nsAA). The survival enabling method comprises providing a system. The system comprises the recombinant utilizer organism and a recombinant producer organism. The system lacks the nsAA and is not supplemented externally with the nsAA. The survival enabling method also comprises producing the nsAA by the recombinant producer organism into the system. The survival enabling method further comprises incubating the recombinant utilizer organism in the system after the production of nsAA. As a result, the recombinant utilizer organism survives in the system.

[0068] Where survival of a recombinant utilizer organism depends on a non-standard amino acid (nsAA), the recombinant utilizer organism may have a survival rate of less than about 10%, 5%, 1%, 0.1% or 0.01% in the absence of the nsAA, or the survival rate of the recombinant utilizer organism may be reduced by at least about 90%, 95%, 99%, 99.9% or 99.99% within a predetermined time, for example, about 0.5, 1, 2, 3, 4, 5, 6. 12, 24, 48, 72 or 96 hours after the nsAA is removed from the recombinant utilizer organism.

[0069] According to the survival enabling method, the recombinant utilizer organism may grow' in the system.

[0070] The survival enabling method may further comprise incorporating the nsAA produced by the recombinant producer organism into a target protein in the recombinant utilizer organism. The target protein may be an enzyme. Exemplary enzymes include luciferase, nitrogenase, phytase, acid phosphatase, surfactin synthetase, phosphopantetheinyl transferase, nitroreductase, polyethylene terephthalate esterase (PETase), cutinase, and allosteric transcription factors. Alternatively, the target protein may be anon-enzymatic protein, such as delta endotoxin (insecticide), green fluorescent protein, other fluorescent proteins, or other reporter proteins. The target protein may be essential to a biological activity of the recombinant utilizer organism. The activity of the biological function of the recombinant utilizer organism may be reduced by at least about 90%, 95%, 99%, 99.9% or 99.99% within a predetermined time, for example, about 0.5, 1, 2, 3, 4, 5, 6, 12, 24, 48, 72 or 96 hours after the nsAA is removed from the recombinant utilizer organism, or the increased by at least about 10. 100, 1.000 or 10,000 folds after the nsAA is incorporated into the target protein provided to the recombinant utilizer organism.

[0071] The survival enabling method may further comprise adjusting the production of the nsAA by the recombinant producer organism such that the growth of the recombinant utilizer organism is controlled.

[0072] According to the survival enabling method, the recombinant producer organism and the recombinant utilizer organism may be in direct contact. The recombinant producer organism and the recombinant utilizer organism may be in close proximity to each other, for example, within a distance of about 0, 0.00001, 0.0001, 0.001, 0.01, 0.1, 1, 10 or 100 cm, or about 0-100, 0.00001-100, 0.0001-100. 0.001-100, 0.01-100, 0.1-100. 1-100, 10-100.

[0073] 0.00001-10, 0.0001-10, 0.001-10, 0.01-10, 0.1-10, 1-10, 0.00001-1, 0.0001-1, 0.001-1, 0.01-1, 0.1-1, 0.00001-0.1, 0.0001-0.1, 0.001-0.1, 0.01-0.1, 0.00001-0.01, 0.0001-0.01, or 0.00001-0.0001 cm.

[0074] According to the survival enabling method, the nsAA may be present in the system at a desirable concentration, which may vary depending on what is required by the nsAA utilizer organism and / or conceivable to attain via biosynthesis. The nsAA concentration in the system may be the range of about 0.01-1,000,000 nM, 0.1-100,000 nM, 1-100,000 nM, 10-10.000 nM or 100-1,000 nM.

[0075] According to the survival enabling method, the nsAA may be any of the nsAAs known for genetic code expansion using orthogonal translation systems, such as O-methyl- tyrosine (OMeTyr), p-azido-L-phenylalanine, p-acetyl-L-phenylalanine, 3,4- dihydroxyphenylalanine, sulfotyrosine, p-amino-L-phenylalanine, p-ethylthiocarbonyl-L- phenylalanine, p-(3-oxobutanoyl)-L-phenylalanine, 1,5-dansyl-alanine, 7-amino-coumarin- alanine, 7-hydroxy-coumarin-alanine, o-nitrobenzyl-serine, O-(2-nitrobenzyl)-L-tyrosine. p- carboxymethyl-L-phenylalanine, m-cyano-L-phenylalanine, p-cyano-L-phenylalanine, biphenylalanine, 3-amino-L-tyrosine, bipyridyalanine, p-(2-amino-l -hydroxy ethyl)-L- phenylalanine, p-isopropylthiocarbonyl-L-phenylalanine. For example, the nsAA may be O- methyl-tyrosine (OMeTyr).

[0076] According to the survival enabling method, the recombinant producer organism may be a bacterium or plant. The bacterium may be gram negative or gram positive. The bacterium may be Escherichia coli or Bacillus subtilis. The plant may be Arabidopsis thaliana. Where the recombinant producer organism is a plant, for example, Arabidopsis thaliana, the recombinant utilizer organism may be a rhizobacterium, for example, Bacillus subtilis. In one embodiment, the recombinant producer organism is Arabidopsis thaliana and the recombinant utilizer organism is Bacillus subtilis.

[0077] According to the survival enabling method, the recombinant utilizer organism may be a microorganism. In one embodiment, the recombinant utilizer organism and the recombinant producer organism are both microorganisms. The recombinant utilizer organism and the recombinant producer organism may be derived from the same microorganism, for example, Escherichia coli or Bacillus subtilis.

[0078] The survival enabling method may further comprise one or more sets of microbes within a target microbiome. The microbiome may be selected from the group consisting of soil microbiomes, gut microbiomes, oral microbiomes, vaginal microbiomes, and skin microbiomes.

[0079] The survival enabling method may further comprise an additional recombinant utilizer organism dependent on the recombinant producer organism. Growth, survival, or biological activity of the additional recombinant utilizer organism may depend on the recombinant producer organism. The additional recombinant utilizer organism may be dependent on the nsAA produced by the recombinant producer organism.

[0080] The present invention further provides a method of enabling a biological function of a recombinant utilizer organism. The biological function of the recombinant utilizer organism depends on a non-standard amino acid (nsAA). The biological function enabling method comprises providing a system. The system comprises the recombinant utilizer organism and a recombinant producer organism. The system lacks the nsAA and is not supplemented externally with the nsAA. The biological function enabling method also comprises producing the nsAA by the recombinant producer organism into the system. The biological function enabling method further comprises incubating the recombinant utilizer organism in the system after the production of nsAA. As a result, the recombinant utilizer organism generates a biological function in the system.

[0081] Where a biological function of a recombinant utilizer organism depends on a nonstandard amino acid (nsAA), the activity of the biological function of the recombinant utilizer organism may be reduced by at least 90%, 95%, 99%, 99.9% or 99.99% within a predetermined time, for example, about 0.5, 1, 2, 3, 4, 5, 6, 12, 24, 48, 72 or 96 hours after the nsAA is removed from the recombinant utilizer organism, or the increased by at least 10, 100, 1,000 or 10,000 folds after the nsAA is provided to the recombinant utilizer organism.

[0082] According to the biological function enabling method, the recombinant utilizer organism may grow in the system.

[0083] According to the biological function enabling method, the nsAA may be essential to survival of the recombinant utilizer organism. The recombinant utilizer organism may have a survival rate of less than about 10%, 5%, 1%, 0.1% or 0.01% in the absence of the nsAA, or the survival rate of the recombinant utilizer organism may be reduced by at least about 90%, 95%, 99%, 99.9% or 99.99% within a predetermined time, for example, about 0.5, 1, 2, 3, 4, 5, 6, 12, 24, 48, 72 or 96 hours after the nsAA is removed from the recombinant utilizer organism.

[0084] The biological function enabling method may further comprise incorporating the nsAA produced by the recombinant producer organism into a target protein in the recombinant utilizer organism. The target protein may be an enzyme. Exemplary enzymes include luciferase, nitrogenase, phytase, acid phosphatase, surfactin synthetase, phosphopantetheinyl transferase, nitroreductase, polyethylene terephthalate esterase (PETase), cutinase, and allosteric transcription factors. Alternatively, the target protein may be anon-enzymatic protein, such as delta endotoxin (insecticide), green fluorescent protein, other fluorescent proteins, or other reporter proteins. The target protein may be essential to a biological activity of the recombinant utilizer organism. The activity' of the biological function of the recombinant utilizer organism may be reduced by at least about 90%, 95%, 99%, 99.9% or 99.99% within a predetermined time, for example, about 0.5, 1, 2. 3, 4, 5, 6. 12, 24, 48, 72 or 96 hours after the nsAA is removed from the recombinant utilizer organism, or the increased by at least 10, 100, 1,000 or 10,000 folds after the nsAA is incorporated into the target protein provided to the recombinant utilizer organism. The biological function enabling method may further comprise adjusting the production of the nsAA by the recombinant producer organism such that the growth of the recombinant utilizer organism is controlled.

[0085] According to the biological function enabling method, the recombinant producer organism and the recombinant utilizer organism may be in direct contact. The recombinant producer organism and the recombinant utilizer organism may be in close proximity to each other, for example, within a distance of about 0, 0.00001, 0.0001, 0.001, 0.01. 0.1, 1, 10 or 100 cm, or about 0-100, 0.00001-100, 0.0001-100, 0.001-100, 0.01-100, 0.1-100, 1-100, 10-100, 0.00001-10, 0.0001-10, 0.001-10, 0.01-10, 0.1-10, 1-10, 0.00001-1, 0.0001-1, 0.001-1, 0.01-1, 0.1-1, 0.00001-0.1, 0.0001-0.1, 0.001-0.1, 0.01-0.1, 0.00001-0.01, 0.0001- 0.01, or 0.00001-0.0001 cm.

[0086] According to the biological function enabling method, the nsAA may be present in the system at a desirable concentration, which may vary depending on what is required by the nsAA utilizer organism and / or conceivable to attain via biosynthesis. The nsAA concentration in the system may be the range of about 0.01-1,000,000 nM, 0.1-100,000 nM, 1-100.000 nM, 10-10,000 nM or 100-1,000 nM.

[0087] According to the biological function enabling method, the nsAA may be any of the nsAAs known for genetic code expansion using orthogonal translation systems, such as O- methyl-tyrosine (OMeTyr), p-azido-L-phenylalanine, p-acetyl-L-phenylalanine, 3,4- dihydroxyphenylalanine, sulfotyrosine, p-amino-L-phenylalanine, p-ethylthiocarbonyl-L- phenylalanine, p-(3-oxobutanoyl)-L-phenylalanine, 1,5-dansyl-alanine, 7-amino-coumarin- alanine, 7-hydroxy-coumarin-alanine, o-nitrobenzyl-serine, O-(2-nitrobenzyl)-L-tyrosine. p- carboxymethyl-L-phenylalanine, m-cyano-L-phenylalanine, p-cyano-L-phenylalanine, biphenylalanine, 3-amino-L-tyrosine, bipyridyalanine, p-(2-amino-l -hydroxy ethyl)-L- phenylalanine, p-isopropylthiocarbonyl-L-phenylalanine. For example, the nsAA may be O- methyl-tyrosine (OMeTyr).

[0088] According to the biological function enabling method, the recombinant producer organism may be a bacterium or plant. The bacterium may be gram negative or gram positive. The bacterium may be Escherichia coli or Bacillus subtilis. The plant may be Arabidopsis thaliana. Where the recombinant producer organism is a plant, for example, Arabidopsis thaliana, the recombinant utilizer organism may be a rhizobacterium, for example, Bacillus subtilis. In one embodiment, the recombinant producer organism is Arabidopsis thaliana and the recombinant utilizer organism is Bacillus subtilis.

[0089] According to the biological function enabling method, the recombinant utilizer organism may be a microorganism. In one embodiment, the recombinant utilizer organism and the recombinant producer organism are both microorganisms. The recombinant utilizer organism and the recombinant producer organism may be derived from the same microorganism, for example, Escherichia coli or Bacillus subtilis.

[0090] The biological function enabling method may further comprise one or more sets of microbes within a target microbiome. The microbiome may be selected from the group consisting of soil microbiomes, gut microbiomes, oral microbiomes, vaginal microbiomes, and skin microbiomes.

[0091] The biological function enabling method may further comprise an additional recombinant utilizer organism dependent on the recombinant producer organism. Growth, survival, or biological activity of the additional recombinant utilizer organism may depend on the recombinant producer organism. The additional recombinant utilizer organism may be dependent on the nsAA produced by the recombinant producer organism.

[0092] The term “about” as used herein when referring to a measurable value such as an amount, a percentage, and the like, is meant to encompass variations of ±20% or ±10%. more preferably ±5%, even more preferably ±1%, and still more preferably ±0.1% from the specified value, as such variations are appropriate.

[0093] Example 1. Engineering exclusive communication or reliance of one bacterium on another bacterium or one bacterium on a plant

[0094] A strategy has been demonstrated to control the survival of a synthetic auxotroph in a target non-sterile environment by transitioning its chemical dependence into a biological dependence on another engineered organism. This was accomplished by creating an orthogonal and obligate commensalism through pairing the biosynthesis of an orthogonal nsAA by one “producer” microbe and synthetic auxotrophy in a second “utilizer” microbe within a community (Fig. IB). Excitingly, this proof of principle demonstration using two Escherichia coli strains proves to withstand the presence of other microbes, as exclusive dependence of the utilizer microbe on the producer microbe was observed even within a model soil consortium. This proof-of-concept lays the groundwork for multiple strategies of environmental deployment that offer sustained biocontainment in ways that neither physical containment nor chemical-based biocontainment can currently provide.

[0095] The generalizability of this method to other pairs of organisms has further been demonstrated by creating nsAA utilizer strains of the rhizobacterium Bacillus subtilis along with nsAA producer lines of the model plant species Arabidopsis thaliana.

[0096] 1. Methods a. Strains and plasmids

[0097] Escherichia coli strains and plasmids used are listed in Table SI. Molecular cloning and vector propagation were performed in DH5a. Polymerase chain reaction (PCR) based DNA replication was performed using KOD XTREME Hot Start Polymerase. Cloning was performed using Gibson Assembly with constructs and oligos for PCR amplification shown in Table S2. Genes were purchased as G-Blocks or gene fragments from Integrated DNA Technologies (IDT) (Coralville, IA) or Twist Bioscience (San Francisco, CA) and were optimized for E. coli K12 using the IDT Codon Optimization Tool with sequences shown in Table S3 and Table S4. The new synthetic auxotrophs constructed in this study were derived from the published adk.d6 and DEP.e5 synthetic auxotrophs, which were kindly provided by George Church’s laboratory at Harvard Medical School. b. Materials and Chemicals

[0098] The following compounds were purchased from MilliporeSigma (Burlington, MA, USA): kanamycin sulfate, chloramphenicol, carbenicillin disodium, streptomycin sulfate salt, potassium phosphate dibasic, potassium phosphate monobasic, M9 salts, magnesium sulfate, calcium chloride dihydrate, imidazole, glycerol, sodium dodecyl-sulfate. Tris base, HEPES, trifluoroacetic acid, L-tyrosine, L-methionine, L-tryptophan, and KOD EXTREME Hot Start polymerase. Agarose and ethanol were purchased from Alfa Aesar (Ward Hill, MA, USA). D-glucose was purchased from TCI America (Portland, OR, USA). Anhydrotetracycline (aTc) and isopropyl B-d-1 -thiogalactopyranoside (IPTG) were purchased from Cayman Chemical (Ann Arbor, MI, USA). Hydrochloric acid was purchased from RICCA (Arlington, TX, USA). Acetonitrile, sodium chloride, (9-methyl-L- tyrosine, biphenylalanine (no longer available), LB Broth powder (Lennox), and LB Agar powder (Lennox) were purchased from Fisher Chemical (Hampton, NH, USA). L- Arabinose was purchased from VWR. A MOPS EZ-rich defined medium kit and components were purchased from Teknova (Hollister, CA, USA). Trace Elements A was purchased from Coming (Coming, NY, USA). Taq DNA ligase was purchased from GoldBio (St. Louis, MO, USA). Phusion DNA polymerase and T5 exonuclease were purchased from New England BioLabs (NEB) (Ipswich. MA, USA). Sybr Safe DNA gel stain was purchased from Tnvitrogen (Waltham, MA, USA). Kapa2G fast multiplex mix was purchased from Roche Diagnostics (Indianapolis, IN, USA). DNeasy Blood and Tissue Kit purchased from QIAGEN Sciences (Germantown, MD, USA). c. Biosynthesis of OMeTyr

[0099] Specified strains of E. coli (see Table S I) were transformed with plasmids expressing OMeTyr production pathway genes. For OMeTyr production, strains were inoculated from frozen stocks and grown overnight in 3 mL Lysogeny broth- Lennox media (LB: 10 g / L bacto tryptone, 5 g / L sodium chloride and 5 g / L yeast extract) and specified antibiotics at 34 °C with shaking at 250 RPM. These overnight cultures were used to inoculate experimental cultures targeting an OD600 value of 0.1. To investigate the impacts of the co-expression of aroG* on OMeTyr production, the following specified strains w ere inoculated in 3 mL M9 minimal media (Coming Trace Elements A (1.60 pg / mL CuSO4 ■ 5H2O. 863.00 pg / mL ZnSO4 • 7H2O, 17.30 pg / mL Selenite • 2Na and 1155.10 pg / mL ferric citrate), 5X M9 salts, 2 mM MgSO4. and 0. 1 mM CaC12) with 1% glucose (w / v), 0.004 mM biotin, and 0.1 pg / mL aTc. In addition, sAMFl was grown with 1 mM IPTG, 15 pg / mL kanamycin, and 17 pg / mL chloramphenicol, sAMF2 was grown with 30 pg / mL kanamycin, and Recoded RARE (DE3) was grown with 50 pg / mL carbenicillin. Cultures were incubated at 34 °C with shaking at 250 RPM for 24 h. To investigate the impacts of a constitutive promoter on OMeTyr production, sAMF3 and sAMF4 were each inoculated in 4 mL LB with 30 pg / mL kanamycin. 0.1 pg / mL aTc was also added to the cultures containing sAMF3 at inoculation. Cultures were incubated at 34 °C with shaking at 250 RPM. Supernatant samples were taken every other hour for the first 7 h post inoculation with a final sample taken at 24 h. To compare OMeTyr production in M9 minimal media to production in LB, sAMF4 was inoculated in 4 mL M9 minimal media with 1% glucose (w / v), 0.004 mM biotin, and 30 pg / mL kanamycin and the steps for the time course described above were repeated. All biosynthesis of OMeTyr was quantified via supernatant sampling, which involved pelleting via centrifugation using an Eppendorf 5430R refrigerated centrifuge, and extracting supernatant media to be analyzed using HPLC. d. nsAA incorporation assays For the initial synthetase screen, the orthogonal AARS / tRNA pairs, cloned with pEVOL plasmids were transformed into an E. coll MG1655 (DE3) strain with a pZE plasmid expressing a reporter protein fusion consisting of a ubiquitin domain, followed by an in-frame amber suppression codon, followed by GFP (pZE-Ub-AG-GFP). Sequences of the synthetases screened can be found in Table S4 and the reporter protein used in this study can be found in Table S5. These transformations were inoculated from frozen stocks and grown overnight in 3 mL LB with 30 pg / mL kanamycin and 34 pg / mL chloramphenicol at 37 °C with shaking at 250 RPM. After overnight growth, the cells were inoculated at lOOx dilution at 37 °C in 300 pL MOPS EZ Rich media with aromatic amino acid (Phe, Trp, and Tyr) dropout in deep 96-well plates with a specified concentration of OMeTyr (0 mM or 1 mM), 30 pg / mL kanamycin, 34 pg / mL chloramphenicol, and 2% glucose (w / v) with shaking at 1 ,000 RPM and an orbital radius of 3 mm. At mid-exponential growth (OD600 ~0.5), 0.1 pg / mL aTc and 0.2% (wt / vol) L-arabinose were added to induce transcription of mRNA that requires UAG suppression to form full-length GFP. Cultures were returned to 37 °C and grown for 18 h before pelleting via centrifugation. GFP fluorescence was quantified using a Spectramax i3x plate reader with Softmax Pro 7.0.3 software (excitation = 485 nm, emission = 525 nm).

[0100] For the comparison of the BipARS, NapARS, and NapARS’ aminoacyl-tRNA synthetases, strains sAMF8-10 were inoculated from frozen glycerol stocks and grown overnight in 3 mL LB media with 15 pg / mL kanamycin and 25 pg / mL carbenicillin at 34 °C with shaking at 250 RPM. Following overnight growth, the cells were used to inoculate culture tubes containing 15 pg / mL kanamycin, 25 pg / mL carbenicillin, 0.1 pg / mL aTc, 0.2% (wt / vol) L-arabinose, and either 0 or 0.5 mM OMeTyr grown again at 34 °C with shaking at 250 RPM. All three strains were inoculated in triplicate for both conditions targeting an ODeoo of 0.05. Samples were taken 4 h, 8 h. and 24 h post inoculation, which involved removing a 300 pL sample from each culture tube and pelleting cells via centrifugation and resuspending the pellets in PBS. Both ODeoo and GFP fluorescence were quantified using a Spectramax i3x plate reader with Softmax Pro 7.0.3 softw are (excitation = 485 nm. emission = 525 nm). e. HPLC analysis

[0101] OMeTyr was quantified via high-performance liquid chromatography (HPLC) using an Agilent 1100 Infinity model equipped with a Zorbax Eclipse Plus-C18 column (part number: 959701-902, 5 pm, 95A, 2.1 x 150 mm). We used an initial mobile phase of solvent A / B = 100 / 0 (solvent A, 50 mM potassium phosphate, pH 7.0; solvent B, 0.1% trifluoroacetic acid in acetonitrile, flow 1 mL min’1) and maintained for 1 min. A gradient elution was performed (A / B) with: gradient from 100 / 0 to 85 / 15 (flow 1 mL min’1to 0.5 mL min’1) for 1-6 min, gradient from 85 / 15 to 30 / 70 (flow 0.5 mL min’1) for 6-25 min, gradient from 30 / 70 to 100 / 0 (flow 0.5 mL min’1to L0 mL min’1) for 25-26 min, and equilibration at 100 / 0 for 26-27 min. Absorption was monitored at 276 and 250 nm. f. Co-culture for GFP production

[0102] Strains sAMF4-6 were inoculated from frozen stocks and grown overnight in 3 mL MOPS EZ rich media with 30 pg / mL kanamycin and 2% glucose (w / v) at 34 °C with shaking at 250 RPM. Prior to inoculation, all overnighted cells were diluted to OD 1.0. From the diluted cultures, the strains were used to inoculate MOPS EZ rich media with aromatic amino acid (Phe, Trp, and Tyr) dropout in a 96-well plate with 30 pg / mL kanamycin, 2% glucose (w / v), 0.004 mM biotin, and 0.1 pg / mL aTc with 200 pL total working volume at indicated ratios (1: 1, 1:2, and 1:5), with 2 uL of the OMeTyr (non-)producer strain used for each ratio. Cultures were grown at 34 °C using a SpectraMax i3x plate reader for 18 h, with OD600 and GFP fluorescence (excitation = 485 nm, emission = 525 nm) measurements taken every’ f O min, with medium plate shaking between reads for 12 h. g. Purification of GFP with OMeTyr incorporation

[0103] Strains sAMF5 and sAMF7 were inoculated from frozen stocks and grown overnight in 3 mL LB containing 30 pg / mL kanamycin. The next day, these cultures were used to inoculate 150 mL of an aromatic amino acid (Phe, Trp, Tyr) dropout version of MOPS EZ rich media containing 2% glucose, 30 pg / mL kanamycin. 0.1 mM Trp. 0.004 mM biotin and 0. 1 pg / mL aTc targeting an inoculation OD600 of 0.05 for each strain. The culture was incubated in a 34 °C shaking incubator at 250 RPM for 18 h overnight. The next day, the cells were pelleted by centrifugation using an Avanti J-15R refrigerated Beckman Coulter centrifuge at 4 °C and 4000 RPM for 10 min. Supernatant was aspirated and the pellets were resuspended in 10 mL of lysis buffer (25 mM HEPES, 10 mM imidazole, 300 mM NaCl, 10% glycerol, pH 7.4) and disrupted via sonication using a QSonica Q125 sonicator with 6 cycles of 5 s at 75% amplitude and 10 s off for 2 min cycles. The lysate was distributed into microcentrifuge tubes and centrifuged for 1 h at 18,213 g at 4 °C. The protein-containing supernatant was then removed and loaded into a HisTrap Ni-NTA column using an AKTA Pure GE FPLC system. After performing an isocratic wash at 25 mM and 50 mM imidazole. GFP was eluted in 250 mM imidazole in 1.5 mL fractions, which was clear to see via visible inspection. The fractions were combined and applied to an Amicon column (10 kDa MWCO) and the buffer was diluted at least LOOOx with a 25 mM HEPES buffer at pH 7.2. The protein was flash frozen using ultracold (-80 °C) ethanol and stored at -80 °C. h. Mass spectrometry

[0104] For intact protein MS measurements, samples were submitted to a Waters Acquity UPLC H-Class with an Acquity Protein BEH C4 column (part number: 186004495, 1.7 pm, 300 , 2.1 x 50 mm) coupled to a Xevo G2-XS Quadrupole Time-of-Flight (QToF) mass spectrometer. Protein sample was injected into a Waters Acquity with an initial mobile phase of solvent A / B = 85 / 15 (solvent A, water, 0.1% formic acid; solvent B, acetonitrile, 0.1% formic acid) held at 85 / 15 for 1 minute followed by a gradient elution from (A / B) 85 / 5 to 5 / 95 over 5 min. Flow rate was maintained at 0.5 mL min'1. The QToF mass spectrometer measurements were obtained following positive electrospray ionization (ES+) with source temperature of 150 °C. The spectra were analyzed from m / z 500 to 2000 and the spectra was deconvoluted using the maximum entropy function in MassLynx software. Here, an assumed uniform gaussian distribution (width = half height of 0.60 Da with 30% intensity' right and left channels) was used to evaluate a mass range from 37000 to 38000 Da (resolution: 0.1 Da / channel) at an elution time of 3.2 minutes.

[0105] For LC-MS / MS measurements, protein samples (20 pg / lane) were first submitted to SDS-PAGE. The gel was then stained with BIO-SAFE™ Coomassie stain (Bio-Rad), and de-stained with water. Then, protein bands corresponding to the molecular weight of the protein of interest w ere cut from the polyacrylamide gel and the gel fragments were digested using an in-gel tryptic digestion kit (Thermo Scientific: p / n 89871). Following digestion, extracted peptides w ere then desalted using PIERCE™ pipette tips (ThermoFisher), dried, and resuspended in 20 pL of 0.5% acetic acid (pH 4.5) and then subjected to LC-MS / MS using a Thermo Scientific Orbitrap Eclipse Tribrid Mass Spectrometer (MS; Thermo Scientific) with an Ultimate 3000 nano-liquid chromatography (nano-LC) and a FAIMS Pro Interface (Thermo Scientific). The LC-MS / MS analysis was performed using an Orbitrap Eclipse MS (Thermo Scientific). Peptides were first loaded onto a trap column (PepMap C 18) and then separated by an analytical column (PepMap C18, 2.0 um; 15 cm x 75mm I.D.; Thermo Scientific) at 300 nl / min flow rate using a binarybuffer system (buffer A, 0. 1% formic acid in water; buffer B, 0. 1% formic acid in acetonitrile) with a 165-min gradient (1 % to 10% in 8 min; then to 25% buffer B over 1 17 min; 25% to 32% buffer B in 10 min, then to 95% buffer B over 3 min; back to 1% B in 5 min. and stay equilibration at 1% B for 20 min). Multiple CVs (-40, -55 and -75) were applied for FAIMS separation. For all experiments, the survey scans (MSI) were acquired over a mass range of 375-1500 m / z at a resolution of 60,000 in the Orbitrap. The maximum injection time was set to Dynamic, and AGC target was set to Standard. Monoisotopic peak selection was set to Peptides, and the charge state filter was set to 2-7. For MS / MS acquisition, precursors were isolated with a width of 1.6 m / z, fragmented with HCD using 30% collision energy with a maximum injection time of 100 ms. and collected in Orbitrap at a resolution of 15,000. The dynamic exclusion was set to 60 s and can be shared across different FAIMS experiments. LC-MS / MS data was collected in n=3 technical replicates.

[0106] Proteomic analysis was performed in the MaxQuant- Andromeda software suite (version 1.6.3.4) with the default parameters. An E. coli reference proteome (strain BL21- DE3; taxonomy _id:469008) was used for database search. Other parameters include: trypsin as an enzy me with maximally two missed cleavage sites; protein N-terminal acety lation, methionine oxidation, and OMeTyr substitution for tyrosine were entered as variable modifications; cysteine carbamidomethylation as a fixed modification; and peptide length was set to a minimum of 7 ammo acids. The false-discovery rate of high-confidence protein and peptide identification was 1%. Peptide intensity values derived from MaxQuant were used for quantification. i. Escape assays

[0107] For kinetic, liquid media escape assays, the strains being tested were inoculated from frozen stocks and grown overnight in 4 mL of their permissive LB media (34 pg / mL chloramphenicol, 0.005% w / v sodium dodecyl sulfate (SDS), 20 mM tris-HCl buffer, 0.2% w / v arabinose and 0.1 mM BipA for B-adk.d6 and 30 pg / mL kanamycin, 0.005% w / v sodium dodecyl sulfate (SDS), 20 mM Tris-HCl buffer, 0.1 pg / mL aTc and 0.5 mM OMeTyr for N-adk.d6) at 34 °C with shaking at 250 RPM. Overnight cultures were washed by pelleting the cells through centrifugation at 4 °C, 4000 RPM for 5 min, aspirating the spent media, and resuspending the pellet in fresh LB. Following two washes, the cells were diluted to an ODeoo of 1.0, then they were used to inoculate experimental cultures in 200 pL volumes in a 96-well plate at lOOx dilution. Cell growth was monitored for 48 h in the synthetic auxotroph s non-permissive media (permissive media lacking required nsAA) in a SpectraMax i3x plate reader with medium plate shaking at 34 °C with ODeoo readings taken every 10 min. Three biological replicates were used to visualize the escape of each strain.

[0108] For solid media escape assays, all strains were inoculated from frozen stocks and grown overnight in their specified permissive media at 34 °C with shaking at 250 RPM. Cells were then washed as described for liquid media escape assays and resuspended in fresh LB media following washes. Three biological replicates were plated on permissive media following 10-fold serial dilutions in LB to determine the number of viable CFU’s plated and monitored for at least 2 days at 34 °C. To conduct serial dilutions, cells were pipetted up and down 10 times in each new well, new pipette tips were used for each round of serial dilutions, and multiple serial dilutions for each condition were plated to ensure cells were appropriately diluted 10-fold. The cell counts from permissive plates were also used to calculate the assay limit as l / (average total viable CFU). Three biological replicates were also plated on non-permissive media without prior serial dilutions and monitored for up to 14 days at 34 °C. Colonies were counted after the specified number of days and escape frequencies were calculated as colonies observed on non-permissive media lacking the synthetic auxotroph’s nsAA per average viable CFU plated for each strain. j. MAGE

[0109] Strain sAMF12 was inoculated from a frozen glycerol stock and grown overnight in permissive media at 34 °C with shaking at 250 RPM. The following day, a starter culture was inoculated in its permissive LB media (15 pg / mL kanamycin, 17 pg / mL chloramphenicol, 0.005% w / v sodium dodecyl sulfate (SDS), 20 mM Tris-HCl buffer, 0.1 pg / mL aTc and 0.5 mM OMeTyr) targeting an OD600 of 0.05 and grown at 34 °C to midexponential phase (OD600 ~0.4). Three cycles of MAGE were performed as previously described using 5 pM of the oligonucleotide targeting adk.d6 in Table S2. Following three rounds of MAGE, colonies were screened for the introduction of the TAG codon using multiplex allele-specific colony PCR followed by sequence verification. Following sequence confirmation, passaging in permissive LB media in the absence of kanamycin was used to remove the pORTMAGE-ECl plasmid for the creation of strain N-adk.d7. This process was repeated for strain sAMF13 using the oligonucleotide targeting hisD in Table S2. Following five rounds of MAGE, colonies were screened and passaged as described for the generation of strain N-adk.d7 to create the hisD strain. k. Co-culture assays

[0110] For the co-cultures involving N-adk.d6, the strain was inoculated from its frozen stock and grown overnight in 4 mL of its permissive LB media. Strains sAMF4 and sAMFl 1 were also inoculated from frozen stocks and grown overnight in 4 mL LB with 34 pg / mL chloramphenicol. As described for the escape assays, prior to inoculation strains were washed and diluted to an ODeoo of 1.0. A 96-deep-well plate with non-permissive media and no antibiotics was inoculated with 296 pL media and 4 pL culture volume. The conditions tested include a 1: 1 co-culture and individual monocultures of each strain. Within the same plate we included a permissive media control for both the co-culture and N-adk.d6 monoculture. Cultures were incubated for 24 h at 34 °C with shaking at 1000 RPM and an orbital radius of 3 mm. Following incubation, 10 pL samples from each well were plated on solid, permissive media containing 30 pg / mL kanamycin and solid media containing 34 pg / mL chloramphenicol to separately check for grow th of N-adk.d6 or sAMF4 and sAMFl 1 respectively. To determine colony counts, 10-fold dilutions from the N-adk.d6 monoculture and the co-culture containing N-adk.d6 and sAMF4 were also plated on permissive media containing 30 pg / mL kanamycin. The solid media plates were incubated at 34 °C for at least 24 h prior to counting colonies. To test different ratios, the inoculation volume of the (non-)producer was always held constant at 2 pL and the inoculation volume of N-adk.d6 was increased to match the designated ratio.

[0111] For the co-culture investigation involving N.dl-adk.d6, the strain was inoculated from its frozen stock and grow n overnight in 4 mL of its permissive LB media. Strains sAMF4 and sAMFl 1 were also inoculated from frozen stocks and grown overnight in 4 mL LB with 34 pg / mL chloramphenicol. As described for the escape assays, prior to inoculation strains were washed and diluted to an ODeoo of 1 .0. A 96-deep-well plate with non-permissive media and no antibiotics w as inoculated ith 295 pL media and 10 pL of N.dl-adk.d6 and 2 pL (non-)producer. The conditions tested include a 1:5 co-culture and individual monocultures of each strain. Within the same plate we included a permissive media control for both the co-culture and N-adk.d6 monoculture. Cultures were incubated for 24 h at 34 °C with shaking at 1000 RPM and an orbital radius of 3 mm. Following incubation, 10 pL samples from each w ell were plated on solid, permissive media containing 30 pg / mL kanamycin and solid media containing 34 pg / mL chloramphenicol to separately check for growth of N.dl-adk.d6 or sAMF4 and sAMFl 1, respectively. To determine colony counts, 10-fold dilutions from the N.dl-adk.d6 monoculture and the coculture containing N.dl-adk.d6 and sAMF4 were also plated on permissive media containing 30 pg / mL kanamycin. The solid media plates were incubated at 34 °C for at least 24 h prior to counting colonies.

[0112] For the co-culture investigation between two synthetic auxotrophs. DEP.e5 was first transformed with the optimized OMeTyr production plasmid to generate the DEP.e5 producer. Then, the DEP.e5 producer and N-adk.d6-strep w ere inoculated from frozen stocks and grown overnight in 4 rnL of their own permissive LB media (17 pg / mL chloramphenicol, 15 pg / mL kanamycin, 0.005% w / v sodium dodecyl sulfate (SDS), 20 mM tris-HCl buffer. 0.2% w / v arabinose and 0.01 mM BipA for the DEP.e5 producer and 15 pg / mL kanamycin, 47.5 pg / mL streptomycin, 0.005% w / v sodium dodecyl sulfate (SDS), 20 mM Tris-HCl buffer, 0. 1 pg / mL aTc and 0.5 mM OMeTyr for N-adk.d6-strep). Prior to co-culture inoculation, cells were w ashed 3 times in the same manner described for the escape assays and diluted to an ODsoo of 1.0 in fresh LB. Following washes, cells were inoculated into a 96-deep-w ell-plate with wells charged with 300 pL LB media without antibiotics containing 0.005% w / v sodium dodecyl sulfate (SDS), 20 mM tris-HCl buffer, 0.2% w / v arabinose, 0.1 pg / mL aTc, and 0.005 mM BipA (only when specified). A volume of 2 pL each strain was used to inoculate any well containing that strain. The culture was incubated for 48 h at 34 °C with shaking at 1000 RPM and an orbital radius of 3 mm. Following incubation, 10 pL samples from each well were plated on solid, permissive media selective for N-adk.d6 containing 15 pg / mL kanamycin, 47.5 pg / mL streptomycin. Plates were monitored at 34 °C for 72 h.

[0113] For the mutualism co-culture. N.dl-adk.d6 was inoculated from its frozen stock and grown overnight in 3 mL of its permissive LB media, hi l) and the hi l) producer strains were also inoculated from frozen stocks and grown overnight in 3 rnL LB with 50 pg / mL carbenicillin and 34 pg / mL chloramphenicol respectively. As described for the escape assays, prior to inoculation strains were washed and diluted to ODeoo of 1.0, however washes and dilutions were performed in PBS rather than LB. From the diluted cultures, the strains w ere used to inoculate M9 minimal media with 1% glucose (w / v), 0.004 mM biotin, 0.005% w / v sodium dodecyl sulfate (SDS), 20 mM Tris-HCl buffer, and 0.1 pg / mL aTc in a 96-well plate with 200 pL total working volume with 2 uL of each strain to create the A / 7z.s£>:N.dl-adk.d6 (1: 1) and the hisD producer:N.dl-adk.d6 (1 : 1) co-culture conditions. Both permissive (OMeTyr+ or His+) and non-permissive (OMeTyr- or His-) monoculture controls of the individual strains were also tested. Cultures were grown at 34 °C using a SpectraMax i3x plate reader for 48 h, with OD600 measurements taken every' 10 min, with medium plate shaking between reads. Immediately following 48 h growth on the plate reader, serial dilutions and 10 pL samples were plated on each strains’ permissive LB agar using antibiotic selection to determine colony counts.

[0114] 1. Consortium Experiments

[0115] Strains sAMF4, sAMFl 1, and N.dl-adk.d6-strep were inoculated from frozen stocks and grown overnight in 4 mL of their own permissive LB media. The soil microbes specified for this assay (S’, maltophilia, H. frisingense, C. pusilium, and P. putidd) were streaked on solid media from frozen stocks approximately 36 h prior to the experiment, with a single colony of each species grown overnight in 3 mL LB media at 34 °C with shaking at 250 RPM. Sub cultures of the soil microbes were grown the day of the experiment in 4 mL LB media at 34 °C with shaking at 250 RPM to at least 108cells / mL. Prior to culture inoculation, each soil microbe was diluted in fresh LB to the ODeoo corresponding to 108cells / mL. As described for the escape assays, E. coli strains were washed and diluted to an ODeoo of 1.0 and soil microbes were diluted to their ODeoo value corresponding to 108cells / mL. The soil microbes were combined at equal volumes into a 1: 1: 1 : 1 consortium following each of their individual dilutions. A 96-deep-well-plate with N.dl-adk.d6-strep’s non-permissive LB media without antibiotics was charged with 298 pL media for monoculture controls and 278 pL media for consortium grow th. The conditions tested include a 10: 1: 1 consortia containing 20 pL of N.dl-adk.d6-strep and 2 pL each of sAMF4. and the pre-combined soil consortium and a 10: 1: 1 consortium containing 20 pL of N.dl- adk.d6-strep, and 2 pL each of sAMFl 1 , and the pre-combined soil consortium. Cultures were incubated for 24 h at 34 °C with shaking at 1000 RPM and an orbital radius of 3 mm. Following incubation, 10 pL samples from each well w ere plated on solid, permissive media containing both 15 pg / mL kanamycin and 47.5 pg / mL streptomycin as well as solid LB media without antibiotics to separately check for growth of N-adk.d6-strep compared to the rest of the consortium. To determine colony counts, 10-fold dilutions from the consortium triplicates w ere also plated on the same permissive media described above. The solid media plates were incubated at 34 °C prior to counting colonies, which were checked at 24 h and 72 h after plating. To more rigorously check for escape of the synthetic auxotroph following growth in the consortium containing the non-producer. the entire remaining culture volume from each well was washed 2x in the same way described for escape assays. Cells were resuspended in 240 pL fresh LB following each wash. After the second wash, the 240 pL culture volume was spread evenly across 3 non-permissive plates selective for N.dl-adk.d6 containing 15 pg / mL kanamycin and 47.5 pg / mL streptomycin. All plates were monitored at 34 °C for 7 days.

[0116] 2. Results a. nsAA biosynthesis and incorporation

[0117] To design an orthogonal and obligate commensalism, we established criteria for an nsAA that could mediate a metabolic interaction between two organisms. An ideal nsAA would be: (1) uncommon in nature, to lower the risk of natural cross-feeding; (2) chemically inert, to prevent unwanted reactivity; (3) biosynthetically accessible by some combination of known metabolic pathways, to enable its engineered biosynthesis; (4) amenable to sitespecific incorporation within proteins using existing genetically encoded tools, to enable its utilization; (5) capable of supporting survival of a synthetic auxotroph, to force commensalism; and (6) biodegradable, to maintain strict reliance on the nsAA producer for survival and to avoid environmental pollution. Previously reported E. coli synthetic auxotrophs use orthogonal translation systems (OTSs) for the incorporation of a variety of different nsAAs, all of which have unclear biosynthetic routes from simple carbon sources. Instead, we built a system based on the nsAA O-methyl-L-tyrosine (OMeTyr), which met the first four criteria and was one of the first nsAAs site-specifically incorporated into proteins. We hypothesized that our chosen nsAA could create permissive conditions for growth of an existing or new E. coli synthetic auxotroph because of the known polyspecificity of several OTSs as well as the documented ability of some synthetic auxotrophs to grow by utilizing several different nsAAs. When we began our study, the biosynthesis of OMeTyr in heterologous hosts had not been reported. However, OMeTyr biosynthesis in cells that were supplied tyrosine-containing media was recently reported, as was the site-specific incorporation of biosynthesized OMeTyr within proteins in multiple heterologous hosts. For the design of an orthogonal obligate commensalism, de novo synthesis of OMeTyr from simple carbon sources must be achieved so that biosynthesis is guaranteed in any nutritionally complete environment.

[0118] To investigate OMeTyr biosynthesis, we engineered an E. colt strain for heterologous expression of an O-methyltransferase from the deep sea sediment-derived Streptomyces drozdowiczii, MfnG. This enzyme catalyzes the transfer of a methyl group from the donor co-factor S-adenosyl-L-methionine (SAM) to the substrate L-tyrosine (Fig. 1C). To identify potential bottlenecks in the biosynthesis of OMeTyr, we investigated the effects of external supplementation of L-methionine (a SAM precursor) and L-tyrosine in LB media. We found that supplementation of 5 mM L-tyrosine to LB media increased the OMeTyr concentrations measured at 18 h of aerobic fermentation. Having observed that L- tyrosine availability was a bottleneck, we employed a well-established strategy to increase flux through the aromatic amino acid biosynthesis pathway: expression of a feedbackresistant DAHP synthase (AroG_D146N, or AroG*). We sought to further improve the rate of OMeTyr biosynthesis as it would likely decrease the lag time and exclusion via resource competition of an OMeTyr-dependent microbe during eventual co-culture. We designed a single operon to contain all genes of the required enzymes (Fig. 1C) and compared the use of a constitutive promoter with an inducible promoter. We found that the use of a constitutive promoter increased the rate of OMeTyr biosynthesis in LB media, generating 0.398 ± 0.013 mM OMeTyr within the first 7 h after inoculation (Fig. ID) compared to the use of the original, inducible promoter when induced at inoculation. While OMeTyr biosynthesis was slower in M9 minimal media compared to LB, the cultures reached similar final concentrations of 0.843 ± 0.021 mM in M9 minimal media and 0.911 ± 0.004 mM in LB (Fig. ID). Overall, these results showed that our engineered strains could achieve OMeTyr biosynthesis autonomously, meaning from simple carbon sources, in media lacking tyrosine, and without the need for an additional chemical inducer. We were also optimistic that the improved production kinetics would be sufficient to sustain an OMeTyr- dependent strain during a co-culture in which dependence was unidirectional.

[0119] After optimizing OMeTyr biosynthesis, we focused on the first of two elements that would be required to create an OMeTyr-dependent E. coli strain: site-specific incorporation of OMeTyr within target proteins. We screened a library of previously reported OTSs comprised of engineered Methanoccildococcus jannaschii aminoacyl-tRNA synthetase and tRNA pairs using the widely accepted approach of suppression of an amber stop codon located within a superfolder GFP reporter (GFP_TAG) (Fig. IE). We chose to investigate the dose-response of our best performing synthetase, NapARS, which exhibited a 22-fold increase in ODeoo normalized GFP fluorescence (FL / ODsoo) when supplied with 1 mM OMeTyr compared to a 0 mM OMeTyr control. Importantly, this synthetase also had the lowest normalized fluorescence in the 0 mM OMeTyr condition among all the synthetases tested, indicating the highest fidelity for OMeTyr incorporation. Upon supplying lower concentrations of OMeTyr, we observed a greater than 13-fold increase in normalized GFP fluorescence even with as little as 0.01 mM OMeTyr supplied compared to the 0 mM nsAA condition (Fig. IF). b. Distribution of nsAA biosynthesis and incorporation functions across two strains After constructing OMeTyr biosynthesis and incorporation technology in separate strains, we next monitored and optimized the dynamics of biosynthesis, exchange, and incorporation of OMeTyr within a co-culture. We constructed one OMeTyr '‘producer” strain that harbored the plasmid containing constitutively expressed biosynthetic genes and another OMeTyr “utilizer” strain that harbored the plasmid containing inducible orthogonal translation machinery and reporter genes, each with common antibiotic resistance genes. We measured fluorescence to evaluate successful incorporation of biosynthesized OMeTyr in the utilizer strain (Fig. 2A). As a negative control, we also used a “non-producer” strain that carried a plasmid that harbored our desired antibiotic resistance but not the mfnG gene. When we varied the inoculation ratio of producer: utilizer strains in our co-culture, we saw an increase in fluorescence over the course of culture grow th that was indicative of OMeTyr incorporation within GFP. We observed a 13.5-fold increase in maximum fluorescence between the co-culture containing the OMeTyr producer strain compared to the co-culture containing the non-producer strain (Fig. 2B). We also found that increasing the inoculation ratio of utilizer to producer increased the rate of GFP production (Fig. 2B). To verify the fidelity of our OMeTyr biosynthesis and incorporation system distributed across two strains, we conducted a larger scale co-culture, purified the GFP produced in co-culture, and performed intact protein mass spectrometry. The observed mass was 37322 Da, which aligned with the expected mass for GFP containing OMeTyr (Fig. 2C). As additional verification we performed in-gel tryptic digestion and LC-MS / MS on the same GFP protein, where 99.7 % of the protein purified contained OMeTyr and only 0.3 % contained misincorporated L-tyrosine. Having demonstrated that producer and utilizer organisms could achieve robust and high-fidelity incorporation of an orthogonal building block within a co-culture, we next set out to create a strain that would rely on OMeTyr for its survival. c. Synthetic auxotroph dependent on OMeTyr

[0120] To create a synthetic auxotroph, we modified our system such that the utilizer would incorporate OMeTyr into an essential protein rather than GFP. Because existing synthetic auxotrophs exhibit low or undetectable rates of escape from biocontainment, we decided to investigate previously characterized synthetic auxotrophic markers. Specifically, we investigated the ability of the previously reported adk.d6 (Fig. 3 A) single-marker synthetic auxotrophic strain that was originally designed to depend on biphenylalanine (BipA) to grow on OMeTyr. Prior work has revealed the promiscuity of the BipARS synthetase harbored in these strains. Therefore, we hypothesized that the BipARS synthetase may be able to incorporate OMeTyr based on its poly specificity. However, the synthetic auxotrophic marker must also accept OMeTyr and be active for the synthetic auxotroph to survive. Interestingly, we found that both strains were able to grow on OMeTyr. The adk.dd synthetic auxotroph maintained the same growth profile when provided OMeTyr as when provided BipA (Fig. 3B).

[0121] Since the growth of the adk.d6 synthetic auxotroph was not impacted by the incorporation of OMeTyr, we chose to move forward with this strain, which we will denote as “B-adk.d6’' because of its BipARS aminoacyl-tRNA synthetase. We next carried out a traditional assay to measure the frequency of escape from biocontainment for B-adk.d6. We cultured cells to full density in permissive liquid media, washed cells repeatedly to remove nsAA, and then plated cells on both permissive (with nsAA) and non-permissive (without nsAA) solid media. The escape rate of B-adk.d6 was 4.0 x 10"6escapees / CFU after 24 h of incubation on solid media (Fig. 3C). Since this escape rate does not meet the NIH standard of 1 in 10’8escapees / CFU, we considered it necessary to modify this synthetic auxotroph to decrease its escape rate. Given our previous incorporation results, we hypothesized that replacing the original synthetase with NapARS would decrease the escape rate. Using transformation, passaging, and selection, we replaced the plasmid harbored by B-adk.d6 (which contained the BipARS -M tRNA^' / , OTS under an arabinose inducible promoter and chloramphenicol resistance) with a plasmid that contained the NapARS -MjtRNACUAOTS under the control of an anhydrotetracy cline-inducible promoter and kanamycin resistance. These modifications generated the new N-adk.d6 synthetic auxotroph (Fig. 3D). Sequencing of N-adk.d6 revealed that its synthetase was a chimera of NapARS and BipARS (NapARS’) that exhibited superior performance to BipARS.

[0122] We next investigated the ability of our new synthetic auxotroph to remain biocontained in the absence of OMeTyr by monitoring grow th for 48 h in non-permissive LB media (without OMeTyr) in 96-well microplates. We were pleased to discover that, in stark contrast with the B-adk.d6 strain, the N-adk.d6 strain did not appear to escape during 48 h of culturing under these conditions (Fig. 3E). When we compared escape frequency more quantitatively using larger culture volumes and CFU counting assays, we found that N-adk.d6 resulted in 6.0 x 10'9escapees / CFU after 24 h of incubation on solid media (Fig. 3F), w hich was reduced by three orders of magnitude compared to B-adk.d6 (Fig. 3C). However, the 48 h escape rate ofN-adk.d6 was 1.2 x 10’5escapees / CFU. We used whole genome sequencing to investigate the mechanism of escape for five randomly selected escapee colonies. The results showed that one escapee had a mutational reversion of the UAG codon to UUG (leucine) in the adk.dG gene, two escapees had single mutations in the mutS gene, and tw o escapees had no mutations.

[0123] Since sequencing results revealed that in some cases mutation was not required for escape from biocontainment, we hypothesized that instead low' levels of mischarging the orthogonal tRNA with standard amino acids catalyzed by the NapARS’ may be responsible. Accordingly, we next investigated the influence of orthogonal aminoacyl-tRNA synthetase expression level on escape frequency. When we plated N-adk.d6 on non-permissive solid media that lacked both the inducer of the synthetase and nsAA, we observed a 14-day escape rate of 3.0 x 10'8escapees / CFU. This 14-day escape rate without synthetase induction improved upon the 2-day escape rate with synthetase induction for N-adk.d6 by three orders of magnitude. Unfortunately, we found that the permissive condition required the presence of the inducer for growth of the bacterium, so we considered two engineering approaches to low er escape rate in the presence of the inducer.

[0124] To decrease the likelihood of translation of Adk.d6 in the absence of OMeTyr, we scanned the coding sequence for candidate sites to substitute a second in-frame UAG codon. We found that amino acid position 99, which was predicted to be highly permissive based on sequence conservation analysis, tolerated the substitution of an alanine to OMeTyr in what was originally an apparently spacious pocket w ith considerable distance from the ligand binding pocket. This substitution was performed using multiplex automatable genome engineering (MAGE) and resulted in the N-adk.d7 synthetic auxotroph, which exhibited a 14-day escape rate of 6.4 x 10'8escapees / CFU under conditions where permissive and non-permissive media are identical other than inclusion of OMeTyr (Fig. 3F). While this was a significant improvement upon the initial escape rate of N-adk.d6, we also pursued a second strategy7to reduce escape by lowering synthetase activity in the absence of OMeTyr. We again scanned the coding sequence for candidate sites to substitute an in-frame UAG and identified a target location in a spacious loop region far from any ligand binding locations. We inserted a UAG at position 251 of the NapARS’, creating an OMeTyr-dependent NapARS’. This strategy resulted in our best performing synthetic auxotroph, N.dl-adk.d6, which exhibits a 14-day escape rate of 2.8 x 10'9escapees / CFU (Fig. 3F). Aside from a few engineered strains reported to exhibit escape rates below detection limits of 10'11escapees / CFU after similar durations of monitoring, this M-day- escape rate is comparable with some of the best-performing biocontained strains previously7reported. d. Obligate commensalism based on OMeTyr

[0125] Having engineered robust producer and utilizer strains, we next investigated our concept of orthogonal obligate commensalism within a microbial co-culture. We hypothesized that, in media lacking OMeTyr, N.dl-adk.d6 should rely on the OMeTyr producing strain (Fig. 4A). For this to be true, the N.dl-adk.d6 strain must not be able to circumvent its dependence on OMeTyr through cross-feeding of other nutrients from prototrophic E. coli cells, and it also must not be able to grow using the OMeTyr that it had previously been exposed to. To help address the latter requirement, we measured the persistence of N.dl-adk.d6, which is the length of time that the cells are able to survive in the absence of OMeTyr under conditions that do not generate viable escapees. We used this to determine an inoculation volume where no persistence would be observed after 24 h of liquid culturing. We then constructed two sets of co-cultures at a 1 :5 inoculation ratio in LB media. These featured: (1) N.dl-adk.d6 (KanR) and a non-producer E. coli strain (CmR); (2) N.dl-adk.d6 and the OMeTyr producer strain (CmR). We also included a monoculture of N.dl-adk.d6 as a negative control. In each case, we assessed the abundance of each strain using antibiotic selection: OMeTyr and Kan for the utilizer. Cm for the producer. For comparison, since N-adk.d6 exhibited inferior biocontainment but superior growth, we also performed experiments with N-adk.d6 in place of N.dl-adk.d6 using a 1: 1 inoculation ratio. We performed a preliminary7assessment of biocontainment by plating 10 pL samples from each condition and were excited to observe that, after 24 h of culturing, N- adk.d6 and N.dl-adk.d6 (Fig. 4B) only survived in LB with either OMeTyr supplementation or co-inoculation with the producer strain. We more rigorously examined the rate of escape from biocontainment for N-adk.d6 from the non-permissive liquid co-culture scenario with the non-producer by plating the entire remaining 270 pL culture volume from each triplicate well on non-permissive media (OMeTyr ) with antibiotic selection for N-adk.d6. We observed no colony forming units throughout the entirety of the 7-day monitoring period at an assay detection limit of 7.56 x 10'6escapees / CFU calculated based on the number of cells used to inoculate the co-culture. We also examined the desired permissive conditions - the co-cultures of OMeTyr producer and N-adk.d6 or N.dl-adk.d6 utilizer strains- more thoroughly to quantify utilizer abundance after 24 h of liquid culturing, which was 2.1 x 104CFU / mL and 3.7 x 103CFU / mL respectively (Fig. 4C). For rigorous characterization, we also plated the remaining culture volume from each triplicate well on non-permissive media with antibiotic selection for the synthetic auxotroph to determine if exposure to the producer strain may have facilitated escape from biocontainment. Here, we observed considerable escape from the N-adk.d6 co-culture however, we were very excited that N.dl-adk.d6 only resulted in a total of 17 viable escapees across all replicates. We briefly investigated whether these escapees had acquired the MfnG containing plasmid that encoded chloramphenicol resistance from the producer strain and found that none of them exhibited chloramphenicol resistance. e. Control of orthogonal symbiosis culture composition

[0126] We were pleased to see our obligate commensalism function and next sought to improve the final abundance of the utilizer. We considered strategies including choosing a faster growing utilizer, varying inoculation ratio, slowing the growth rate of the producer, and engineering mutualism. To ascertain the influence of harnessing a faster growth rate and inoculation ratio on the final abundance of the utilizer, we turned to the faster growing N-adk.d6 and mathematical modeling. Specifically, we modeled the observed resource competition of our co-culture system by assuming standard Monod growth kinetics for the producer strain and non-interactive, double substrate-limited growth kinetics for the utilizer. We estimated the model parameters using Monod growth kinetics to determine substratedependent growth kinetics of the producer strain, utilizer strain, and OMeTyr production. The model predicted that increasing inoculation ratio between utilizer and producer strains would yield higher utilizer abundance, which we experimentally verified by testing ratios of 1 : 1, 1 :5. 1: 10. and 1:20 (Fig. 4D). We were pleased to see that the use of a 1 :20 inoculation ratio resulted in 1 . 1 x 106CFU / mL of N-adk.d6 after 24 h.

[0127] We next examined the influence of changing the producer grow th rate by transforming a previously published synthetic auxotroph that requires biphenylalanine (BipA) with the OMeTyr biosynthesis plasmid. In this manner, we could use exogenously supplied biphenylalanine concentrations to tune producer growth rate and also advance towards containment of the producer strain, albeit in its current form without autonomy. We chose DEP.e5 as the host, which was engineered to rely on BipA and further evolved to require very low concentrations. In principle, our obligate commensalism would then contain tw o synthetic auxotrophs that require different nsAAs. with N-adk.d6 dependent on the DEP.e5 producer for survival. We transformed the optimized plasmid for OMeTyr production into DEP.e5 to create the DEP.e5 OMeTyr producer (CmRand KanR) (Fig. 4E). We found that it was able to biosynthesize above our minimum target of 0.03 mM OMeTyr within 7 h, ultimately producing 0.286 ± 0.018 mM in 24 h in its permissive LB media (with 10 pM BipA supplemented).

[0128] We next investigated whether our obligate commensalism w ould hold between these two synthetic auxotrophs after confirming that they could not grow' on the non-cognate nsAA. To continue using antibiotic selection for strain identification following a co-culture, we transformed a plasmid harboring streptomycin resistance into N-adk.d6 to generate N- adk.d6-strep (KanRand StrepR). Then, we prepared a co-culture of the DEP.e5 producer and N-adk.d6-strep with a 1: 1 inoculation ratio in media that was permissive to DEP.e5 (5 pM BipA) but non-permissive to N-adk.d6-strep (0 pM OMeTyr). After 48 h of culturing in liquid media, we discovered via antibiotic selective plating that N-adk.d6-strep was able to survive when co-inoculated with the DEP.e5 producer (Fig. 4F), but had not survived in monocultures that contained BipA but lacked OMeTyr (Fig 4F). As anticipated, when N- adk.d6 was co-cultured with another highly engineered strain, the overall abundance of N- adk.d6 following the co-culture increased to 6.13 x 106CFU / mL, two orders of magnitude larger than the final abundance resulting from the initial co-culture.

[0129] After observing the dependence of N-adk.d6-strep on the DEP.e5 producer, we also hypothesized that by adjusting the BipA concentration supplied, and therefore the growth rate of the DEP.e5 producer, we might have greater tunability over the relative abundance of the biocontained strains. We examined this theoretically using an adapted version of the mathematical model developed for the original utilizer and producer strains. By modeling the growth of the DEP e5 producer using the same non-interactive double substrate limited grow th kinetics as the utilizer, our model predicted that tuning BipA concentration can result in a change in biomass accumulation of each species in the co-culture.

[0130] As another strategy to increase utilizer abundance, we investigated the influence of engineering mutualism. To construct this relationship, we used MAGE to knock out the hisD gene required for histidine production in E. coli to generate a histidine auxotroph (AHisD). We first used a kinetic assay to monitor ODeoo and confirm that AHisD would not grow without external histidine supplementation in M9 minimal media, but that it would be able to grow when histidine was supplied. We then transformed our OMeTyr production machinery into AHisD and confirmed its ability to biosynthesize OMeTyr. Finally, we cocultured both N.dl-adk.d6 and the AHisD OMeTyr producer together (Fig. 4G) in M9 minimal media and again monitored ODeoo to track growth. We were excited to see observable growth when the synthetic auxotroph and OMeTyr producing AHisD were cocultured together and that no growth was seen when N.dl-adk.d6 was co-cultured with AHisD that lacked the OMeTyr production machinery (Fig. 4H). Excitingly, following the co-culture with the AHisD OMeTy r producer, the final concentration ofN.dl-adk.d6 was nearly identical to the final concentration of N.dl-adk.d6 in monoculture externally supplied with OMeTyr at 8 x 106CFU / mL and 7 x 106CFU / mL respectively. This mutualistic relationship resulted in a final abundance of N.dl-adk.d6 three orders of magnitude higher than the final abundance following the obligate commensalism relationship. This successful demonstration of obligate mutualism illustrates the potential for further tuning of the system to improve survival of the synthetic auxotroph when considering eventual deployment in various formats. f. Orthogonal function within soil consortium

[0131] After successfully demonstrating the concept of an obligate commensalism that was orthogonal to non-producing E. coli, we then sought to ascertain whether the system would function orthogonally to other microbes within a consortium (Fig. 5A). To begin to understand the potential for orthogonality to natural environments, we chose four microbes from a simplified maize root soil consortium: Stenotrophomonas maltophilia, Curtobacterium pusilium, Herbaspirillum frisingense, and Pseudomonas putida. These four microbes exhibited growth in LB media and were susceptible to our chosen antibiotics, which allowed us to continue using antibiotic selection to monitor CFUs of the synthetic auxotroph. We first examined the orthogonality of OMeTyr production to the soil consortium with and without the OMeTyr producing E. coli strain present. While the microbial consortium did not produce any OMeTyr on its own, OMeTyr did appear to be consumed by the consortium since we observed that only about 0.25 mM OMeTyr remained 24 h after 0.5 mM was supplied (Fig. 5B). When a 1 : 1 ratio of the OMeTyr producer soil consortium were co-inoculated, the concentration of OMeTyr that remained in culture after 24 h was 0.05 mM (Fig. 5B). The low' OMeTyr concentration observed may have been due to increased competition for resources experienced by the producer strain as well as nsAA consumption. However, because the OMeTyr concentration exceeded the minimum threshold of 0.03 mM required for efficient incorporation within proteins (Fig. IF), we deemed these conditions sufficient to test the orthogonality of our obligate commensalism. We inoculated each of the soil microbes in pairwise co-cultures with N-adk.d6 (KanR) in permissive media to screen for interactions that would exclude either microbe. We confirmed that all species were capable of co-existence after a 24 h co-culture in the N- adk.d6 permissive environment by plating on selective solid media. However, we noticed some potential minimal growth with a light film formation from the C. pusilium monoculture plated with only kanamycin present. Therefore, we decided to move forward with consortium experiments using the N-adk.d6-strep (KanRand StrepR) strain because we did not see any exhibited growth of the soil microbes on agar containing both kanamycin and streptomycin.

[0132] We then investigated whether the growth of N.dl-adk.d6 in the soil consortium would be conditional upon the inclusion of the OMeTyr producer in non-permissive media. To continue using antibiotic selection for strain identification, we transformed a plasmid harboring streptomycin resistance into N.dl-adk.d6 to generate N.dl-adk.d6-strep (KanRand StrepR). We performed a preliminary assessment by plating 10 pL samples from both conditions. Excitingly, we saw that N.dl-adk.d6-strep was able to survive in the consortium when the OMeTyr producing strain was present but was unable to survive when a nonproducing E. coli strain replaced the producing strain (Fig. 5C). Through dilutions we determined that approximately 3.3 x 103CFU / mL N.dl-adk.d6-strep remained in the consortium after 24 h compared to no remaining synthetic auxotroph colonies when the OMeTyr producer was excluded from the consortium (Fig. 5D). We examined the rate of escape from biocontainment for N.dl-adk.d6-strep following its growth in the consortium containing the non-producer or the producer by plating the entire remining 270 uL culture volume from each triplicate well on non-permissive media (OMeTyr") with antibiotic selection for N.dl-adk.d6-strep. When the synthetic auxotroph was in the consortium with the non-producer. no escape was seen for 7-day monitoring period following the 24 h growth in liquid culture at an assay detection limit of 2.4 x 10"8escapees / CFU calculated based on the number of cells used to inoculate the consortium. When the synthetic auxotroph w as cultured in the consortium containing the producer strain, no escapee colonies were seen on non-permissive media until the third day of monitoring on non- permissive agar plates. After 7 days of monitoring approximately 200 escapee colonies formed per triplicate. Collectively, these results demonstrate that our proof-of-concept synthetic ecological relationship exhibits orthogonality and is obligate for the survival of the nsAA-dependent strain. At the same time, exposure of the nsAA-dependent strain to the nsAA producer strain results in a low level of escape that does not appear to be mediated by horizontal gene transfer. The basis of this escape and potential strategies to eliminate it will be the subject of future investigations. g. Strains of Bacillus subtilis engineered for nsAA-dependent reporter protein expression

[0133] We have also demonstrated that nsAA utilization and reliance can be engineered within bacteria other than E. co / i. such as the more environmentally relevant rhizobacterium and spore-former Bacillus subtilis. First, we show nsAA-dependent reporter activity in multiple strains of B. subtilis microbe at >2-sigma above background when supplied nsAA. Given that domesticated strains might be easier to engineer and that undomesticated strains may exhibit superior growth in soil, we chose two strains for investigation: B. subtilis PY79, a well-established laboratory strain, and B. subtilis UDI 022, an undomesticated isolate that exhibits plant growth promoting properties in certain contexts. In the lacA locus of each strain, we genomically integrated a selectable cassette that features constitutive expression of an orthogonal translation system (OTS), specifically the NapARS / tRNATyrcuA system previously reported to enable incorporation of naphthylalanine at positions indicated by the TAG codon (Fig. 6A). In our previous w ork, we have observed that this OTS also efficiently enables incorporation of our desired nsAA, G-methyl-L-tyrosine (OMeTyr or OMY), with relatively low rates of incorporation of standard amino acids occurring in the absence of provided nsAA. In addition to the OTS. we also genomically integrated at the amyE locus an mNeonGreen reporter that we modified to contain a TAG codon at position 17 so that its full-length translation would depend on availability of aminoacylated orthogonal tRNA. This reporter is under control of an IPTG-inducible promoter, pHyperSpank, and can be measured by plate reader fluorimetry at excitation and emission wavelengths of 485 nm and 525 nm, respectively. Finally, for both strains we included an important positive control which is an mNeonGreen reporter that contains no TAG codons, allowing estimation of wild-type levels of protein expression from this reporter under various conditions.

[0134] We began by assessing the roles of OMY concentration, media composition, and host strain on nsAA incorporation within the mNeonGreen reporter. To perform these experiments, we picked fresh colonies from LB plates to make a starter culture in LB. We grew these cultures to an optical density' at 600 nm (ODeoo) of 0.6-0.8. We then prepared experimental cultures with 1 mM IPTG and a variable concentration of nsAA ranging from 0-1 mM in either S750 or 1 / 2X MS + 10 mM glutamate (l / 2x MS + glut) media, seeding these cultures from starter cultures at an OD of 0.02. We then cultured experimental cultures overnight under shaking incubation at 25 °C for 18 hours. At the experimental endpoints, we diluted cultures 1 : 1 with PBS and then measured fluorescence (FL) and ODeoo using a Biotek Synergy Hl plate reader. Error is represented as a technical triplicate of each media and nsAA concentration.

[0135] Our results demonstrate that both the engineered PY79 and UDI 022 strains exhibit significant increases in fluorescence as a function of increasing OMY concentration under both the S750 and 'A MS + glut cases (Figs. 6B-C). The dynamic range (or fold-change relative to the no nsAA condition) is highest for PY79 in the S750 media, which is not that surprising given that this combination of strain and media is most well-adapted to laboratory experiments featuring B. subtilis. The inclusion of ! MS + glut, an adaptation of a defined media used for plant culturing called Murashige and Skoog, is designed to begin probing how these strains will fare under conditions beginning to resemble plant culturing conditions. UD1022 exhibits less preference for either media condition and its performance in 'A MS + glut is comparable to the performance of PY79. Additionally, it takes as little as 0.05 mM OMY to result in at least 2-fold activation for both strains under both conditions tested. Knowledge of the dose-response under these conditions helps inform what biosynthetic titer of OMY will be required of an engineered plant.

[0136] To more rigorously evaluate nsAA incorporation under soil-like conditions, 'e next performed assays either in soil extract (SESOM) or directly in sterilized soil. To our knowledge, the performance of genetic code expansion technologies have never previously been evaluated under such conditions. We designed a workflow to allow cells to incubate with OMY in autoclave sterilized soil (Espoma organic potting mix) for varying numbers of days, followed by an extraction and filtration with a 10 micron filter to recover cells from larger soil debris, and then performed analysis of cell populations by flow cytometry (Fig. 7 A). Our first step was to identify the forward and side scatter values that would correspond to B. subtilis singlets and to set gates accordingly, which we did by preparing cells directly in PBS rather than soil (Fig. 7B). Next, we performed various trials of incubating B. subtilis in sterilized soil and implementing our extraction, filtration, and flow cytometric workflow7. Under optimized conditions (Fig. 7C), we observed considerable background noise in our samples but were successfully able to recover cells within the gates that we had previously set. Using these conditions and gates, we began to incubate four different engineered strains of B. subtilis in sterilized soil: (1) PY79 harboring mNeonGreen; (2) PY79 harboring mNeonGreen-lTAG; (3) UD1022 harboring mNeonGreen; (4) UD1022 harboring mNeonGreen- 1 TAG. For the strains harboring reporters that contain 1TAG codon, we tested incubation with 500 pM OMY (equivalent to 500 nmol / g OMY / soil) or with no exogenously supplied OMY. For every case, we performed experiments in triplicate.

[0137] The UDI 022 strain background outperformed the PY79 strains in all cases involving sterilized soil. This result is consistent with UDI 022 being an undomesticated plant growth promoting rhizobacterial strain that is likely better adapted to soil. The population distributions, represented as histograms, shed interesting light on the performance of nsAA incorporation under these conditions. Fig. 7D show's one of the replicates from three different populations analyzed 3 days after incubation. The top distribution serves as a reference for the level of green fluorescent protein produced by the engineered UDI 022 strain under these conditions in a manner independent of nsAA incorporation. The middle distribution show's the fluorescence resulting from the population of cells that contain the mNeonGreen- 1 TAG reporter in the absence of OMY. One can see that the middle distribution is nearly two orders of magnitude left-shifted (based on the median fluorescence) relative to the top distribution. Encouragingly, the bottom distribution, which corresponds to the same genetic background as the middle distribution but instead in the presence of OMY, is relatively narrow and centered at nearly the same level of fluorescence as the top distribution.

[0138] To calculate this difference more rigorously, we opted to plot the average of median fluorescence resulting from all three replicates at each condition for UDI 022 (Fig. 7E). Here, it is worth noting that at the 1-day timepoint, there was not yet a clear nsAA- dependent reporter expression. On the other hand, at the 3-day timepoint, the average median fluorescence of the UD1022 1TAG reporter populations in the absence of OMY was 1200 ± 190 (a.u ). whereas the average median fluorescence of the UD1022 1TAG reporter populations in the presence of OMY was 47738 ± 7855 (a.u.). Thus, the difference between the population means is ~5.9 times sigma using the larger population standard deviation.

[0139] Collectively, our experiments conclusively demonstrate that our engineered B. subtilis strains are able to efficiently incorporate OMY within target proteins in soil-like conditions. We present several insights that are either novel, useful, or both regarding the influence of host strain and the apparent affinity of our system for OMY. These results provide a strongly encouraging foundation for the project vision of harnessing engineered B. subtilis sentinel strains that rely on the OMY biosynthesized by a plant’s roots. g. Strains of B. subtilis engineered to rely on nsAAs

[0140] We next transitioned from incorporating OMY within reporter proteins to instead incorporating it w ithin essential proteins, thereby making the full-length translation of essential proteins dependent on the provision of OMY. If an insufficient level of full-length translation of one or more essential proteins occurs, then a cell should lose viability. Thus, this engineered reliance on a synthetic building block is a form of intrinsic biological containment known as synthetic auxotrophy, and it should permit a. B. subtilis synthetic auxotroph to grow only in the presence of OMY or alongside a plant that can biosynthesize OMY.

[0141] We constructed candidate synthetic auxotrophs that depend on OMY by transforming engineered B. subtilis cells with a linear DNA construct consisting of an antibiotic resistance cassette, an artificial promoter, and an in-frame TAG codon in the N- terminal region of candidate auxotrophic markers or, if applicable, at a predetermined site further into the gene that had been shown to work for E. coll synthetic auxotrophs (Fig. 8). These strains already contained a genome integrated expression cassette that enabled production of the orthogonal translation system for incorporation of OMY within proteins. If the site selected for the ns AA was more than 8 amino acids into the sequence, we ordered a gBlock to codon shuffle the region of the essential gene to ensure the UAG was appropriately inserted into the native locus during transformation. For those markers previously demonstrated to depend on nsAAs in E. coll, we designed cassettes to insert the UAG of interest in an identical site by performing pairwise sequence alignments of the protein sequences.

[0142] Using this strategy, we identified dxs and metS as two potential genes that can produce partially effective synthetic auxotrophs that are dependent on OMY for some period of time before visible outgrowth of an escapee from biocontainment. We next sought to improve the escape rate by combining the synthetic auxotrophic markers. We successfully generated a double marker synthetic auxotroph (dxs and metS) that we characterized further, including by examining its growth rate as a function of OMY concentration. We found that media conditions and temperature can affect the observed rate of escape from biocontainment, with rates of escape substantially decreasing under conditions more relevant to plant culturing conditions. We performed traditional measurements of the rate of escape from biocontainment (or escape assays) on solid media using l / 2x MS based minimal media agar plates incubated at 25 °C. We first streaked out the synthetic auxotroph from glycerol stocks onto permissive media plates containing 0.8x antibiotic concentrations (Cm: 4 g / mL, Kan: 8 g / mL, Spec 80 g / mL - as specified for the strain) and 300-500 pM of OMY. Non-permissive media is identical media lacking the nsAA. Plates are incubated for 24 h at 37 °C until colonies are large enough to pick. A colony is then picked and used to inoculate a permissive liquid culture comprising l / 2x MS supplemented with 1 .5% sucrose and 10 mM L-glutamate. Cultures are grown for either 4-6 h for the single marker synthetic auxotrophs or overnight for the double marker synthetic auxotrophs at 25 °C shaking or spinning in a rotor at max speed. Once cultures reach an OD of 0.4-0.6, cells are harvested and washed 5 times with non-permissive experimental media via centrifugation at 10,000 rpm for 2 min for each spin. Culture ODs are measured following washes using a BioTek synergy Hl plate reader and resuspended to a final OD of 0.5 or 1.0 depending on the assay conditions. Then, lOx serial dilutions are performed in non-permissive media until at least a 106dilution is reached. For initial tests of an escape rate, all dilutions are plated as 10 L dots on both non-permissive and permissive media. Following the first escape assay, a second experiment is performed in which the dilution that contains a countable concentration of colonies for both the non-permissive and permissive conditions is plated on a full agar plate. The escape rate is calculated by determining the total number of CFUs plated based on the permissive condition first then the escape rate is the number of escapees divided by the total number of CFUs plated. We monitored plates for two weeks. Encouragingly, under these conditions, escape from biocontainment for the dxs / metS double marker synthetic auxotroph was below the limit of detection (~107escapees / CFU) for up to one week (Fig. 9). We only began to observe escape for all three replicates after 10 days on the plates, and the rate of escape never surpassed 10'6escapees / CFU. h. Engineering OMeTyr biosynthesis in plants

[0143] If engineered nsAA-dependent bacteria are deployed in a field environment, we can ensure that their survival and / or activity' is reliant on the presence of a specific, target plant by engineering the plant to produce the nsAA. To that end, we engineered the model plant Arabidopsis thaliana (Columbia / Col-0 ecotype) to produce OMY by expressing the bacterially-derived MfnG enzyme under the control of the constitutive UBIQUITIN 10 (UBQ10) promoter or the root cap-specific BEARSKIN1 (BRN1) promoter. MfnG was fused to sfGFP for fluorescence imaging. Transgenic plants were generated by Agrobacterium tumefaciens-ms aXsA floral dip. Briefly, A. tumefaciens cultures were inoculated from frozen into 14-ml culture tubes containing 2 mL of LB media with antibiotics. Cultures were grown for 24 h at 28°C and 250 r.p.m. After incubation, 400 pL of bacterial culture was plated on 15 cm round LB agar plates containing antibiotics and incubated for two nights at 28°C. Next, the cells were scraped from plates and resuspended in 300 mL transformation buffer (per Liter: 50 g sucrose and 250 pL silwet-77). Arabidopsis inflorescences were submerged in the bacterial solution for 5 seconds, then transferred to a dark, humid container where they were allowed to incubate overnight. The following day, transformed plants were returned to normal growth conditions (22°C in 16 / 8-h light / dark cycles with 60% humidity', in Percival-Scientific growth chambers), where they continued to grow until they7produced seeds. Transgenic seeds were isolated using fluorescence microscopy by selection for the seed-specific TagRFP marker. Expression of the MfnG-GFP fusion in the engineered Arabidopsis was confirmed with fluorescent microscopy (Leica THUNDER stereomicroscope, GFP filter). Transgenic seeds homozygous for MfhG-GFP driven by the UBQ10 and BRN1 promoters were plated on solid growth medium (IX Murashige-Skoog basal salts, 1% sucrose, 0.6% Gelzan), stratified at 4°C for 3 days to break seed dormancy, then grown upright in normal growth conditions for 8 days. As expected, we saw green fluorescence distributed throughout the seedling when MfhG-GFP expression is driven by the UBQ10 promoter, and localized to the root cap when using the root cap specific BRN1 promoter (Fig. 10).

[0144] We then sought to quantify production of OMeTyr in plant root exudate. We sowed seeds onto plastic meshes in individual wells of a 48-well plate for hydroponic growth in 250 pL growth media (IX Murashige-Skoog basal salts, 1% sucrose). Plants were grown at normal growth conditions for 7 days. We pooled media from 10-25 individual seedlings, lyophilized overnight, and extracted with 80% methanol for LCMS analysis (Agilent 6520, Cl 8 column, positive ion mode). Titers for exudate-conditioned media were in the 1-60 pM range (Fig. 11).

[0145] Having confirmed OMY production in the MfhG-expressing Arabidopsis lines, we next sought to determine if the plants could support OMY-dependent activity in bacteria. To visualize OMY -mediated plant-microbe communication, w e used B. subtilis strains that expressed NanoLuc luciferase reporters for OMY incorporation, similar to the mNeonGreen reporter described above. The NanoLuc OMY reporter strains were streaked from frozen onto LB plates and incubated at 37°C overnight. Single colonies were inoculated into 10 mL LB cultures in 55 mL glass tubes, and incubated at 37°C for 4 hours. Cells were pelleted at 4000g for 10 minutes, then washed 3 times with 10 mL IX PBS with 5-minute 4000g spins. Cells were added to autoclaved MS agar (0.5x MS, 1% sucrose, 1% agar) cooled to ~45°C for a final OD=0. 1. Seeds were germinated separately and transferred freshly-made bacteria-agar matrix at 7-8 days after stratification. Plates were then closed with micropore tape and grown upright at normal Arabidopsis grow th conditions for 4 days, to allow time for OMY production and bacterial growth. To visualize bacterial NanoLuc production, we mixed NanoLuc substrate (NanoFuel® NLuc FLASH Substrate) with 0.8% agarose cooled to ~45°C at a 1 :500 dilution and applied it as a thin film that covers the entire plate. The plated w as immediately imaged to capture the resulting luminescence signal with the Chemi Default setting on an iBright. This luciferase reporter assay allow s us to visualize the extent of the impact of nsAA secretion in the rhizosphere. A schematic is shown in Fig. 12.

[0146] For each condition, we investigated three homozygous Arabidopsis lines and three different bacterial strains. The three plant lines were: 1) wild type Col-0, 2) a line where the ubiquitous and constitutive UBQ10 promoter drives OMY production, and 3) a line where root cap specific BRN1 promoter drives OMY production. We also tested three different bacterial strains, starting with derivatives of PY79 (Fig. 13): 1) the base WT PY79, 2) the positive control is a strain that constitutively produces luciferase, and 3) our test strain, which produces functional luciferase only upon incorporating OMY. Encouragingly, we see no signal for the wild type bacteria. We see robust signal from the constitutive strain regardless of plant line. For the OMY-dependent strain, we only see signal on the plates with OMY being produced. The signal is robust for the ubiquitous OMY -production plants, and weaker but perceptible in the root cap specific lines.

[0147] We conducted the same experiment with the same three homozygous Arabidopsis lines and three different bacterial strains, where this time the bacterial strains were derived from the undomesticated field isolate Bacillus subtilis UD1022 (Fig. 14). There were some minor differences in the conditions used for the assay between the UDI 022 and PY79 experiments that do not allow for a direct comparison: The UDI 022 experiment used seedlings that were one day older when transferred to plate than the PY79 experiment; The UDI 022 experiment had twice the volume of luciferin-agarose than the PY79 experiment; The UDI 022 experiment used a lower exposure (2 seconds rather than 10 seconds) to prevent signal saturation. We saw similar results as before: constitutive luciferase is always on, whereas OMY-dependent luciferase is only on in the presence of OMY-producing plant. The third column of the figure shows the 3 different UD1022-derived strains imaged together for a direct comparison of signal intensity.

[0148] By coupling the nsAA-producing Arabidopsis thaliana lines and the strains of B. subtilis that are engineered to rely on nsAA for survival, we would expect to see bacterial growth only in the proximity of the plant in comparable zones as we see for the nsAA- dependent luminescence signal. Thus, we have shown that the exclusive communication and reliance method that we have built is generalizable across taxonomical kingdoms.

[0149] Table SI. Strains and plasmids used in this study.

[0150] Table S2. Oligos used in this study

[0151] * = Phosphorothioate bonds between bases

[0152] Table S3. DNA G-Block used for MfnG cloning in this study.

[0153] Table S4. Sequences of AARS variants used in this study

[0154] Table S5. Sequences of reporter protein used in this study

[0155] All documents, books, manuals, papers, patents, published patent applications, guides, abstracts, and / or other references cited herein are incorporated by reference in their entirety. Other embodiments of the invention will be apparent to those skilled in the art from consideration of the specification and practice of the invention disclosed herein. It is intended that the specification and examples be considered as exemplary only, with the true scope and spirit of the invention being indicated by the following claims.

Claims

WHAT IS CLAIMED:

1. A method of enabling growth of a recombinant utilizer organism, wherein the growth of the recombinant utilizer organism depends on a non-standard amino acid (nsAA), the method comprising:(a) providing a system comprising the recombinant utilizer organism and a recombinant producer organism, wherein the system lacks the nsAA and is not supplemented externally with the nsAA;(b) producing the nsAA by the recombinant producer organism into the system; and(c) incubating the recombinant utilizer organism in the system after step (b), whereby the recombinant utilizer organism grows in the system.

2. A method of enabling survival of a recombinant utilizer organism, wherein the survival of the recombinant utilizer organism depends on a non-standard amino acid (nsAA), the method comprising:(a) providing a system comprising the recombinant utilizer organism and a recombinant producer organism, wherein the system lacks the nsAA and is not supplemented externally with the nsAA;(b) producing the nsAA by the recombinant producer organism into the system; and(c) incubating the recombinant utilizer organism in the system after step (b), whereby the recombinant utilizer organism survives in the system.

3. A method of enabling a biological function of the recombinant utilizer organism, wherein the biological function of the recombinant utilizer organism depends on a non-standard amino acid (nsAA), the method comprising:(a) providing a system comprising the recombinant utilizer organism and a recombinant producer organism, wherein the system lacks the nsAA and is not supplemented externally with the nsAA;(b) producing the nsAA by the recombinant producer organism into the system; and(c) incubating the recombinant utilizer organism in the system after step (b), whereby the recombinant utilizer organism generates the biological funebon in the system.

4. The method of claim 2 or 3, wherein the recombinant utilizer organism grows in the system.

5. The method of claim 1 or 3. wherein the nsAA is essential to survival of the recombinant utilizer organism.

6. The method of any one of claims 1-5, further comprising incorporating the nsAA produced by the recombinant producer organism into a target protein in the recombinant utilizer organism.

7. The method of claim 6, wherein the target protein is an enzyme.

8. The method of claim 6 or 7, wherein the target protein is essential to a biological activity of the recombinant utilizer organism.

9. The method of any one of claims 1-8, further comprising adjusting the production of the nsAA by the recombinant producer organism, whereby the growth of the recombinant utilizer organism is controlled.

10. The method of any one of claims 1-9, wherein the recombinant producer organism and the recombinant utilizer organism are in direct contact.

11. The method of any one of claims 1-10, wherein the nsAA is present in the system at a concentration from 1 nM to 100 mM.

12. The method of any one of claims 1-11, wherein the nsAA is O-methyl- tyrosine (OMeTyr).

13. The method of any one of claims 1-12, wherein the recombinant producer organism is a bacterium.

14. The method of any one of claims 1-13, wherein the recombinant producer organism is a plant.

15. The method of claim 14, wherein the plant is Arabidopsis thaliana.

16. The method of claim 14 or 15, wherein the recombinant utilizer organism is a rhizobacterium.

17. The method of claim 16, wherein the rhizobacterium is Bacillus subtilis.

18. The method of any one of claims 1-17, wherein the recombinant utilizer organism is a microorganism.

19. The method of any one of claims 1-18, further comprising one or more sets of microbes within a target microbiome selected from the group consisting of soilmicrobiomes, gut microbiomes, oral microbiomes, vaginal microbiomes, and skin microbiomes.

20. The method of any one of claims 1-19, wherein the system further comprises an additional recombinant utilizer organism dependent on the recombinant producer organism.

Images