Patent Application
20180105825
The present disclosure provides synthetic repressor constructs and the proteins encoded therein, as well as synthetic repressible promoter constructs for use in combination with the synthetic repressor constructs/synthetic repressors disclosed herein. Various combinations of synthetic repressor constructs and synthetic repressible promoter constructs are also provided in synthetic genetic circuits for modifying expression of a protein of interest in a plant cell.
A paper copy of the sequence listing and a computer readable form of the same sequence listing are appended below and herein incorporated by reference. The information recorded in computer readable form is identical to the written sequence listing, according to 37 C.F.R. 1.821(f).
As plant biotechnologists desire to install more transgenes together, the transcriptional control of these genes becomes increasingly important. Numerous synthetic genetic systems allowing controlled activation, and thereby increased levels, of gene expression exist. However, some applications require that a gene be expressed at high levels than repressed. Accordingly, there remains a need in the art for novel, synthetic genetic circuits in plants that function independently from or even replace endogenous networks.
In an aspect, the present disclosure encompasses a synthetic repressor construct for modifying gene expression in a plant, comprising a nucleic acid encoding a transcriptional repressor domain linked to a DNA-binding domain, wherein the transcriptional repressor domain and the DNA-binding domain are operable in the same plant species.
In another aspect, the present disclosure encompasses a synthetic repressible promoter construct for use in combination with a synthetic repressor construct, the synthetic repressible promoter construct comprising: (a) a nucleic acid sequence encoding a core promoter capable of conferring constitutive gene expression in a plant species, the core promoter optionally comprising a TATA box; and (b) a synthetic regulatory element comprising at least one copy of a binding element having a nucleic acid sequence capable of specifically binding its cognate DNA-binding domain, the copy of the at least one binding element inserted at a position upstream of the core promoter, downstream of the core promoter but before the translation start site for a protein of interest, or proximal to the 5′ end of the optionally present TATA box when present.
In another aspect, the present disclosure encompasses an artificial genetic circuit for modifying expression of a protein of interest in a plant, comprising: a promoter operably linked to a synthetic repressor construct, the synthetic repressor construct comprising a nucleic acid encoding a transcriptional repressor domain linked to a DNA-binding domain; a nucleic acid construct comprising a nucleic acid encoding a protein of interest; a synthetic repressible promoter construct operably linked to the nucleic acid encoding the protein of interest, the synthetic repressible promoter construct comprising: (i) a nucleic acid sequence encoding a core promoter capable of conferring constitutive gene expression in a plant species, the core promoter optionally comprising a TATA box, and (ii) a synthetic regulatory element comprising at least one copy of a binding element having a nucleic acid sequence capable of specifically binding the DNA-binding domain of the synthetic repressor, the copy of the at least one binding element inserted at a position upstream of the core promoter, downstream of the core promoter region but before a translation start site for the protein of interest, or proximal to the 5′ end of the optionally present TATA box when present; and wherein the transcriptional repressor domain of the synthetic repressor, the DNA-binding domain of the synthetic repressor, the promoter operably linked to the synthetic repressor construct, and the core promoter of the synthetic repressible construct are each operable in the same plant species.
In another aspect, the present disclosure encompasses a transgenic plant cell comprising any one of a synthetic repressor construct disclosed herein, a synthetic repressible promoter construct disclosed herein, or a synthetic genetic circuit disclosed herein, or any combination thereof.
In another aspect, the present disclosure encompasses a transgenic plant comprising any one of a synthetic repressor construct disclosed herein, a synthetic repressible promoter construct disclosed herein, or a synthetic genetic circuit disclosed herein, or any combination thereof.
In another aspect, the present disclosure encompasses a kit comprising any one of a synthetic repressor construct disclosed herein, a synthetic repressible promoter construct disclosed herein, or a synthetic genetic circuit disclosed herein, or any combination thereof.
In another aspect, the present disclosure encompasses a method for modifying expression of a protein of interest in a plant, the method comprising introducing a synthetic genetic circuit as disclosed herein, into a cell of the plant, wherein the promoter operably linked to the synthetic repressor construct is an inducible promoter.
In another aspect, the present disclosure encompasses a method for creating a library comprising a plurality of synthetic repressible promoter constructs, the method comprising: providing a construct comprising a core promoter capable of conferring constitutive gene expression in a plant species, wherein the promoter optionally comprises a TATA box, and modifying the construct a plurality of times by (a) introducing one or more copies of a binding element having a nucleic acid sequence capable of specifically binding a DNA-binding domain, the copy of the binding element inserted at a position upstream of the core promoter, downstream of the core promoter but before a translation start site for a protein of interest, or proximal to the 5′ end of the optionally present TATA box when present, and then (b) varying the number of binding elements at a given position and/or the spacing between the 2 or more binding elements at a given position.
Other aspects and iterations of the invention are described more thoroughly below.
The application file contains at least one photograph executed in color. Copies of this patent application publication with color photographs will be provided by the Office upon request and payment of the necessary fee.
The present disclosure provides synthetic promoters and their corresponding synthetic repressor proteins. These promoters allow continuous gene expression in plants and plant cells in the absence of the corresponding repressor. Expression of genes operably linked to the synthetic promoters can be repressed by the presence of the corresponding repressor protein.
Unless defined otherwise, all technical and scientific terms used herein have the meaning commonly understood by a person skilled in the art to which this invention belongs. The following references provide one of skill with a general definition of many of the terms used in this invention: Singleton et al., Dictionary of Microbiology and Molecular Biology (2nd ed. 1994); The Cambridge Dictionary of Science and Technology (Walker ed., 1988); The Glossary of Genetics, 5th Ed., R. Rieger et al. (eds.), Springer Verlag (1991); and Hale & Marham, The Harper Collins Dictionary of Biology (1991). As used herein, the following terms have the meanings ascribed to them unless specified otherwise.
When introducing elements of the present disclosure or the preferred embodiments(s) thereof, the articles “a”, “an”, “the” and “said” are intended to mean that there are one or more of the elements. The terms “comprising”, “including” and “having” are intended to be inclusive and mean that there may be additional elements other than the listed elements.
As used herein, the term “construct” refers to any recombinant polynucleotide molecule.
As used herein, the term “endogenous sequence” refers to a chromosomal sequence that is native to the cell.
The term “exogenous,” as used herein, refers to a sequence that is not native to the cell, or a chromosomal sequence whose native location in the genome of the cell is in a different chromosomal location.
A “gene,” as used herein, refers to a DNA region (including exons and introns) encoding a gene product, as well as all DNA regions which regulate the production of the gene product, whether or not such regulatory sequences are adjacent to coding and/or transcribed sequences. Accordingly, a gene includes, but is not necessarily limited to, promoter sequences, terminators, translational regulatory sequences such as ribosome binding sites and internal ribosome entry sites, enhancers, silencers, insulators, boundary elements, replication origins, matrix attachment sites, and locus control regions.
The term “heterologous” refers to an entity that is not endogenous or native to the cell of interest. For example, a heterologous protein refers to a protein that is derived from or was originally derived from an exogenous source, such as an exogenously introduced nucleic acid sequence. In some instances, the heterologous protein is not normally produced by the cell of interest.
The terms “nucleic acid” and “polynucleotide” refer to deoxyribonucleotide or ribonucleotide polymer, in linear or circular conformation, and in either single- or double-stranded form. For the purposes of the present disclosure, these terms are not to be construed as limiting with respect to the length of a polymer. The terms can encompass known analogs of natural nucleotides, as well as nucleotides that are modified in the base, sugar and/or phosphate moieties (e.g., phosphorothioate backbones). In general, an analog of a particular nucleotide has the same base-pairing specificity; i.e., an analog of A will base-pair with T.
The term “nucleotide” refers to deoxyribonucleotides or ribonucleotides. The nucleotides may be standard nucleotides (i.e., adenosine, guanosine, cytidine, thymidine, and uridine) or nucleotide analogs. A nucleotide analog refers to a nucleotide having a modified purine or pyrimidine base or a modified ribose moiety. A nucleotide analog may be a naturally occurring nucleotide (e.g., inosine) or a non-naturally occurring nucleotide. Non-limiting examples of modifications on the sugar or base moieties of a nucleotide include the addition (or removal) of acetyl groups, amino groups, carboxyl groups, carboxymethyl groups, hydroxyl groups, methyl groups, phosphoryl groups, and thiol groups, as well as the substitution of the carbon and nitrogen atoms of the bases with other atoms (e.g., 7-deaza purines). Nucleotide analogs also include dideoxy nucleotides, 2′-0-methyl nucleotides, locked nucleic acids (LNA), peptide nucleic acids (PNA), and morpholinos.
The term “operably-linked”, as used herein, means that expression of a gene is under the control of a promoter with which it is spatially connected. A promoter may be positioned 5′ (upstream) of a gene under its control. The distance between the promoter and a gene may be approximately the same as the distance between that promoter and the gene it controls in the gene from which the promoter is derived. As is known in the art, variation in this distance may be accommodated without loss of promoter function.
The terms “polypeptide” and “protein” are used interchangeably to refer to a polymer of amino acid residues.
The term “promoter”, as used herein, refers to a synthetic or naturally-derived nucleic acid sequence which is capable of conferring, activating or enhancing expression of a gene in a cell. A promoter may comprise one or more specific transcriptional regulatory sequences to activate or enhance expression and/or to alter the spatial expression and/or temporal expression of the gene. In prokaryotes, these regulatory elements may include, but are not limited to, the −35 RNA polymerase recognition sequence, the −10 ribosome binding sequence, and upstream distal enhancer elements, which can be located as much as several thousand base pairs from the start site of transcription. Non-limiting examples of regulatory elements found in eukaryotes include the TATA box, initiator elements, downstream core promoter element, CAAT box, and the GC box. An inducible promoter is a promoter that is induced by the presence or absence of biotic or abiotic factors. An inducible promoter allows for the expression of the gene operably linked to it to be turned on or off (i.e., controlled).
The term “specifically binds”, as used herein in reference to the interaction of DNA-binding molecule and its cognate binding element, means that the interaction is dependent upon the presence of a particular structure. For example, specific binding between a DNA-binding molecule and a synthetic repressible construct may be demonstrated, for example, by the absence of binding of the DNA-binding domain to the synthetic repressible promoter construct when the binding element(s) are removed.
Techniques for determining nucleic acid and amino acid sequence identity are known in the art. Typically, such techniques include determining the nucleotide sequence of the mRNA for a gene and/or determining the amino acid sequence encoded thereby, and comparing these sequences to a second nucleotide or amino acid sequence. Genomic sequences can also be determined and compared in this fashion. In general, identity refers to an exact nucleotide-to-nucleotide or amino acid-to-amino acid correspondence of two polynucleotides or polypeptide sequences, respectively. Two or more sequences (polynucleotide or amino acid) can be compared by determining their percent identity. The percent identity of two sequences, whether nucleic acid or amino acid sequences, is the number of exact matches between two aligned sequences divided by the length of the shorter sequences and multiplied by 100. An approximate alignment for nucleic acid sequences is provided by the local homology algorithm of Smith and Waterman, Advances in Applied Mathematics 2:482-489 (1981). This algorithm can be applied to amino acid sequences by using the scoring matrix developed by Dayhoff, Atlas of Protein Sequences and Structure, M. O. Dayhoff ed., 5 suppl. 3:353-358, National Biomedical Research Foundation, Washington, D.C., USA, and normalized by Gribskov, Nucl. Acids Res. 14(6):6745-6763 (1986). An exemplary implementation of this algorithm to determine percent identity of a sequence is provided by the Genetics Computer Group (Madison, Wis.) in the “BestFit” utility application. Other suitable programs for calculating the percent identity or similarity between sequences are generally known in the art, for example, another alignment program is BLAST, used with default parameters. For example, BLASTN and BLASTP can be used using the following default parameters: genetic code=standard; filter=none; strand=both; cutoff=60; expect=10; Matrix=BLOSUM62; Descriptions=50 sequences; sort by=HIGH SCORE; Databases=non-redundant, GenBank+EMBL+DDBJ+PDB+GenBank CDS translations+Swiss protein+Spupdate+PIR. Details of these programs can be found on the GenBank website.
The present disclosure provides synthetic repressor constructs and synthetic repressor encoded thereby, as well as synthetic repressible promoter constructs for use in combination with the synthetic repressor constructs/synthetic repressors.
In various embodiments, the constructs of the present disclosure can be present in a vector. Suitable vectors include plasmid vectors, phagemids, cosmids, artificial/mini-chromosomes, transposons, and viral vectors (e.g., lentiviral vectors, adeno-associated viral vectors, etc.). The choice of the vector will vary depending upon the intended use (e.g., stable transformation in bacterial cells, stable transformation in plant cells, transient transformation in plant cells, etc.). In one embodiment, the synthetic repressor construct and the synthetic repressible promoter construct are present in a plasmid vector. In another embodiment, the synthetic repressor construct is present in first plasmid vector and the synthetic repressible promoter construct is present in a second plasmid vector. Non-limiting examples of suitable plasmid vectors include pUC, pBR322, pET, pBluescript, pCAMBIA2300, pRI 101, pBI121, pPZP100, and variants thereof. The vector can comprise additional expression control sequences (e.g., enhancer sequences, Kozak sequences, polyadenylation sequences, transcriptional termination sequences, etc.), selectable marker sequences (e.g., antibiotic resistance genes), origins of replication, and the like. Additional information can be found in “Current Protocols in Molecular Biology” Ausubel et al., John Wiley & Sons, New York, 2003 or “Molecular Cloning: A Laboratory Manual” Sambrook & Russell, Cold Spring Harbor Press, Cold Spring Harbor, N.Y., 3rd edition, 2001.
(a) Synthetic Repressor Construct and Synthetic Repressor Protein
In an aspect, the present disclosure provides a synthetic repressor construct for modifying gene expression in a plant. The synthetic repressor construct comprises a nucleic acid encoding a transcriptional repressor domain linked to a DNA-binding domain. The transcriptional repressor domain and the DNA-binding domain are operable (i.e., carry out their intended function) in the same plant species but may or may not be derived from the same source and may or may not be native to the plant species. The nucleic acid encoding the transcriptional repressor domain can be linked directly to the DNA-binding domain or indirectly to the DNA-binding domain (e.g., separated by three or more nucleotides). The polynucleotide sequence encoding the transcriptional repressor domain can be 5′ of the polynucleotide sequence encoding the DNA-binding domain, or vice versa. In certain embodiments, the nucleic acid encoding a transcriptional repressor domain linked to a DNA-binding domain further encodes a nuclear localization signal. In all embodiments, the polynucleotide sequence encoding the transcriptional repressor domain and the polynucleotide sequence encoding the DNA-binding domain are in-frame such that when a promoter is operably linked to a synthetic repressor construct, the expressed nucleic acid will be translated as single protein (e.g., a fusion protein).
In another aspect, the present disclosure provides a synthetic repressor comprising a transcriptional repressor domain and a DNA-binding domain. The transcriptional repressor domain can be linked directly to the DNA-binding domain or indirectly to the DNA-binding domain. In embodiments where the transcriptional repressor domain and the DNA-binding domain are indirectly linked, the domains are connected by a flexible linker comprised of one or more amino acids (e.g., 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, or more amino acids). In certain embodiments, the flexible linker can be a nuclear localization signal and/or a marker domain.
Transcriptional repressor domains are protein domains that are sufficient to confer the capacity for repression of transcription when linked to a heterologous DNA-binding domain. In general, a transcriptional repressor domain interacts with transcriptional control elements and/or transcriptional regulatory proteins (e.g., transcription factors, RNA polymerases, etc.) to decrease and/or terminate transcription of a gene. Non-limiting examples of transcriptional repressor domains include inducible cAMP early repressor (ICER) domains, Kruppel-associated box A (KRAB-A) repressor domains, YY1 glycine rich repressor domains, Sp1-like repressors, E(spl) repressors, IκB repressor, MeCP2, BRD transcriptional repressor domains, OVATE family protein (OFP) transcriptional repressor domains, mSin interaction domains (SID), EAR transcriptional repressor domains (including SRDX domains), and derivatives thereof.
A DNA-binding domain is an independently folded protein domain that contains at least one structural motif that recognizes a double- or single-stranded DNA. Non-limiting examples of suitable DNA-binding domains include a helix-turn-helix motif, a zinc finger domain, a leucine zipper domain, a winged helix domain, a winged helix-turn-helix domain, a helix-loop-helix domain, an HGM-box, a WOr3 domain, an OB-fold domain, a B3 DNA binding domain, or a transcriptional activator-like effectors (TALE) DNA binding domain. Suitable DNA-binding domains can also be computationally designed. See, for example, Huang, et al., (2016a) Nature, 537: 320-327; Huang, et al. (2016b) Nat Chem Biol, 12: 29-34; Rose, et al. (2017) Nat Chem Biol, 13: 119-12. DNA-binding domains can recognize a specific DNA sequence (i.e., a recognition sequence) or have a general affinity to DNA, though sequence-specific DNA binding domains are preferred. Suitable DNA-binding domains may be derived from any known DNA-binding protein, including but not limited to viral, prokaryotic or eukaryotic transcription factors or endonucleases. In various embodiments, a suitable DNA-binding domain may be derived from a plant DNA-binding protein, an insect DNA-binding protein, a yeast DNA-binding protein, a fungal DNA-binding protein, a bacterial DNA-binding protein, a nematode DNA-binding protein a mammalian DNA-binding protein. Alternatively, a suitable DNA-binding domain may be any other type of sequence-specific repressor known in the art including, but not limited to a CRISPR/cas protein lacking endonuclease activity. For example, Cas9 can be modified by mutating the RuvC and HNH domains such that they no longer possess nuclease activity. While CRISPR/cas proteins themselves do not recognize double- or single-stranded DNA, they can be used to target any DNA sequence with the help of guide-RNAs, which have sequence homology to the target site. Methods for designing DNA-binding domains and other sequence-specific repressors to target various sites are known in the art.
The present disclosure contemplates various combinations of transcriptional repressor domains and DNA-binding domains known in the art, provided the transcriptional repressor domain and the DNA-binding domain are both operable in the same plant species. In certain embodiments, the DNA-binding domain is not native to the plant species in which expression is desired. In further embodiments, the DNA-binding domain is a sequence-specific DNA binding domain. In still further embodiments, the DNA binding domain is a sequence-specific, DNA-binding domain with a recognition sequence that is not specifically recognized by a transcription factor native to the plant species in which expression is desired.
In an exemplary embodiment, a synthetic repressor construct comprises a nucleic acid encoding a transcriptional repressor domain linked to a DNA-binding domain, wherein the DNA-binding domain is a yeast Gal4 DNA-binding domain or a bacterial LexA DNA-binding domain, and the transcriptional repressor domain is an EAR transcriptional repressor domain, an OFP transcriptional repressor domain, or a BRD transcriptional repressor domain. In a further embodiment, the OFP transcriptional repressor domain is SEQ ID NO: 35. In another exemplary embodiment, a synthetic repressor construct comprises a nucleic acid encoding a transcriptional repressor domain linked to a DNA-binding domain, wherein the transcriptional repressor domain is an OFP transcriptional repressor domain and the DNA-binding domain is a bacterial LexA DNA-binding domain. In a further embodiment, the OFP transcriptional repressor domain is SEQ ID NO: 35. In another exemplary embodiment, a synthetic repressor construct comprises a nucleic acid encoding a transcriptional repressor domain linked to a DNA-binding domain, wherein the transcriptional repressor domain is a BRD transcriptional repressor domain and the DNA-binding domain is a yeast Gal4 DNA-binding domain. In another exemplary embodiment, a synthetic repressor construct comprises a nucleic acid selected from the group consisting of SEQ ID NO: 27-34.
Synthetic repressor constructs of the present disclosure may further comprise a polynucleotide encoding additional domains and/or other genetic elements. Additional domains include, but are not limited to, nuclear localization signal or other signal sequences for targeting proteins to subcellular compartments, a cell-penetrating domain, and a marker domain. Other genetic elements include, but not limited to, promoter sequences, terminators, translational regulatory sequences such as ribosome binding sites and internal ribosome entry sites, enhancers, silencers, insulators or other boundary elements, polyadenylation signals, locus control regions, etc.
In some embodiments, a promoter is operably linked to a synthetic repressor construct of the present disclosure. The promoter can be a constitutive promoter or an inducible promoter, and can be a synthetic promoter or a promoter from a native gene (e.g. a viral promoter, a bacterial promoter, an archaeal promoter, a plant promoter, an insect promoter, a nematode promoter, a yeast promoter, a fungal promoter, a mammalian promoter, etc.). In certain embodiments, the promoter is capable of conferring expression in a plant species. In further embodiments, the promoter is an inducible promoter. In still further embodiments, the promoter is sensitive to a computationally designed ligand binding transcription factor. See, for example, Bick, et al., (2017) eLife, 6; Feng, et al., (2015) eLife, 4, e10606.
Any inducible promoter for plants can be utilized in the instant invention. The promoter may be responsive to an internal or an external factor (inducer). Several systems for induction of transgene expression in plants are known in the art. See, for example, Borghi et al, Method Mol Biol, 2010, 655:65-75; Gatz Curr Opinion Biotechnology 1996, 7:168-172; U.S. Pat. No. 5,750,385, U.S. Pat. No. 5,420,034, U.S. Pat. No. 5,753,475, U.S. Pat. No. 6,281,410, each hereby incorporated by reference in its entirety. Additional, non-limiting examples include AlcR/AlcA (ethanol inducible); GR fusions, GVG, and pOp/LhGR (dexamethasone inducible); XVE/OlexA (beta-estradiol inducible); as well as known promoters responsive to a variety of environmental factors, including but not limited to light-inducible or stress-inducible (e.g. water deficit, cold, heat, salt, pest, disease, nutrient stress, etc.) promoters. Suitable inducible promoters are also described in the examples.
When expression of a synthetic repressor of the present disclosure is desired in a eukaryotic cell (e.g., in an isolated plant cell, a cell of a plant seed, or a cell in a whole plant), the synthetic repressor also comprises a nuclear localization signal. In general, a nuclear localization signal comprises a stretch of basic amino acids. Nuclear localization signals are known in the art. The nuclear localization signal can be located at the N-terminus, the C-terminus, or in an internal location of the synthetic repressor.
Transport of protein produced by transgenes to a subcellular compartment such as the chloroplast, vacuole, peroxisome, glyoxysome, cell wall or mitochondrion, or for secretion into the apoplast, is accomplished by means of operably linking the nucleotide sequence encoding a signal sequence to the 5′ and/or 3′ region of a gene encoding the protein of interest. Targeting sequences at the 5′ and/or 3′ end of the structural gene may determine, during protein synthesis and processing, where the encoded protein is ultimately compartmentalized. The presence of a signal sequence directs a polypeptide to either an intracellular organelle or subcellular compartment or for secretion to the apoplast. Any signal sequence known in the art is contemplated by the present invention.
In still other embodiments, the synthetic repressor can further comprise at least one marker domain. Non-limiting examples of marker domains include fluorescent proteins, luciferase enzymes, purification tags, and epitope tags. In some embodiments, the marker domain can be a fluorescent protein. Non limiting examples of suitable fluorescent proteins include green fluorescent proteins (e.g., GFP, GFP-2, tagGFP, turboGFP, EGFP, Emerald, Azami Green, Monomeric Azami Green, CopGFP, AceGFP, ZsGreenl), yellow fluorescent proteins (e.g., YFP, EYFP, Citrine, Venus, YPet, PhiYFP, ZsYellowl,), blue fluorescent proteins (e.g., EBFP, EBFP2, Azurite, mKalamal, GFPuv, Sapphire, T-sapphire,), cyan fluorescent proteins (e.g. ECFP, Cerulean, CyPet, AmCyanl, Midoriishi-Cyan), red fluorescent proteins (e.g., mKate, mKate2, mPlum, DsRed monomer, mCherry, mRFP1, DsRed-Express, DsRed2, DsRed-Monomer, HcRed-Tandem, HcRedl, AsRed2, eqFP611, mRasberry, mStrawberry, Jred), and orange fluorescent proteins (e.g., mOrange, mKO, Kusabira-Orange, Monomeric Kusabira-Orange, mTangerine, tdTomato) or any other suitable fluorescent protein. In other embodiments, the marker domain can be a luciferase enzyme. Non-limiting examples include firefly luciferase, Renilla luciferase, Nanoluc luciferase, and derivatives thereof. In other embodiments, the marker domain can be a purification tag and/or an epitope tag. Exemplary tags include, but are not limited to, glutathione-S-transferase (GST), chitin binding protein (CBP), maltose binding protein, thioredoxin (TRX), poly(NANP), tandem affinity purification (TAP) tag, myc, AcV5, AU1, AU5, E, ECS, E2, FLAG, HA, nus, Softag 1, Softag 3, Strep, SBP, Glu-Glu, HSV, KT3, S, S1, T7, V5, VSV-G, 6×His, biotin carboxyl carrier protein (BCCP), and calmodulin. The marker domain can be located at the N-terminus, the C-terminal, or in an internal location of the synthetic repressor.
In still other embodiments, the synthetic repressor can further comprise at least one cell penetrating domain. In one embodiment, the cell-penetrating domain can be a cell-penetrating peptide sequence derived from the HIV-1 TAT protein. In another embodiment, the cell-penetrating domain can be TLM, a cell-penetrating peptide sequence derived from the human hepatitis B virus. In still another embodiment, the cell-penetrating domain can be MPG. In an additional embodiment, the cell-penetrating domain can be Pep-1, VP22, a cell penetrating peptide from Herpes simplex virus, or a polyarginine peptide sequence. The cell-penetrating domain can be located at the N-terminus, the C-terminus, or in an internal location of the synthetic repressor.
(b) Synthetic Repressible Promoter Construct
In another aspect, the present disclosure provides a synthetic repressible promoter construct for use in combination with a synthetic repressor construct of Section I(a). In various embodiments, the 3′ end of a synthetic repressible promoter construct described below may be proximal to a cloning site. Alternatively, or in addition, a synthetic repressible promoter construct described below may be operably linked to a nucleic acid encoding a protein of interest.
Synthetic repressible promoter constructs of the present disclosure comprise (i) a nucleic acid sequence encoding a core promoter region capable of conferring constitutive gene expression in a plant species, the promoter region optionally comprising a TATA box, and (ii) a synthetic regulatory element comprising at least one copy of a binding element having a nucleic acid sequence capable of specifically binding the DNA binding domain of the cognate synthetic repressor. In some embodiments, a synthetic repressible promoter construct is operably linked to a nucleic acid encoding a protein of interest. In preferred embodiments, the binding element is a recognition sequence and the DNA-binding protein is a sequence-specific, DNA-binding protein.
A core promoter is typically a minimal set of nucleotides capable of driving accurate transcription initiation when bound by a basal transcription factor. A core promoter may contain a TATA-box, a GA element, a CAAT box, or other core promoter elements known in the art. Suitable core promoters are operable in the plant species in which expression is desired, but may or may not be native to the plant species. Further, the core promoter may or may not be derived from the same species as the transcriptional repressor domain, DNA-binding domain, or any other element of the synthetic repressor construct; however, in order to function together each of these elements must be operable in the plant species in which expression is desired.
Suitable core promoters capable of conferring constitutive gene expression in a desired plant species are known in the art. Non-limiting examples of suitable core promoters include promoters from plant viruses (e.g., Cauliflower Mosaic Virus (CaMV 35S) promoter, Figwort Mosaic Virus (FMV) promoter, etc.) bacterial plant pathogens (e.g., Nopaline Synthase (NOS) promoter, etc.), the promoters from such genes as rice actin (e.g., OsACT2.1), maize ubiquitin (e.g., ZmUBI1), and corn H3 histone, and also the ALS promoter, a XbaI/NcoI fragment 5′ to the Brassica napus ALS3 structural gene (or a nucleotide sequence that has substantial sequence similarity to the XbaI/NcoI fragment). In some embodiments, the core promoter confers constitutive gene expression only in one tissue or cell type of the plant. In other embodiments, the core promoter confers constitutive gene expression in one or more cell types of the plant. In other embodiments, the core promoter confers constitutive gene expression in one or more tissues of the plant. In other embodiments, the core promoter confers constitutive gene expression in all cell types of the plant.
Any tissue-specific or tissue-preferred promoter can be utilized in the instant invention. Exemplary tissue-specific or tissue-preferred promoters include, but are not limited to, a seed-preferred promoter such as that from the phaseolin gene; a leaf-specific and light-induced promoter such as that from cab or rubisco; an anther-specific promoter such as that from LAT52; a pollen specific promoter such as that from Zm13 or a microspore-preferred promoter such as that from apg.
At a position upstream of the core promoter, downstream of the core promoter but before a translational start site for a protein of interest (e.g., within a 5′ UTR, etc.) or a cloning site, proximal to the 5′ end of the optionally present TATA box when present, or any combination thereof, a repressible promoter of the present disclosure comprises at least one copy of a binding element having a nucleic acid sequence capable of specifically binding the DNA binding domain of its cognate synthetic repressor. While binding elements are placed proximal to the 5′ end of the promoter in various embodiments disclosed in the Examples, the present disclosure contemplates one or more binding elements at various positions upstream of a core promoter at greater distances (e.g., about 20 bp, 40 bp, 60 bp, 80 bp, 100 bp or more). The presence of one of more binding elements creates a repressible promoter from the constitutive core promoter. Tunable expression from the promoter can be achieved by independently varying the number of different binding elements, the copy number of a binding element at a given position, and/or the spacing between binding elements at a given position.
In some embodiments, a synthetic repressor promoter construct may comprise 2 or more different types of binding elements. For example, a synthetic repressor promoter construct may comprise 2, 3, 4, 5, 6, 7, 8, 9, 10, or more different types of binding elements. In certain embodiments, a synthetic repressor promoter construct may comprise 2 or more recognition sequences. Alternatively or in addition, a synthetic repressor promoter construct may comprise 2 or more copies of any given binding element. For example, a synthetic repressor promoter construct may comprise 2, 3, 4, 5, 6, 7, 8, 9, 10, or more copies of a binding element.
The total number of binding elements in a synthetic repressor construct can and will vary depending upon the level of repression desired and the type of binding element(s). Generally speaking, increasing the number of binding elements will result in greater repression. However, the number of binding elements should not substantially weaken the constitutive activity of the core promoter in the absence of repressor. For illustrative purposes only, if about 6-10 copies of a binding element that is 20 nucleotides in length begins to adversely weaken a core promoter in the absence of a repressor, then a similar effect may be seen with about 3-5 copies of a binding element that is 40 nucleotides in length. In some embodiments, a synthetic repressor promoter construct may comprise at least 2 and no more than 10 copies of all binding elements. In still other embodiments, a synthetic repressor promoter construct may comprise at least 2 and no more than 8 copies of all binding elements. In still other embodiments, a synthetic repressor promoter construct may comprise at least 2 and no more than 6 copies of all binding elements. In each of the above embodiments, the binding elements may be the same or different.
In embodiments that comprise 2 or more copies of a binding element, or 2 or more different binding elements, the binding elements may be at the same position or at different positions. As a non-limiting example, if a synthetic repressor promoter construct comprises a core promoter, a TATA box and 2 copies of a binding element, then (1) both copies may be at a single position selected from upstream of the core promoter, downstream of the core promoter but before a translational start site for the protein of interest/cloning site, or proximal to the 5′ end of the TATA box; (2) one copy may be upstream of the core promoter and one copy may be downstream of the core promoter but before a translational start site for the protein of interest/cloning site; (3) one copy may be upstream of the core promoter and one copy may be proximal to the 5′ end of the TATA box; or (4) one copy may be downstream of the core promoter but before a translational start site for the protein of interest/cloning site and one copy may be proximal to the 5′ end of the TATA box. As another non-limiting example, if a synthetic repressor promoter construct comprises a core promoter, a TATA box and 1 copy each of two different binding elements, then (1) both binding elements may be at a single position selected from upstream of the core promoter, downstream of the core promoter but before a translational start site for the protein of interest/cloning site, or proximal to the 5′ end of the TATA box; (2) one binding element may be upstream of the core promoter and one binding element may be downstream of the core promoter but before a translational start site for the protein of interest/cloning site; (3) one binding element may be upstream of the core promoter and one binding element may be proximal to the 5′ end of the TATA box; or (4) one binding element may be downstream of the core promoter but before a translational start site for the protein of interest/cloning site and one binding element may be proximal to the 5′ end of the TATA box.
Alternatively or in addition to the above, in embodiments that comprise 2 or more copies of a binding element at any one position (or 2 or more different binding elements at any one position), the binding elements may be separated from each other by a nucleic acid spacer sequence of about 2 to about 20 nucleotides, preferably about 2 to about 10 nucleotides.
In an exemplary embodiment, a synthetic repressible promoter construct comprises (a) a nucleic acid sequence encoding a core promoter region capable of conferring constitutive gene expression in a plant species, wherein the core promoter region is selected from the group consisting of Cauliflower Mosaic Virus (CaMV35S) promoter, Figwort Mosaic Virus (FMV) promoter, Nopaline Synthase (NOS) promoter, Ubiquitin-1 promoter from maize (ZmUBI1), and Actin 2.1 promoter from rice (OsACT2.1); and (b) a synthetic regulatory element comprising at least two copies of recognition sequence for a bacterial LexA DNA-binding domain or at least two copies of a recognition sequence for a yeast Gal4 DNA-binding domain, wherein the two or more copies of the recognition sequence is inserted at a position upstream of the core promoter, downstream of the core promoter but before a translation start site for a protein of interest or a cloning site, or proximal to the 5′ end of the optionally present TATA box when present. In another exemplary embodiment, a synthetic repressor construct comprises a nucleic acid selected from the group consisting of SEQ ID NO: 27-34.
Synthetic repressible promoter constructs of the present disclosure may further comprise a polynucleotide encoding additional genetic elements including, but not limited to, terminators, translational regulatory sequences such as ribosome binding sites and internal ribosome entry sites, enhancers, silencers, insulators or other boundary elements, polyadenylation signals, locus control regions, etc.
(c) Synthetic Genetic Circuit
In another aspect, the present disclosure provides a synthetic genetic circuit for modifying expression of a protein of interest in a plant cell. A genetic circuit, as used herein, refers to a set of expression cassettes (genes) that interact to produce an output. The output of a genetic circuit is typically controlled, or “tuned”, through inducible transcription factors and cis-regulatory elements. For example, in operation, an input signal may trigger the expression of, or otherwise alter the generation of, a product from a first expression cassette. The first product may then directly or indirectly alter the expression of a product encoded for by another part of the genetic circuit (e.g., a second expression cassette).
The synthetic genetic circuit of the present disclosure comprises: (a) a nucleic acid construct comprising a nucleic acid encoding a protein of interest; (b) a synthetic repressible promoter construct operably linked to the nucleic acid encoding the protein of interest; and (c) a promoter operably linked to a synthetic repressor construct. Suitable synthetic repressor constructs are described in Section I(a) and suitable synthetic repressible promoter constructs are described in Section I(b). In all embodiments, the transcriptional repressor domain and the DNA-binding domain of the synthetic repressor are operable in the same plant species as the promoter operably linked to the synthetic repressor construct and the core promoter of the synthetic repressible construct.
By assembling genetic circuits as detailed herein, a range of quantitatively distinct responses can be produced. How well a synthetic repressor actually works on a quantitative level is measurable and allows for the assembling of components into digital-like systems. For example, using synthetic repressors allows one to produce a NOT gate. NOT gates can be combined to produce NOR gates. NOR gates are Boolean complete, meaning that any computational function can be designed (e.g. any LOGIC function found in electronic circuits, i.e., design circuit boards). For example, see, Brophy et al. (2014) Nat Methods, 11: 508-520; Moon et al. (2012) Nature, 491:249-253; Tasmir et al. (2011) Nature, 469: 212-215.
In an exemplary embodiment, the transcriptional repressor domain of the synthetic repressor is an OFP transcriptional repressor domain and the DNA-binding domain of the synthetic repressor is a bacterial LexA DNA-binding domain. In further embodiments, the core promoter is cauliflower mosaic virus 35S promoter. In still further embodiments, the OFP transcriptional repressor domain is SEQ ID NO: 35.
In another exemplary embodiment, the transcriptional repressor domain of the synthetic repressor is a BRD transcriptional repressor domain and the DNA-binding domain of the synthetic repressor is a yeast Gal4 DNA-binding domain.
In each of the above embodiments, the promoter operably linked to the synthetic repressor construct can be a constitutive promoter or, preferably, an inducible promoter. In embodiments where an inducible promoter is operably linked to a synthetic repressor construct, the synthetic repressor produced therefrom is conditionally expressed such that the repressor protein is only be available to the synthetic repressible promoter under certain conditions and, therefore, drive expression of the protein of interest under those same conditions. Suitable inducible promoters are described in Section I(a). In some embodiments, the inducible promoter confers gene expression only in one tissue or cell type of the plant. In other embodiments, the inducible promoter confers gene expression in one or more cell types of the plant. In other embodiments, the inducible promoter confers gene expression in one or more tissues of the plant. In other embodiments, the inducible promoter confers gene expression in all cell types of the plant.
Any tissue-specific or tissue-preferred promoter can be utilized in the instant invention. Exemplary tissue-specific or tissue-preferred promoters include, but are not limited to, a seed-preferred promoter such as that from the phaseolin gene; a leaf-specific and light-induced promoter such as that from cab or rubisco; an anther-specific promoter such as that from LAT52; a pollen specific promoter such as that from Zm13 or a microspore-preferred promoter such as that from apg.
In further embodiments, the protein of interest negatively regulates the inducible promoter operably linked to the synthetic repressor construct.
Transport of protein produced by transgenes to a subcellular compartment such as the chloroplast, vacuole, peroxisome, glyoxysome, cell wall or mitochondrion, or for secretion into the apoplast, is accomplished by means of operably linking the nucleotide sequence encoding a signal sequence to the 5′ and/or 3′ region of a gene encoding the protein of interest. Targeting sequences at the 5′ and/or 3′ end of the structural gene may determine, during protein synthesis and processing, where the encoded protein is ultimately compartmentalized. The presence of a signal sequence directs a polypeptide to either an intracellular organelle or subcellular compartment or for secretion to the apoplast. Any signal sequence known in the art is contemplated by the present invention.
A further aspect of the present disclosure encompasses kits comprising any one of the constructs, or synthetic genetic circuits, detailed above.
In some embodiments, a kit comprises a synthetic repressor construct of Section I(a) and/or a synthetic repressible promoter construct of Section I(b).
In other embodiments, a kit comprises a vector comprising a promoter operably linked to a synthetic repressor construct of Section I(a) and/or a vector comprising a synthetic repressible promoter construct of Section I(b) and cloning site proximal to the 3′ end of a synthetic repressible promoter construct.
In other embodiments, a kit comprises a vector comprising (a) a promoter operably linked to a synthetic repressor construct of Section I(a) and (b) a synthetic repressible promoter construct of Section I(b) and cloning site proximal to the 3′ end of a synthetic repressible promoter construct.
In other embodiments, a kit comprises a plurality of synthetic genetic circuits of Section I(c), wherein each synthetic genetic circuit of the kit varies from the other synthetic genetic circuits in the number of binding elements at a given position and/or the spacing between the binding elements at a given position.
In each of the above embodiments, a kit can further comprise cells competent for transformation or transfection, transformation or transfection reagents, restriction enzymes, inducers, buffers, and the like.
In further embodiments, a kit can also comprise a construct comprising an inducible promoter operably linked to a reporter. The inducible promoter can be same as, or different than, the inducible promoter operably linked to the synthetic repressor construct. Non-limiting examples of suitable reporters include fluorescent proteins, purification tags, epitope tags, and the like. Exemplary fluorescent proteins, purification tags, epitope tags are described in Section I(a).
A further aspect of the present disclosure encompasses a transgenic plant cell comprising a synthetic repressor construct of Section I(a). In another aspect, the present disclosure provides a transgenic plant cell comprising a synthetic repressible promoter construct of Section I(b). In another aspect, the present disclosure provides a transgenic plant cell comprising a synthetic genetic circuit of Section I(c). The plant cell can been an isolated plant cell, a cell in a whole plant, or cell in plant structure (e.g., seed, flower, fruit, etc.). In various embodiments, the plant cell can be a parenchyma cell, a collenchyma cell, a sclerenchyma cell, a xylem cell, or a phloem cell.
In each of the above aspects, the transgenic plant cell can be derived from a monocot or a dicot. In certain embodiments, the plant cell is crop plant cell. Non-limiting examples of crop plants include grain crops (e.g., rice, Jowar, wheat, maize, barley, millets, etc.), pulse/legume crops (e.g., green gram, black gram, soybean, pea, cowpea, etc.), oil seed crops (e.g., groundut, mustard, sunflower, sesamum, linseed, etc.), forage crops (e.g., fickler, hay, silage, etc.), fiber crops (e.g., cotton, steam, jute, mesta, sun hemp, etc.), root crops (e.g., sugar beet, carrots, turnips, etc.), tuber crops (e.g., potato, yam, etc.), sugar crops (e.g., sugarcane, sugar beet, etc.), vegetable crops, green manure crops, medicinal and aromatic crops (e.g., cinchona, isabgoli, opium poppy, senna, belladonna, rauwolfra, iycorice, lemon grass, citronella grass, palmorsa, Japanese mint, peppermint, rose geranicem, jasmine, henna etc.).
In embodiments encompassing whole plants or plant structures, one or more of the cell types may comprise a synthetic repressor construct of Section I(a), a synthetic repressible promoter construct of Section I(b), or The synthetic genetic circuit of Section I(c). In other embodiments encompassing whole plants or plant structures, one or more of the tissue types may be comprised of a transgenic plant cell comprising a synthetic repressor construct of Section I(a), a synthetic repressible promoter construct of Section I(b), or The synthetic genetic circuit of Section I(c).
Methods for transiently or stably transforming plant cells are well known in the art. For example, numerous methods for plant transformation have been developed, including biological and physical, plant transformation protocols. See, for example, Miki et al., “Procedures for Introducing Foreign DNA into Plants” in Methods in Plant Molecular Biology and Biotechnology, Glick, B. R. and Thompson, J. E. Eds. (CRC Press, Inc., Boca Raton, 1993) pages 67-88. In addition, expression vectors and in vitro culture methods for plant cell or tissue transformation and regeneration of plants are available. See, for example, Gruber et al., “Vectors for Plant Transformation” in Methods in Plant Molecular Biology and Biotechnology, Glick, B. R. and Thompson, J. E. Eds. (CRC Press, inc., Boca Raton, 1993) pages 89-119. Additional methods are further detailed in the examples.
Synthetic genetic circuits disclosed herein can be used to engineer transgenic plants that express various phenotypes of agronomic interest, or to engineer transgenic plants for recombinant protein production.
In some embodiments, the nucleic acid encoding a protein of interest is a gene that confers resistance to pests or disease including, but not limited to, those that encode: (a) plant disease resistance genes; (b) a lectin; (c) a vitamin-binding protein; (d) an enzyme inhibitor; (e) an insect-specific hormone or pheromone, mimetic based thereon, antagonist or agonist thereof; (f) an insect-specific peptide or neuropeptide which, upon expression, disrupts the physiology of the affected pest; (g) an insect-specific venom produced in nature by a snake, a wasp, etc.; (h) an enzyme responsible for a hyperaccumulation of a monterpene, a sesquiterpene, a steroid, hydroxamic acid, a phenylpropanoid derivative or another non-protein molecule with insecticidal activity; (i) an enzyme involved in the modification, including the post-translational modification, of a biologically active molecule, for example, a glycolytic enzyme, a proteolytic enzyme, a lipolytic enzyme, a nuclease, a cyclase, a transaminase, an esterase, a hydrolase, a phosphatase, a kinase, a phosphorylase, a polymerase, an elastase, a chitinase and a glucanase, whether natural or synthetic; (j) a molecule that stimulates signal transduction; (k) a hydrophobic moment peptide; (l) a membrane permease, a channel former or a channel blocker; (m) a viral-invasive protein or a complex toxin derived therefrom; (n) an insect-specific antibody or an immunotoxin derived therefrom; (o) a virus-specific antibody; (p) a developmental-arrestive protein produced in nature by a pathogen or a parasite; or (q) a developmental-arrestive protein produced in nature by a plant.
In other embodiments, the nucleic acid encoding a protein of interest is a gene that confers resistance to herbicides including, but not limited to: (a) a herbicide that inhibits the growing point or meristem, such as an imidazalinone or a sulfonylurea; (b) glyphosate (resistance imparted by mutant 5-enolpyruvl-3-phosphikimate synthase (EPSP) and aroA genes, respectively) and other phosphono compounds such as glufosinate (phosphinothricin acetyl transferase (PAT) and Streptomyces hygroscopicus phosphinothricin acetyl transferase (bar) genes), and pyridinoxy or phenoxy proprionic acids and cyclohexones (ACCase inhibitor-encoding genes); or (c) a herbicide that inhibits photosynthesis, such as a triazine (psbA and gs+ genes) and a benzonitrile (nitrilase gene).
In other embodiments, the nucleic acid encoding a protein of interest is a gene that confers or contributes to a value-added trait including, but not limited to (i) introduction of a phytase-encoding gene would enhance breakdown of phytate, adding more free phosphate to the transformed plant; (ii) a gene could be introduced that reduces phytate content; or (iii) modified carbohydrate composition effected, for example, by transforming plants with a gene coding for an enzyme that alters the branching pattern of starch.
In other embodiments, the nucleic acid encoding a protein of interest is a gene that encodes or produces a therapeutic such as an antibody, a hormone, etc.
In another aspect, the present disclosure provides a method for modifying expression of a protein of interest in a plant. The method comprises introducing a synthetic genetic circuit of Section I(c) into a cell of the plant, wherein the promoter operably linked to the synthetic repressor construct is an inducible promoter. In some embodiments, the method further comprises varying the level of expression of the protein of interest across different cell types or tissues types in the plant by varying the number of binding elements at a given position and/or the spacing between the 2 or more binding elements at a given position.
In another aspect, the present disclosure provides a method for creating a library comprising a plurality of synthetic repressible promoter constructs. The method comprises providing a construct comprising a core promoter capable of conferring constitutive gene expression in a plant species, wherein the promoter optionally comprises a TATA box, and modifying the construct a plurality of times by (i) introducing one or more copies of a binding element having a nucleic acid sequence capable of specifically binding a DNA-binding domain, the copy of the binding element inserted at a position upstream of the core promoter, downstream of the core promoter but before a translation start site for a protein of interest, or proximal to the 5′ end of the optionally present TATA box when present, and then (ii) varying the number of binding elements at a given position and/or the spacing between the 2 or more binding elements at a given position. In various embodiments, the core plant promoter is operably linked to a nucleic acid encoding a protein of interest. Alternatively, or in addition, the 3′ end of the core promoter can be proximal to a cloning site. In further embodiments, the construct comprising a core promoter capable of conferring constitutive gene expression in a plant species is provided in the form of a vector. In still further embodiments, the vector further comprises a synthetic repressor construct, the synthetic repressor construct comprising a promoter operably linked to a nucleic acid encoding a transcriptional repressor domain and a DNA-binding domain; wherein the DNA-binding domain specifically binds to a binding element of the synthetic repressible promoter construct, and wherein the transcriptional repressor domain of the synthetic repressor, the DNA-binding domain of the synthetic repressor, the promoter operably linked to the synthetic repressor construct, and the core promoter of the synthetic repressible construct are each operable in the same plant species.
In another aspect, the present disclosure provides a method for selecting in vitro a synthetic gene circuit for expression in a plant. Generally speaking the method comprises (a) introducing a plurality of synthetic genetic circuits into isolated plant cells to produce a plurality of isolated plant cells that have only one type of synthetic genetic circuit, each synthetic genetic circuit comprises (i) a nucleic acid construct comprising a nucleic acid encoding a first reporter; (ii) a synthetic repressible promoter construct operably linked to the nucleic acid encoding the first reporter; and (iii) a promoter operably linked to a synthetic repressor construct; (b) further introducing into each of the isolated plant cells a construct comprising an inducible promoter operably linked to a second reporter, wherein the inducible promoter of the construct is the same as the inducible promoter of the synthetic genetic circuits; (c) for each plant cell produced from step (b), adding inducer to an in vitro culture of the plant cell, culturing the plant cell for a sufficient amount of time, measuring the amount of each reporter, and determining a quantitative parameter of the synthetic genetic circuit; and (d) selecting a synthetic genetic circuit based on the quantitative parameter determined in step (c). In various embodiments, the order of steps (a) and (b) may be interchanged. Synthetic repressor constructs are detailed in Section I(a) and synthetic repressible promoter constructs are detailed in Section I(b). The transcriptional repressor domain and the DNA-binding domain of the synthetic repressor are operable in the same plant species as the promoter operably linked to the synthetic repressor construct and the core promoter of the synthetic repressible construct. Further, each synthetic genetic circuit varies from the other synthetic genetic circuits in the number of binding elements at a given position and/or the spacing between the binding elements at a given position.
Methods for extracting quantitative parameters of synthetic genetic circuits are known in the art, and further detailed in the examples.
The following examples are included to demonstrate preferred embodiments of the invention. It should be appreciated by those of skill in the art that the techniques disclosed in the examples that follow represent techniques discovered by the inventors to function well in the practice of the invention. Those of skill in the art should, however, in light of the present disclosure, appreciate that changes may be made in the specific embodiments that are disclosed and still obtain a like or similar result without departing from the spirit and scope of the invention. Therefore, all matter set forth or shown in the accompanying drawings is to be interpreted as illustrative and not in a limiting sense.
The following examples illustrate various iterations of the invention.
Synthetic Biology promises to bring new understanding of living organisms while allowing the design of predictable biological function. Yet to date all quantitatively defined genetic circuits have been produced in unicellular organisms (bacteria, yeast) or mammalian cells in culture1-4. This raises a question as to whether synthetic genetic circuits with predictable function can be produced in multicellular organisms. In multicellular organisms, sexual reproduction proceeds through meiosis, and in plants this is accompanied by further development into gametophyte and sporophyte stages, potentially affecting the stability and function of synthetic genetic elements in the genome. Even so, genetic circuits with highly predictable function in plants could have profound applications towards sustainable life on earth. For example, such circuits could be used to control biofuel production so that optimal biomass is produced prior to the trait induction. Moreover, it is unlikely that natural production of biomaterials is optimum. For example, cotton fibers are only produced from ovule epidermal cells rather than the considerably more abundant epidermal cells of leaves.
The ability to design and produce predictable genetic circuits in plants requires a deep understanding of plant biology and rigorous quantitative data. The latter requirement imposes a significant challenge that lead some to believe that quantitative predictable function of synthetic genetic circuits in plants is unattainable. The concerns are myriad. Plants develop continuously, move regulatory molecules between cells and tissues, and control much of their differentiation by positional information with input from their local environment5. Plant epigenetics, known to affect many processes, is not fully understood6. Hence, a predictable genetic circuit should function independent of inputs from endogenous genetic and epigenetic regulators, positional information, and the environment.
The standardized parts-based approach used in synthetic biology requires accurate quantification of the dynamic behavior of each part in vivo in order to rationally and predictably design higher-level complex circuits. Therefore, the first step towards producing predictable function in higher-level circuits is quantification of the input-output characteristics of genetic parts.
Quantitative analysis of a large number of stably integrated genetic parts (e.g., promoters, terminators, UTRs) in plants would require decades of efforts, even for fast growing species such as Arabidopsis. Current methods for rapid transient gene expression in higher plants, such as particle bombardment, Agrobacterium infiltration, VIGS (virus-induced gene silencing)-based systems, and protoplasts7-10 could be used for parts characterization. However, particle bombardment and VIGS do not easily allow high throughput analysis. Agrobacterium infiltration methods, including AGROBEST11, could be scaled up to test a large number of parts, but quantification of function is difficult. Transient expression in plant leaf protoplasts has been scaled up for use with FACS12, yet quantitative data are also difficult to obtain due to the large auto-fluorescence signal from abundant chlorophyll, which typically overlaps with signal from the fluorescent proteins used as readouts.
To overcome these limitations, we developed an enhanced throughput transient expression assay in plant protoplasts, using luciferase outputs, to rapidly test the behavior of plant genetic parts. We combine this assay with a rigorous mathematical analysis to account for significant stochastic factors, allowing quantitative analyses of plant parts function. Here, we describe this methodology and demonstrate its use to quantitatively characterize 128 new synthetic plant promoter-repressor pairs. We further show that these parts can be computationally selected and then assembled to produce predictable function in planta.
Orthogonality, i.e., the ability of a genetic component to function in an organism independent of endogenous regulation, is an important principle of synthetic biology13,14. To apply this principle, genetic components were chosen with some prior characterization from bacteria, yeast, and plant viruses as sources for engineering synthetic transcriptional repressor proteins and repressible promoters for plants. Synthetic plant transcriptional repressors were created by making translational fusions of the DNA binding domain from the yeast Gal4 or the bacterial LexA transcription factors to previously characterized repression domains found in Arabidopsis repressor proteins (EAR, OFP, BRD)15-18. Synthetic plant repressible promoters were engineered from promoters naturally found in plant viral and bacterial pathogens [Cauliflower Mosaic Virus (CaMV35S), Figwort Mosaic Virus (FMV), and Nopaline Synthase (NOS)], which have been shown to drive constitutive expression of genes in plants19-21. Synthetic promoters were engineered to be both constitutive and repressible in plants by inserting multiple copies of the Gal4 or LexA binding elements (recognized by the above repressors) at specific positions in the CaMV35S, FMV, and NOS promoters.
The transcriptional repressor proteins are built out of two genetic components: DNA binding (DB) and repressor domains (RD). The DNA binding domains of the yeast Gal4 and the bacterial LexA transcription factors were used to create orthogonal repressor proteins for the plant system. Ethylene-responsive element binding factor-associated amphiphilic repression (EAR), plant-specific B3 repression domain (BRD), two variants of the Arabidopsis OVATE Family proteins (AtOFP1 and AtOFPx) were used as repressor domains. AtOFPx represents a consensus sequence of the OVATE family repressor proteins demonstrating the highest levels of repression16. Sequence optimized Gal4 and LexA DB, and the two OVATE RD were synthesized as double stranded gBlocks (GeneArt/Life Technologies and IDT). The synthetic repressor domains were fused in frame to one of two mentioned synthetic DNA binding domains using overlapping extension PCR with compatible BsaI restriction enzyme sites built into the primers for downstream cloning. The small-sized EAR and B3 repressor domains were incorporated into reverse primers used to amplify DNA binding domains, creating in-frame C-terminal fusions. The hybrid products were sub-cloned and sequenced in pJET2.1 vector using pJET forward and reverse primers [Thermo Scientific]. Two core repressor modules containing an upstream transcription block25, estrogen inducible promoters, repressors and NOS terminator26 were synthesized (GeneArt,
The constitutively active repressible promoters were constructed by introducing DNA binding elements (operators) in the backbone of Cauliflower Mosaic Virus 35S (CaMV35S), Nopaline Synthase (NOS) and Figwort Mosaic Virus (FMV) promoters20. The DNA binding elements containing two copies of Gal4, and two or eight copies of LexA, were synthesized as a gene block with appropriate restriction sites included (IDT). A library of repressible promoters was generated by varying the number of DNA binding elements, the spacing between each binding element and its position relative to the transcription start site. Plasmid backbone (
Two pBluescript SK+ backbone plasmids (
To quantitatively measure the input-output function of the synthetic repressible promoters, we constructed a test bed by linking each synthetic promoter to Renilla luciferase (R-luc) to provide a direct readout of the promoter's quantitative behavior (
Repressible promoters and their cognate repressors were cloned into a single plasmid and transiently expressed in Arabidopsis leaf protoplasts (
Luminescence of both F-luc and R-luc was measured using a single photon ICCD Camera (Stanford Photonics, Inc.). With increasing inducer concentrations increasing levels of F-luc (input) was observed coupled with decreasing levels of R-luc (output). Initial results from the protoplast experiments showed large variability (noise) in the data (
Luminescence values are typically reported in relative luciferase units (RLUs) within an area over collection time (RLU/area*s). To estimate luciferase activity in physical units, assays with purified recombinant F-luc and R-luc were used to quantify the relation between the luminescence and the number of luciferase molecules (
There are several possible sources for the variability we observe in
Before rigorous mathematical models for gene function could be developed, experiments were designed to isolate these three potential sources of noise and determine their actual experimental magnitude. A plasmid that contains all elements found in the promoter-repressor test system were used, except the repressor, and prepared protoplast transformations with one DEX-inducible gene circuit and one OHT-inducible gene circuit. Luminescence data was collected with no inducer added, and the experiment was repeated on three different days. In the absence of any noise, all wells should display identical F-luc and identical R-luc luminescence.
The most apparent source of noise in the data originates from distinct batches of protoplasts prepared on different days (
While the data clearly show that variation from preparation of protoplast batches is the greatest source of noise, the origin of this noise has not been precisely identified. Because the protoplasts are prepared from leaves, each day's preparation likely contains distinct compositions of differentiated cells within a leaf (e.g., mesophyll, palisade parenchyma, bundle sheath) produced from plants that experience micro-climatic variations within our growth chambers. While all protoplasts are pooled and treated equally, the data represent a bulk measurement of all protoplasts in an individual well. As such, different protoplast cell populations may be represented in each day's preparations, giving rise to the random batch effect. Based on the above analysis, a quantitative model of the protoplast data to test the input-output characteristics of the repressor-promoter pairs was constructed.
To develop a model, the experimental data were first defined with mathematical symbols as follows. In all cases the indices ij refer to the j-th well of the i-th plasmid. Concentrations are in molecules per well and RLU units are RLU/(area*sec).
1. Concentration of repressor=Rij
4. R-luc luminescence in RLU units=Lij
5. F-luc luminescence in RLU units=Fij
It was desired to determine the quantitative relationship between the concentration of the repressor and the expression of R-luc as driven by the constitutively active repressible promoter. It was assumed this relationship is represented by a Hill function3,24. Hence, for a single plasmid in a protoplast:
Here βi represents the maximal expression of the R-luc protein in the absence of the repressor, while K is the concentration of repressor required for half-maximal expression of R-luc, and n is the Hill coefficient. We assume that every protoplast batch has associated with it a batch effect multiplicative factor, α. Further, we assume that each well has Nij plasmids. Then the total R-luc luminescence in the j-th well of the i-th transformation is related to the concentration of R-luc by Lij=C2αiNij[Rluc]ij where C2 is the slope of the R-luc standard curve.
Despite being controlled by the same promoter, the concentrations of F-luc and the repressor are not necessarily identical, but should be linearly proportional to each other. For a single plasmid therefore we can write [Fluc]={tilde over (C)}R, where {tilde over (C)} is the proportionality constant. Generalizing again to the j-th well of the i-th transformation, the F-luc luminescence can be expressed as: Fij=C1αiNij{tilde over (C)}Rij, where C1 is the slope of the F-luc standard curve.
Now a Hill functional form was fit between the variables Lij and Fij. This function is written as
The parameter Hi represents the half-maximal whole-well R-luc luminescence, while the parameter Bi represents the whole-well R-luc luminescence of the i-th plasmid. In terms of the parameters for a single plasmid therefore, a best-fit estimate of Bi is given by Bi=C2Nij
αiβi where the angled brackets indicate a mean over the j-index, i.e., over the wells in each plasmid. Similarly, Hi=C1αi
Nij
{tilde over (C)}Ki.
Though the estimated parameters are proportional to the single promoter parameters β and K, they also contain the unknown multiplicative batch effect term α, which complicates comparison of the repressible promoter strength β, between plasmids. Thus, we considered normalization methods that remove or reduce the effect of this parameter on our data. We first tested the possibility that the batch effect can be removed by normalization with the total protein content of the wells.
It was hypothesized that the factor controlling the variation from different batches of protoplasts, α, is related to the overall biological health of the protoplasts. Therefore, the term αiNij describes the number of plasmids in viable protoplasts in the j-th well of the i-th transformation. We define γi as the basal expression of the inducible promoter, which is a constant since the inducible promoter that controls F-luciferase is exactly the same on all gene circuits (i.e., the same for all gene circuits induced by OHT and the same for all gene circuits induced by DEX). Hence, the luminescence of F-luc without inducer, Fij0, can be described by Fij0=C1αiNij0γi and the distribution of its values for all i and j is proportional to the distribution of αiNij0, where the 0 superscript refers to the wells with zero inducer.
Here it is assumed that the distribution of Nij and αi are independent of each other and that the distribution of Ni0 is sharply peaked, so that Ni0
≈
Ni0
ij. Next, we divide every value of F-luc and R-luc luminescence for the i-th plasmid by the normalization factor λi, which replaces the batch effect factor αi by its mean
αi
. The variance in our data due to batch effects is therefore expected to vanish, and is replaced by a constant.
Fitting the relevant non-linear function to the normalized data now estimates Bi=C2Nij
αi
βi, Hi=C1
αi
Nij
{tilde over (C)}Ki and the Hill coefficient n. This procedure allows us to rank the different promoters and compare them. The single-plasmid parameters βi (promoter strength) and Ki (the repressor concentration at half-maximum) could also be estimated if required by measuring the value of the different constants in the expressions above.
Tests of the normalization method against simulated data (
We fit the normalized data using the nonlinear least squares package in Matlab. We tested alternative methods of fitting but found that they did not improve analysis and predictability (data not shown). We also implemented a number of quality control steps to remove assays that appeared to have failed, or promoters that did not work properly (detailed in Example 12). From about 120 gene circuits tested in Arabidopsis protoplasts, about 20 of them met all the criteria. Out of these 20 we selected only those promoter-repressor pairs whose fits had Hill coefficients that were significantly different from zero with a p-value of 0.1 or less for further use.
Protoplast assays are useful only if they provide quantitative data that is a reliable measure of the circuit performance in stably transformed plants. To test whether our predictions are a reliable guide to the performance in transgenic plants, we compared the predictions of repressor characteristics in the transient assay with their performance in stably transformed Arabidopsis μlants. Protoplasts were prepared from both wild-type Arabidopsis μlants and transgenic plants stably transformed with a single copy of the repressible promoter circuit described above. Wild-type protoplasts were transiently transformed with the plasmids containing the circuits and induced with increasing concentrations of OHT or DEX. Transgenic protoplasts were also induced with OHT or DEX. F-luc and R-luc luminescence were measured, and parameters were derived using the mathematical analysis described above for both types of protoplasts. The data were normalized using a slightly different method in order to accurately compare the parameters obtained from stably transformed plants with those obtained from transient assays, since the number of working circuits each protoplast contains is significantly different in the two cases (
Comparison of the parameters from the normalized data for the stable transgenic plants and for the transient assay are shown in
Arabidopsis protoplast isolation and transfection were carried out according to the protocol described by Yoo et. al.31, with some modifications to allow higher throughput testing of synthetic components in 96-well plates. Wild type Columbia plants were grown in short days (10 h light/14 h dark), and 20-25 leaves, approximately 4 cm in length, were used. In brief, leaves in W5 solution were cut into ˜1 mm strips using a scalpel blade. Enzyme solution [0.4 M Mannitol, 20 mM KCl, 20 mM MES (pH 5.7), 1.5% Cellulase R-10 (Yakult Honsha), 0.4% Macerozyme R-10 (Yakult Honsha), 10 mM CaCl2, 1 mg/ml BSA] was added, a slight vacuum was applied, and incubated at room temperature with gentle shaking (40 rpm) for 3 hours. Resulting protoplasts were filtered through a 70 μm cell strainer (BD Biosciences) and harvested by centrifugation at 600×g. After two washes in W5 solution, the protoplasts were resuspended in MMg solution, and the concentration adjusted to 2×105 protoplasts/ml. Protoplast transfection with plasmids of interest was performed in 15-ml conical centrifuge tubes by carefully mixing 50 μl of protoplasts (approx. 10,000 cells), 5 μl of plasmid DNA (1 μg/μl), and 55 μl of 40% PEG solution for one reaction. Larger-scale (14 reactions) transfections were used to allow testing of multiple concentrations of inducers. Transfected protoplasts were resuspended in 200 μl of WI solution/reaction, and plated on black, clear-bottom, 96-well Costar assay plates (Corning), using a multi-channel pipette. Inducers (4-OHT or dexamethasone) were added also using a multi-channel pipette, and plates were incubated overnight in the dark, with gentle agitation (50 rpm).
All test plasmids used in this work had Firefly and Renilla luciferases as the measurable outputs. Therefore, the Dual-Luciferase Reporter Assay system (Promega) was used to lyse the protoplasts and provide both substrates for luciferase imaging. After overnight incubation of protoplasts with the inducers, cell lysis was carried out on the assay plates by removing 160 μl of supernatant from each well, followed by addition of 50 μl of 2× Passive Lysis Buffer, and incubation at room temperature for 30 minutes. Quantitative measurements of Firefly and Renilla Luciferase expression were obtained by the addition of LAR II and Stop & Glo reagents, respectively, and imaged using a Stanford Photonics XR/Mega-10 ICCD Camera System and available Piper software (v. 2.6.17). Regions of interest, ROI's, are drawn around each well of a 96-well plate. Pixel intensity values for the 1st minute of collection time are summed and divided by the area of the ROI and time collected to give us the RLU/(area*s) value in each well. The data then go through post-image correction (below).
Five primary systematical parameter values were first determined: 1) r1 (Radius of well opening); 2) r2 (Radius of well bottom); 3) h1 (Height of the camera relative to the surface of the 96 wells); 4) h2 (Depth of well); 5) V (Reaction reagent total volume). These parameters were then fed into the function V(d) (Example 12) to yield the secondary parameter d (depth of solution added in the well). Next, values of r1, r2 and h1 were substituted into function Av (Example 12) resulting in Av (s,D). Then h2 and d were substituted as the lower and upper integration limits, respectively, for the integration of Av(s)ds, resulting in Vvtotal(D). Here, the function is fully parameterized and the only input needed is D, the positional parameter corresponding to each well in this algorithm. It was assumed the well in the ith row, jth column from the top left corner of the microplate (as the origin of 96-well-plate plane) holds a coordinate of (xij,yij) and the projected camera center onto the plate holds a coordinate of (x,y). Then, Dij for the well (i, j) can be calculated as Dij=√{square root over ((xij−x)2+(yij−y)2)}. Substitute Dij into Vvtotal(D) to generate the total visible volume for well (i, j). The ratio of this total visible volume to V (total reagent volume) is then used for camera correction.
The noise measurements shown in
Data is processed in the following steps using MATLAB. (1) Camera corrected F-luc RLU/(area*s) and R-luc RLU/(area*s) values, and inducer type and concentration are stored in different .csv files for each promoter tested. (2) DEX and OHT data are separated. (3) Fold Change (FC) values are calculated for each promoter. Promoters with a FC>1.3 are stored for further analysis. (4) Data from promoters that do not meet the threshold criteria are tagged and kept for further processing. (5) RLU data are converted to Molecule Number/well via the RLU vs concentration standard curves shown in
The detailed analysis of promoter-repressor pairs in isolated plant cells attempts to determine if gene circuits with predictable behavior can be produced in multicellular eukaryotic organisms such as plants. The data in the examples were obtained from protoplast assays, where the levels of synthetic repressors were varied by using external inducers, and the quantitative behavior of the synthetic promoters was determined by measuring R-luc luminescence. The level of the repressor protein was measured by controlling F-luc by the same promoter. Readouts in the form of R-luc and F-luc luminescence provided quantitative measures of the promoter-repressor circuit.
The results show that quantitative data obtained from a rapid transient protoplast assay, when combined with rigorous analysis of noise and mathematical modeling, allows fast and quantitative estimation of the parameters of synthetic gene parts. The data show that it is possible to reliably assess repressor strength using the suite of experimental methods presented herein, and these quantitative measures stand up to experimental analysis in plants. The results support the mathematical model as a reasonable depiction of the experiment, suggesting that further direct measurements of the unknown terms in the equations could significantly reduce model uncertainties. The procedures described here therefore are immediately applicable for the development of comprehensive quantitatively characterized libraries of synthetic plant gene parts. The quantitative parameters of each promoter-repressor pair can be then used for in silico testing of the suitability of its use in more complex genetic circuits, such as a genetic toggle switch.
1. Luminescence Imaging Correction:
False-colored images of the 96 well plates with luminescing protoplasts appeared to show a systematic difference based on well position in the plate. To measure the extent of this difference we designed a simple “flip-plate” experiment. Two protoplast assays with 200 ul per well were collected for 5 minutes with well A1 in the top left hand corner and again for 5 minutes with well A1 in the bottom right hand corner. Only the first minute of each collection time was used to calculate the Relative Luminescence Units (RLU) for each well. We then calculated the percent change of F-Luc (in RLU/(area*s)) between the two values for each well. We repeated the experiment using purified recombinant F-Luc protein diluted to be in the range of our protoplast data. In both cases, we found significant differences between the measured luminescence of the two positions for the outer wells. The graph in
Since we image for five minutes (though only use the first minute of data for comparisons) we need to calculate the possible degradation of the luminescence signal during this time. Our data reveals that the degradation of the F-Luc signal has an average of 7% over a five-minute period (
We theorize that the most likely reason for these systematic imaging errors are that the camera does not pick up as many photons from wells that are farther away from its central axis when compared to wells that are closer. This could be seen from the dark crescents in the images themselves (
Another complication arises when the camera center does not coincide with the center of the 96-well plate. We estimate where the camera center lies from the pattern of percent changes of each well on the plate, since the wells closest to the camera center should have the minimum percent changes (zero if the camera center lies directly above any well). We correct the luminescence data by using this formula to calculate the original luminescence of each well from its position on the plate and the observed intensity. We also built a frame for the 96-well plate that we used for all subsequent imaging in order to keep the plate center in a fixed position in relation to the camera center.
2. Image Correction Method:
The formula derived below for estimating the imaging correction is based on the 96-well plate geometry as shown in
Step 1—Calculate the distance D between the center of the targeting well and projection of the camera center using similar triangles (
Then the upper edge of the well is projected to the bottom along the sight line between the camera and its closest point on the upper edge. 1 is the shift distance on the well bottom of the closest point along the sight line. Then the visible portion of the bottom edge is the part enclosed by the projection of the upper edge and itself. The area of this portion is calculated as follows.
This portion can be separated into two parts (A1 and A2) by the connecting line between the two intersections of the two circles. A1 and A2 can then be calculated using the differences of their corresponding sectors and triangles (
y
1
+y
2
=r
1
−r
2
+l
r
1
2
−y
1
2
=r
2
2
−y
2
2
=x
2
Yields,
With a=r1−r2+l.
Based on these equations, the central angles of these two sectors can be calculated:
The areas of the two sectors can be expressed as: ½α1r12 and ½α2r22.
The two portions of the visible area on this plane can be calculated as:
A
1=½α1r12−y1x
A
2=½α2r22−y2x
The total visible area on the bottom is:
A
v
=A
1
+A
2=½α1r12−y1x+½α2r22−y2x=½α1r12+½α2r22−αx.
To get the visible volume from the visible area, integration is needed from the bottom of the well to the liquid surface. Therefore, it is necessary to calculate the depth of the reagent inside the well. Taking note of the “imaginary cone” as shown in
A
v(s)ds
V
vtotal=∫h
3. Conversion of Luminescence Values to Physical Units:
The function of the promoter-repressor pairs were experimentally characterized using luminescence from two types of luciferase. Luminescence values are typically reported in RLUs, or relative luciferase units. For the collection system (Stanford Photonics ICCD Camera), RLU is the sum of pixel intensity values within an area over collection time RLU/area*s), and represents the activity of F-Luc and R-Luc for each protoplast sample. RLUs were converted to molecules of luciferase by quantifying the relationship between the luminescence and the luciferase activity from purified recombinant F-Luc and R-Luc.
4. Testing the Sources of Noise:
Protoplast transformations were prepared with one DEX-inducible gene circuit and one OHT-inducible gene circuit (enough for 48 wells each). Luminescence data were collected with no inducer added, and repeated the experiment on three different days. In the absence of any noise, all wells should display identical F-Luc and identical R-Luc luminescence, with R-Luc expression at its maximum. Thus, variations between luminescence values from wells containing the same gene circuit on the same plate represent within-plate noise (the first source of noise). The difference between the mean R-Luc luminescence measured from the DEX-inducible gene circuit and the OHT-inducible gene circuit in the same batch represents the between-transformation noise (the second source of noise). Finally, the difference between the mean luminescence of the three batches represents the between-batch noise (the third source of noise)
Because the batch effect is a random variation that affects the entire population of protoplasts in a batch, it can be represented mathematically by a random number α such that the observed luminescence in the j-th well of the i-th batch can be represented by Rij=αiBR+δij and Fij=αiBF+δ′ij, for R-luc and F-luc luminescence, respectively. Here, BR, BF are the steady state number of luciferase molecules in the well in the absence of any noise for the R-luc and F-luc promoters, respectively; αi is a random number that represents a multiplicative batch effect, while δij, δ′ij are random variables that represent additive noise terms that could arise from the remaining noise sources. If the R-luc and the F-luc luminescence are averaged for each batch and plot them, αiBR+δij
j would be plotted against αiBF+
δ′ij
j (where the subscript on the angled brackets indicates the index being averaged). If this plot is approximately linear we can conclude that the batch effect is identical for both R-luc and F-luc, and dominates the additive noise terms.
5. Testing the Normalization Scheme with Simulated Data:
To generate simulated data, single-plasmid data were first generated using Equation 1 with assumed parameter values. Then the single-plasmid data was multiplied by a normally distributed random number representing the number of plasmids in each well (Nij) and another random number drawn from a log normal distribution representing the batch effect factor (αi). The latter was assumed smaller than 1 based the our analysis in the main text.
For simplicity all the constants C1, C2 and {tilde over (C)} were set to 1. 1000 sets of data were simulated, consisting of six inducer levels and two technical replicates, similar to the experimental data. For each set one value of a was chosen from a log normal distribution with a mean less than one. Because the log normal distribution is unbounded in the positive infinity direction, a 95% cut-off for the distribution of a was assumed. To test the normalization scheme with different levels of noise, the standard deviation was increased to obtain a series of distributions with a decreasing population mean and increasing variance of αi. Since each well in the experiments has approximately 10,000 protoplasts, this was set to be the mean of Nij and simulated various levels of noise by changing the standard deviation of the normal distribution.
The fitting procedure produced fits with an unreasonably high Hill coefficient at high levels of noise in the simulated data. We therefore imposed the criteria that the fitted Hill coefficient of the repressible promoter should lie between 1 and 6.
Due to the high levels of noise that ca be artificially generated in the simulated data, there are also “bad fits” within 1<n<6. These can be further characterized by unreasonably high fitted values of B which are far away from the well-formed distribution of most fitting results. It was observed that the fitting results of each parameter form log normal distributions similar to the assumed distribution of α. Therefore, logarithmic transformation were carried out to the fitted values of B and applied outlier tests following Peirce's method (Peirce, 1852). Specifically the R-code written by Dardis and Muller (r-forge.r-project.org/projects/pierce/) was used, which extends the development of Pierce's method by S. M. Ross in 2003 (Ross, 2003).
Fits that meet the criterion of n and pass the outlier tests are deemed successful and this defines the Number of Successes (NOS) among the 1000 repeats carried out. Within these biologically feasible results, the mean and standard deviation of the three fitted parameters are compared, namely B, H and n in Equation 2 in the main text.
The variation in the magnitude of the parameters is a measure of the effect of experimental noise on our estimates. The coefficient of variation of the estimated parameters was therefore plotted against the level of noise introduced in the simulated data in
6. Normalization for Comparison with Stably Transformed Plants:
A key difference in the mathematical description of the stably transformed plants is the number of working circuits each protoplast contains. For the heterozygous plants used in our study, a single copy of the inducer-repressor circuit is expected, as suggested from genetic segregation data (not shown). However, for the transient protoplast assay it is expected that on average multiple copies of the plasmid would be found in each protoplast. The data shown in αi
Nij
for the construct, estimates of B from transient data are expected to be overestimates of the estimates of B for stable constructs. In agreement with this expectation, tests on simulated data with varying levels of mean Nij showed that Bi was systematically overestimated as the mean number of plasmids became larger (FIG. 11). In order to correct for this overestimation we normalized our stably transformed plant data with the mean of the distribution coming from the transient assay. In other words, we defined a normalization factor λ*i such that
Here, the superscript 0t refers to the zero-inducer values for the transient assay, and 0s refers to the zero inducer values of the stable transformation assay. Dividing the data by λ*i therefore not only replaces αi by αi
but multiplies each F-luc and R-luc value by the fraction by which the plasmids in an average well in the transient assay exceed those in the stably transformed assay (i.e. the fraction
Nij0t
ij
Nij0s
j). Tests on simulated data (
7. Testing the Normalization Factor λ*i with Simulated Data:
As described elsewhere, there is only a single copy of the repressible promoter circuit in the stably transformed transgenic plant cells and multiple copies of the plasmid are expected in each transiently transformed protoplast. This leads to different multipliers found in parameter B in Equation 2 and hinders the direct comparisons of the estimated parameter values between transient and stably transformed assays. In the main text, we proposed a normalization factor λ*i to correct this bias from plasmid numbers. We will test if this normalization factor λ*i behaves as expected using simulated data similarly to what we did for
Due to the natural differences between the transient and stably transformed assays, we assume the noise levels are positively proportional to the mean numbers of plasmid in each protoplast. Nij of the transient assays is assumed to be high in both mean and variance, and Nij of the stably transformed assays to be low in both. Therefore, we assume the coefficient of variance (COV) is the same between the transient and stably transformed assays, while the absolute levels are different. Therefore, instead of varying the standard deviation of the normal distribution underlying Nij while keeping the mean value the same in Fij0
j from the case with highest mean value (simulated transient assays) rather than with their corresponding
Fij0
j as in previous simulations (Equation 3 in the main test). We applied the same fitting procedure and measurements to both normalized dataset and its corresponding raw dataset. The results are shown in
8. Bootstrapping Data Analysis of Transient Vs Stable Transformants:
Bootstrapping statistical analysis was carried out to generate mean values and confidence intervals for the predictions in stably transformed plants. Bootstrapping is a useful inference method when the underlying distribution of the data is not known or when the sample size is small (Fox, 2008).
As shown elsewhere: (1) The predictions for B are quite accurate in one of the three cases, and overestimated only by a factor of three or four in the other two. (2) The predictions for H are very good in one of the three cases and within a factor of five for the third. (3) The predictions for n also lie within a factor of two or three of the value for the plant data. Bootstrapping then looks into the question: if the data had been sampled differently would the predictions still look the same?
To generate the different sample sets, the original data set was randomly selected to form bootstrap sample sets in the following 3 steps. First the appropriate number of bins to histogram F-Luc values was chosen. The chosen number of bins was the largest number that yielded no bins with zero values in it. This was done to optimize the sampling of the data. Next the number of sample points to draw from each bin was chosen. This was set to be one greater than the minimum number of points in any bin. For example, if one bin in the F-Luc histogram contained only one point, the maximum number of sample points that could be drawn from any bin was set as 2. This was done to avoid drawing the same point an excessive number of times per sample. Histograms were then made of the F-luc data with the number of bins chosen in step 1, and the corresponding R-luc data was placed in a corresponding R-luc bin. Bootstrapped samples were created by drawing the number of sample points fixed in step 2 from each bin, randomly and with replacement.
500 bootstrap samples were created separately from transient and stable data for each construct. Each bootstrap sample was fit using the standard procedure, and the parameters B, H and n estimated. This exercise produces a distribution of fitted values for each parameter. In
The results of the bootstrapping exercise are shown in
This application claims priority to U.S. provisional application No. 62/409,555, filed Oct. 18, 2016, the disclosure of which is hereby incorporated by reference in its entirety.
This invention was made with government support under grant DE-AR0000311 awarded by the Department of Energy and grant W911NF-09-10526 awarded by the US Department of Defense, Defense Threat Reduction Agency. The government has certain rights in the invention.
| Number | Date | Country | |
|---|---|---|---|
| 62409555 | Oct 2016 | US |
