Field of invention

The invention refers to a bistable genetic toggle switch comprising a pair of mutual repressors and a positive feedback loop based on DNA-binding proteins. Each repressor and activator pair binds to its corresponding binding site on DNA. This mechanism constitutes switch functionality.

State of the art

Genetic regulatory networks are hirearchically organised networks of genes that act on each other in order to perform a function of some kind.

Genetic arrangements of this kind can be found in nature, such as a switch in the bacteriophage lambda and a circadian oscillator in cianobacteria (2).

In the context of recent developments in the fields of biotechnology and synthetic biology, there arose a need to construct synthetic genetic regulatory networks with well-defined and complex functions.

Bistable and multistable toggle swithces are the most useful representatives, as they can function as a form of memory in biological systems as well as in the control of expression of endogenous and heterologous genes. The development started with the construction of a bistable switch using endogenous transcription factors in Escherichia coli (3). Kramer et al. constructed a bistable switch out of similiar elements in mammalian cells (4). This switch consists of two constitutive promoters and their corresponding opposing repressors. A genetic network of this kind can assume two stable states, since the activation of the promoter of the first construct activates expression of the repressor that represses the promoter of the second construct. For a practical application of such a switch, an external signal able to push the switch into the desired state is needed. This kind of change should be possible with the addition of small molecules that act as inducers, notably anhydrotetracyclin (aTc) and isopropyl-beta-D-l-thiogalactopyranoside (IPTG) (3). Each of the inducers binds to its corresponding repressor and inactivates it. For example, inducer 2 binds to repressor 2 and inactivates it. Lack of repression of promotor 2 results in expression of repressor 1 , which binds to promotor 1 and blocks the transcription of repressor 2. Similiarly, when inducer 1 is added, it inactivates repressor 1 and promoter 1 activates transciption of repressor 2, which represses promoter 2 and therefore blocks the transcription of repressor 2. It is important that the system exhibits stability -a set state must persists after the removal of the inducer. Kramer et al. prepared a bistable switch in Chinese hamster ovary (CHO) cells (4). This switch functions very similiarly to the switch constructed in Escherichia coli. The switch is made of the same two opposing promoters, followed by binding sites for repressors. Promoter 1 is followed by the repressor 1 binding site, whereas promoter 2 is followed by the repressor 2 binding site. Both promoters and corresponding binding sites are followed by a gene encoding a DNA-binding protein (promoter 1 followed by DNA-binding protein 2 and vice versa for promoter 2). Mammalian repressors are constructed modularly - the repressor and the DNA- binding function are exerted by separate domains of the protein. This provides the oppurtunity to use specific DNA-binding domains, such as zinc-finger domains or TAL (Transcriptional activator like) proteins and combine them with repressor domains (for instance KRAB), or with activator domains (such as VP 16), which results in functional transcriptional repressors or activators, respectively. In the case of the switch designed by Kramer et al. (4), the DNA- binding domains A and B are fused with the repression domain KRAB which results in two repressors: A and B.

Drawbacks of current state of the art

Development in the field continued with the design of gene regulatory networks of higher complexity consisting of several bistable switches. Networks with an odd number of repressors were constructed, exhibiting oscillatory behaviour (5,6). In addition, networks of repressors that were capable of boolean logic operations were constructed (7).

Up to now, all of the designed genetic switches were based on natural bacterial DNA-binding proteins as building blocks of repressors. The number of well characterized natural repressor proteins is limited, which limits the number of independent bistable switches able to operate inside a single cell. Several biochemical properties such as stability, oligomeric state and affinity for DNA vary between different natural repressor proteins, contrary to the the desired balance between the regulatory elements which is needed for a bistable switch to function properly and robustly. The second drawback of bistable switches described to date is the use of only repressors, which results in relatively small differences between the two states. In the study by Kramer et al. (4), it was found that the state to state difference in the expression of a reporter gene under the control of a bistable switch was relatively small.

Literature:

1. Ptashne, M. A Genetic Switch: Phage λ and Higher Organisms (Cell, Cambridge, Massachusetts, 1992).

2. Ishiura, M. et al. Expression of a gene cluster kaiABC as a circadian feedback process in cyanobacteria. Science 281, 1519-1523

3. Gardner, T. S., Cantor, C. R. & Collins, J. J. Construction of a genetic toggle switch in Escherichia coli. Nature 403, 339-42 (2000).

4. Kramer, B. P. et al. An engineered epi genetic transgene switch in mammalian cells. Nature biotechnology 22, 867-70 (2004).

5. Elowitz, M. B. & Leibler, S. A synthetic oscillatory network of transcriptional regulators. Nature 403, 335-8 (2000).

6. Fung, E. et al. A synthetic gene-metabolic oscillator. Nature 435, 1 18-122 (2005).

7. Tamsir, A., Tabor, J. J. & Voigt, C. a. Robust multicellular computing using genetically encoded NOR gates and chemical "wires. 'Nature 1-4 (2010). doi:10.1038/nature09565

Summary of invention.

Artificially designed DNA-binding proteins, e.g. zinc-finger proteins (ZFP) and TAL effectors (US patent application: US201201 10685), which can be prepared in any desired numbers, seem to be the best solution to the problem of limited numbers of natural repressor proteins. However, computer simulations and laboratory experiments have shown that a bistable switch based on synthetic DNA-binding proteins is difficult to implement. Cooperativity of repressors is a prerequisite to for the preparation of a functional bistable toggle switch (US patent US6841376) (Figure 2). Synthetic modular DNA-binding proteins (e.g. TAL effectors, ZFP) bind DNA as monomers and therefore lack binding cooperativity. To solve this problem we propose a designed bistable toggle switch composed of two mutual repressors based on DNA-binding protein domains, and an additional positive feedback loop based on the same DNA-binding protein domains fused with an activator domain (Figure 1). This switch is significantly more stable than a switch based on synthetic DNA repressor proteins without an integrated positive feedback loop. Moreover, the difference in reporter expression in both toggle states of this switch is far greater than in the case of a switch without a positive feedback loopand absolute repression is not necessary for bistability.

Figure 1 represents a scheme of the positive feedback loop switch design. Transcription of both the repressors and the activators is controlled by a minimal promoter that ensures very low expression of the downstream gene. The expression of the downstream gene is greatly increased if a DNA-binding protein fused with an activatory domain binds upstream of the minimal promoter. The switch has binding sites for the appropriate artificial transcription factors, which are composed of designed DNA-binding proteins and located upstream of promoters. The binding sites for the artificial DNA-binding proteins A and B are located upstram of the promoters of the first or of the second group of genes, respectively.

Each state requires transcription of two genes, therefore the switch is composed of four genes. An optional number of genes (effectors or reporters) expressed in either of the two toggle states can be added. (Figure 3).

The following are active in state 1 : repressor B, which is under the control of operator A located upstream of either a minimal or a constitutive promoter. Repressor B represses the expression of all genes characteristic of state 2 as well as the effectors or the reporters of state 2;

activator A, which is under the control of operator A. Activator A activates the expression of all genes characteristic of state 1, i.e. the repressor B and the activator A as well as the effectors of state 1. The activator A's operon is auto-activatory and represents a positive feedback loop that can be inhibited by repressor A.

The following are active in state 2:

- repressor A, which is under the control of operator B located upstream of either a minimal or a constitutive promoter. Repressor A represses the expression of all genes characteristic of state 1 as well as the effectors or the reporters of state 1 ; activator B, which is under the control of operator B. Activator B activates the expression of all genes characteristic of state 2, i.e. the repressor A and the activator B as well as the effectors of state 2. The activator B's operon is auto-activatory and represents a positive feedback loop that can be inhibited by repressor B.

The activator consists of a DNA-binding domain that is fused to an activation domain, such as the VP 16 or the VP64 domain. The repressor consists of a DNA-binding domain that is fused to a transcriptional repression domain, such as the KRAB domain.

Because the repressor B, activator A and the state 1 effector are under the joint transcriptional control of operator A, we can combine them under the same operator and promoter in single or multiple DNA sequences linked with 2A sequences between structural genes, which enable co-translational cleavage of proteins, e.g. the t2A sequence.

We can combine repressor A, activator B and the state 2 effectors under the joint control of a single operator and promoter in the same manner.

Figure descriptions

Figure 1 : A scheme of a mutual repressor switch with a positive feedback loop based on DNA-binding proteins. A) A multiple-operon implementation allowing toggle control with the same operator for each state. B) A single-operon implementation allowing toggle control with the same operator for each state.

Figure 2: A scheme of a classic bistable toggle switch, as implemented in bacterial and mammalian cells, based on bacterial cooperative DNA-binding domains (Gardner et al. 2000, Kramer et al. 2004).

Figure 3: A scheme of a bistable genetic toggle switch comprising a pair of mutual repressors and a positive feedback loop, based on DNA-binding proteins. A) Activator A is expressed in state 1 and binds to DNA binding element A, activating transcription of the structural genes of repressor B, activator A (itself) and effector 1. The expression of repressor B inhibits the expression of repressor A and of activator B. B) Activator B is expressed in state 2 and it binds to DNA binding element B, activating the transcription of the structural genes of repressor A, activator B (itself) and effector 2. The expression of repressor A inhibits the expression of repressor B and of activator A. Figure 4: Implementation of a bistable toggle switch with a pair of mutual repressors and a positive feedback loop, based on TAL DNA-binding proteins, applicable for expression in mammalian cells. Legend: the arrows depict the structural genes coding for repressor (TAL:KRAB), activator (TAL:VP16), effectors (fluorescent proteins) and inducer proteins; 2A represents the position of the self-cleaving peptide; a round symbol represents the position of an operator withDNA binding elements, which along with the promoter controls the expression of structural genes.

Figure 5: Implementation of a bistable toggle switch with a pair of mutual repressors and a positive feedback loop, based on TAL DNA-binding proteins, with joint operons for expression in mammalian cells. Legend: the arrows represent structural genes coding for repressor (TAL:KRAB), activator (TAL:VP16), effectors (fluorescent proteins) and inducer proteins; 2A represents the position of the self-cleaving peptide; a round symbol represents the position of the operator with the DNA binding elements, which along with the promoter controls the expression of structural genes.

Figure 6: Detection of bistability in mammalian cells, which have been transfected with plasmids encoding the neccessary genes for the switch.

Figure 7: Detection of the TAL-repressor efficiency.

Figure 8: Detection of the effect of the DNA binding elements in the operon's operator. Figure 9: Detection of TAL activator efficiency.

Description of the invention

The present invention relates to a bistable switch comprising a pair of mutual repressors and a positive feedback loop based on modular DNA-binding proteins. The bistable switch regulates switching from stable state 1 to stable state 2 and vice-versa. The switching is directed by signals from the environment. The positive feedback loop brings non- linearity to the system, enabling the switch to function with non-oligomeric and non- cooperative DNA-binding domains.

The bistable switch includes: a) An operon for maintaining state 1 with a positive feedback loop, which includes the structural gene for activator A, which forms the positive feedback loop. Transcription of the structural gene for activator A is under regulation of an operator, which includes the DNA binding element A and a promoter. b) An operon for maintaining state 1 with repressor B which inhibits state 2 by repressing the genes under regulation of operator B. Transcription of the structural gene for repressor B is under regulation of an operator, which includes the DNA binding element A and a promoter. c) An operon representing the activation of state 1, which includes the gene for effector 1. Effector 1 can be any gene or a group of genes e.g. different enzymes, fluorescent proteins, signaling proteins etc. Transcription of the structural gene for effector 1 is under regulation of an operator, which includes the DNA binding element A and a promoter. d) An operon for maintaining state 2 with a positive feedback loop, which includes the structural gene for activator B, which forms the positive feedback loop. Transcription of the structural gene for activator B is under regulation of an operator, which includes the DNA binding element B and a promoter. e) An operon for maintaining state 2 with repressor A which inhibits state 1 by repressing the genes under regulation of operator A. Transcription of the structural gene for repressor A is under regulation of an operator, which includes the DNA binding element B and a promoter. f) An operon representing the activation of state 2, which includes the gene for effector 2. Effector 2 can be any gene or a group of genes e.g. different enzymes, fluorescent proteins, signaling proteins etc. Transcription of the structural gene for effector 2 is under regulation of an operator, which includes the DNA binding element B and a promoter.

State 1 or state 2 are activated with the following switch-ON operons: a) An operon for switching to state 1 , which includes a structural gene for activator A to activate state 1, and a structural gene for repressor B which inhibits state 2. The transcription of these structural genes is under regulation of a promoter and an operator, which is active in the presence of inducer A. The promoter can be constitutive or minimal. b) An operon for switching to state 2, which includes a structural gene for activator B to activate state 2, and a structural gene for repressor A to inhibit state 2. The transcription of these structural genes is under regulation of a promoter and an operator, which is active in the presence of inducer B. The promoter can be constitutive or minimal.

Switching operons are activated by inducers: a) Operons of the inducers, which include structural genes for inducer proteins under regulation of a promoter, preferably constitutive or minimal, and optionally an operator.

Operons of the structural genes, which are under regulation of the same operator (activator A, repressor B and effector 1 or activator B, repressor A and effector 2) can be joined in groups of two or more so that they are under regulation of the same operator and are linked with self-cleaving 2A peptides. They are transcribed from structural genes in the form of a single RNA and are translated into polypeptide chains, leading to synthesis of individual proteins in an equal stoichiometric ratio.

The number of DNA binding elements for DNA-binding domains of the activator or repressor ranges from one to several, preferably from one to 20, more preferably from one to 12. The position of the DNA binding elements for transcription factors in regard to the promoter can be upstream or downstream of the promoter, or both. In a preferred embodiment, the DNA binding elements are upstream of the promoter, which is adjacent to transcription initialization site and the structural gene.

The operons composing the bistable switch include a structural gene, which encodes several proteins adjacently linked with a self-cleaving 2 A peptide. Optionally, the above- mentioned polycistronic operons may be divided into seperate operons encoding a single protein or two proteins liked via a self-cleaving 2A peptide with the same operators upstream of the structural gene.

Definitions:

The term »DNA-binding domain« refers to DNA-binding domains of protein families such as TAL effectors, zinc fingers and other transcriptional regulators, their homologues, orthologues and mutants with preserved or enhanced basic functions of DNA-binding proteins.

The term »TAL« refers to synthetic or natural TAL proteins, preferably their central DNA- binding domain with an additional nuclear localization signal. The central domain of a TAL protein is composed of a variable number of TAL repeats. The term »TAL« may also refer to homologues, orthologues and mutants with preserved or enhanced basic function of TAL proteins. The term »TAL« may refer to synthetic TAL domains with any number of TAL repeats in any order, additionally containing a nuclear localization signal.

The term »operator« refers to a DNA sequence containing DNA binding elements located near a promoter. An operator can be located upstream or downstream of a promoter, preferably upstream. For the purposes of the present invention, the operator can contain one or more sequential, either identical or different DNA binding elements. Repeats of different DNA binding elements can either alternate or cluster. The number of DNA binding elements is not limited. The term »DNA binding element« refers to a specific nucleotide sequence on a DNA molecule, which binds to the DNA-binding domain. The nucleotide sequence of the DNA binding element depends on the DNA-binding domain specificity of repressors and activators of the switch. DNA binding elements composing an operator may be separated with a variable number of nucleotides. The number of nucleotides separating the DNA binding elements is between 2 and 100, preferably between 2 and 50.

The term »transcription repression domain« refers to a protein domain, which ensures the inhibition of structural gene transcription, if linked to a DNA-binding domain. The function of a transcription repressor domain, linked to a DNA-binding domain is the inhibition of structural gene transcription by preventing the binding of RNA polymerase to the corresponding promoter. Transcription repressor domains can be chosen from a range of repressors known to persons skilled in the art, preferably from the family of KRAB repressors. The term »KRAB« refers to »Kriippel-associated box« and may also refer to homologues, orthologues and mutants with preserved or enhanced basic function of inhibiting structural gene transcription.

The term »transcription activation domain« refers to a protein domain, which ensures activation of structural gene transcription, if linked to a DNA-binding domain. The function of a transcription activation domain linked to a DNA-binding domain is transcriptional activation of structural genes. Transcription activation domains can be chosen from a range of activators known to persons skilled in the art, preferably from the family of VP 16 and VP64 activators. The term »VP16« refers to a transcription activation domain of viral origin. VP 16 induces formation of a protein complex, which enhances expression of structural genes. The term »VP64« refers to four tandem repeats of the activation region of the VP 16 domain.

The term »repressor« refers to proteins, comprising a DNA-binding domain and a transcription repression domain, preferably KRAB. The term »repressor« refers to proteins with the function of inhibiting structural gene transcription when bound to their respective DNA binding elements.

The term »activator« refers to proteins, comprising a DNA-binding domain and a transcription activation domain, preferably VP 16 or VP64. The term »activator« refers to proteins with the function of activating structural gene transcription when bound to their corresponding DNA binding elements (Garg et al. 2012).

The DNA-binding domain and the transcription repression/activation domain are linked together by a linker peptide, which is any polypeptide of any length and any aminoacid sequence. The term »linker peptide« refers to aminoacid sequences with the function of separating individual domains of a chimeric protein. Optional functions of a linker peptide in a chimeric protein can also be cleavage or posttranslational modification site introduction.

The ratio of the DNA-binding domain and the transcriptional repression/activation domain can be 1 :1 or 1 :2. One or more transcriptional repression/activation domains can be linked to the DNA-binding domain at the N or C terminal end or at the N and C terminal end.

The term »minimal promoter« refers to a DNA sequence of a few nucleotides in length located upstream of a transcription initiation site and is a minimal requirement for the binding of transcription factors and gene transcription. Nucleotide sequences of minimal promoters are known to persons skilled in the art and have been extensively described elsewhere. The term »constituitive promoter« refers to a DNA sequence, which ensures continuous transcription of structural genes. Its location and sequence is known to persons skilled in the art and has been previously described elsewhere. The term »constituitive promoter« refers to an unregulated promoter, enabling continuous expression of the corresponding gene.

The term »inducer« refers to molecules, able to regulate gene expression by binding to proteins, e.g. repressors or activators. The term »inducer« refers to antibiotics and their analogues, natural compounds and their analogues, metalothionines, steroids and analogues; e.g. tetracyclin, doxycyclin, anhidrotetracyclin, rapamycin and analogues, ecdysone and analogues (e.g. ponasteronA, muristeronA), alolactose (lac operon), arabinose (ara operon), cumermycin and novobiocin, RU486 (mifepriston), estrogens and analogues (e.g. 4-hydroxi- tamoxifene), streptogramines (e.g. pristinamycin), macrolides (e.g. erythromycin), vanilinic acid, cumate, phloretin, biotin, arginine, metal ions, polymeric substrates (e.g. pectin, xylan, arabinan) or monomelic units of degraded polymers (e.g. arabinose, xylose, metals and metal ion-protein complexes, cAMP, cyanate, CRP, formate, maltose, acetolactate, urea) and other compounds known to persons skilled in the art. The term »inducer« may also refer to temperature, pH or redox potential, if the change in temperature, pH or redox potential effects the activity of repressors or activators.

The term »self-cleaving peptide« refers to aminoacid peptide sequences, that ensure autocatalytic cleavage of the peptide, such as 2 A sequences (e.g. t2A, e2A, f2A etc). Self- cleaving peptides enable synthesis of a polycistronic mRNA chain, from which individual proteins are synthesised. The polycistronic chain can contain two or more sequentially linked proteins, separated by the self-cleavable peptide. The described composition ensures synthesis of individual proteins in equal stoichiometric ratio.

The term »signal sequence« or »signal peptide« refers to an aminoacid sequence, important for directing the protein to a certain location in the cell. Signal sequences differ depending on the host organism for protein expression. Aminoacid sequences and functions of signal peptides are known to persons skilled in the art and are available in the literature.

The term »tag peptide« refers to aminoacid sequences, added to a protein for simplified purification, isolation or detection. The position of signal sequences, linker peptides and tag peptides can be arbitrary, although they should allow functional expression of the protein, while also preserving the function for which these sequences were selected.

The terms »homologue« and »orthologue« refer to polypeptides, originating from the same or different organism. The term »homologous« also refers to mutated protein segments, where the mutations have a minimal effect on the structure or function of the polypeptide. The term »mutant« refers to a polypeptide, differing from the native protein polypetide in at least one aminoacid.

The term »effector« refers to any protein.

The terms »promoter«, »teminator«, »protein«, »DNA« are generally known to persons skilled in the art and are used as expected.

Embodiments of the invention can contain one or more switches. The switches can function independently or can be interconnected.

An embodiment of the invented switch enables controlled expression of state 1 or state 2, maintains a stable state even when the inducer molecule is removed, and is capable of switching between the two states, depending on the presence of a corresponding inducer. States are defined by expression of one or more effectors, e.g. therapeutic molecules, signal molecules, regulators or any other protein molecules.

The switch can be used for state signalization as a reporter system, e.g. reporting the presence of an inducer. Such inducers might include but are not limited to metal ions, pH, glucose and others.

The term »expression vector« refers to circular or linear DNA plasmids or viral DNA, containing operons listed in the invention and the necessary elements for expression in prokaryotic or eukaryotic cells, which are known to persons skilled in the art. Bacterial vectors contain bacterial control elements, a bacterial replication origin and an antibiotic resistance operon for selection of successfully transformed bacteria. Eukaryotic vectors contain, in addition to a bacterial replication origin, appropriate eukaryotic control elements, and appropriate antibiotic resistance operons for selection of successful bacterial transformation and/or successful eukaryotic transfection.

Embodiements of the invention can be used in prokaryotic as well as in eukaryotic organisms and cell lines. The basic difference is the use of transcription and translation ensuring nucleotide sequences in promoters and terminators, known to persons skilled in the art.

The invention further includes host cells and organisms, which either transiently or stably incorporate the nucleic acids described herein. The appropriate host cells are known to persons skilled in the art and include bacterial and eukaryotic cells. One skilled in the art will appreciate that a protein can be expressed in mammalians cells of the following origins: human, rodent, bovine, pork, poultry, rabbit and similar. Mammalian host cells include cultivated primary cell lines or immortalized cell lines.

Transfer of DNA into host cells is performed with conventional methods well known to persons skilled in the arts, such as transformation or transfection, including: chemical transfer, electroporation, microinjection, DNA lipofection, cell soni cation, gene bombarding, viral DNA transfer etc.

DNA transfer can be either transient or stable. »Transient transfer« refers to transfer of DNA in a vector, that does not undergo cromosomal insertion. »Stable transfer« refers to insertion of DNA into the host genome. DNA transfer to a cell line with a previous stable insertion can be controlled with the presence of markers. »Markers« refer to antibiotic or chemical resistance and can be included in the vector.

Examples of implementations described in detail below are conceived to best describe the invention. These descriptions are not intended to limit the field of the invention or its applicability, but serve to better demonstrate the invention and its appplicability.

Exemplification

Example 1. Preparation of DNA constructs for the switch according to the invention

For the preparation of DNA constructs the inventors used methods of molecular biology, such as: chemical transformation of competent E. coli cells, DNA plasmid isolation, polymerase chain reaction (PCR), reverse transcription - PCR, PCR ligation, determination of nucleic acid concentration, agarose gel electrophoresis of DNA, isolation of DNA fragments from agarose gels, chemical synthesis of DNA, DNA digestion with restriction enzymes, digestion of plasmid vectors, ligation of DNA fragments, purification of plasmid DNA in larger quantities. The protocols of the experimental techniques and methods are well known to person skilled in the art and are described in the manuals of molecular biology.

All work was performed with sterile techniques, which are well known to persons skilled in the art. All plasmids, completed constructs and partial constructs were transformed into bacteria E. coli with chemical transformation. Plasmids and constructs were transfected into cell lines HEK293 and HE 293T using comercially available transfection reagents.

The final gene constructs comprising the operons for the switch as described in the present invention are listed in Table 1 and the proteins transcribed from structural genes are listed in Table 2. All operons have been prepared using techniques and methods known in the art. Operons were inserted into appropriate plasmids suitable for eukaryotic systems. The inventors confirmed adequacy of nucleotide sequences by sequencing and restriction analysis.

The label [A] represents the sequence of the transcription factor-binding DNA element that is the recognition sequence of the DNA-binding protein TALA, which is a TAL effector designed to recognize the chosen DNA sequence. The DNA sequence that is the recognition sequence of the Tt represorrepressor is labeled [TRE] and the DNA sequence that is the recognition sequence of ecdysone is labeled [ECD].

The operon for the maintenance of stable state 1 is comprised of the DNA binding element [A] for the repressor TALArKRAB or the activator TALA:VP16, a minimal promoter and structural genes of a feedback loop, which in turn is comprised of a repressor for the inhibition of expression of structural genes from the operon formaintenance of stable state 2, specifically it comprises TALBrKRAB. Separately, the operon for the maintenance of stable state 1 comprises also a DNA-binding site [A] for the repressor TALA:KRAB or the activator TALA:VP16, a minimal promoter and structural genes of a positive feedback loop to maintain state 1, specifically TALA:VP16, and any effectors (Figure 1, Figure 3, Figure 4, Figure 5). Both operons can be combined into a single operon in which structural genes are separated by a self-cleaving peptide 2A (Figure 2, Figure 3, Figure 4, Figure 5). The operon for the maintenance of stable state 2 is comprised of the DNA binding site [B] for the repressor TALB:KRAB or the activator TALB:VP16, a minimal promoter and structural genes of a feedback loop, which in turn is comprised of a repressor for the inhibition of expression of structural genes from the operon formaintenance of stable state 1 , specifically it comprises TALA:KRAB. Separately, the operon for the maintenance of stable state 2 comprises also a DNA-binding site [B] for the repressor TALB:KRAB or the activator TALB:VP16, a minimal promoter and structural genes of a positive feedback loop to maintain state 2, specifically TALB:VP16, and any effectors (Figure 1 , Figure 3, Figure 4, Figure 5). Both operons can be combined into a single operon in which structural genes are separated by a self-cleaving peptide 2A (Figure 1, Figure 3, Figure 4, Figure 5).

The operator contains 10 DNA binding elements for either the DNA-binding protein TALA or TALB.

The operons for switching into stable state 1 are comprised of a constitutive or minimal promoter, DNA binding elements for an inducer- dependent activator or repressor, and structural genes encoding TALA:VP16 for state 1 activation and (on a separate operon) TALB:KRAB for state 2 inhibition. Operons for switching into state 1 can be combined into a single operon, which contains the DNA-binding elements for an inducer-dependent activator or repressor, minimal or constitutive promoter and structural genes separated by a self- cleaving peptide 2A.

The operons for switching into stable state 2 are comprised of a constitutive or minimal promoter, DNA binding elements for an inducer-dependent activator or repressor, and structural genes encoding TALB:VP16 for state 2 activation and (on a separate operon) TALA:KRAB for state 1 inhibition. Operons for switching into state 2 can be combined into a single operon, which contains the DNA-binding elements for an inducer-dependent activator or repressor, minimal or constitutive promoter and structural genes separated by a self- cleaving peptide 2A.

Inducers for state 1 or state 2 induction are different. Inducers are well known to persons skilled in the art and described in detail in the state of the art. Inducers can be arbitrarily selected, provided they specifically activate one of the two/several states. The TAL DNA-binding domain was obtained from TAL effectors by PCR amplification of the central DNA-binding domain to which a signal sequence for nuclear localization was added. Appropriate DNA binding elements were prepared synthetically so that they match the DNA-binding domains of selected TAL proteins. It is evident from the state of the art that DNA-binding domains based either on TAL or zinc fingers are naturally or synthetically prepared and each have their own recognition binding sites on DNA; the DNA binding elements. The number of DNA binding elements can be varied. It is evident from the state of the art that the effect of a repressor or an activator is improved by increasing the number of its DNA binding elements up to 12. We selected VP16 as the activation domain and KRAB as the repression domain. Both domains alike were fused with a TAL DNA- binding domain.

Table 1 : List of operons, DNA-binding elements and proteins.

operon structure SEQ ID NO.

10x[B]-p _MIN -TALB:VP16 Ϊ

10x[B]-p _m in-TALA:KPvAB-T2 A-mCit 2

10x[A]-p _MIN -TALB:KRAB-T2A-TALA:VP16-T2A-BFP 3

10x[A]-p _MIN -TALB:KRAB 4

10x[A]-p _MIN -TALA:VP16-T2A-BFP 5

10x[B]-p _MIN -TALA:KRAB-T2A-TALB:VP16-T2A-mCit 6

PCMV-[TRE]-TALB:VP16 7

PcMV [TRE] -TAL A: KRAB 8

[ECD] -p _m in-TALB :KRAB 9

[ECD]-p _min -TALA:VP16 10

pCMv-tTR-KRAB

PCMV-ECD-KRAB

[B] 19

[A] 20

pCMV-tTR:KRAB 21

pERV3 22 protein SEQ ID NO.

TALA:VP16 11

TALB:VP16 12

TALA-.KRAB 13

TALB:KRAB 14 mCit 15

BFP 16

TALB:KRAB-T2A-TALA: VP 16-T2A-BFP 17

TALA:KRAB-T2A-TALB : VP 16-T2A-mCit 18

Example 2, Bistable switch in mammalian cells

The methods and techniques of culturing cell cultures are well known to persons skilled in the art, therefore they are only briefly described with the intention of illustrating the implementation example. Cell lines of HEK293 and HEK293T cells were cultured at 37 °C and 5% C0 ₂ . DMEM medium supplemented with 10% FBS, which contains all the necessary nutrients and growth factors was used for cell culturing. Once the cell culture reached an appropriate density, cells were subcultured into a new culture vessel and/or diluted. For the application of cells in experiments the number of cells was determined with a hemocytometer and seeded at 2,5 x lO ⁴ cells per well on a 12 well microtiter plate 24 hours prior to transfection. Seeded plates were incubated at 37 °C and 5% C0 ₂ , until cells were 50-70% confluent for transfection with JetPei transfection reagent (Polyplus transfection). Transfection was performed according to the manufacturer's protocol, modified to a 12 well microtiter plate.

The HEK293 and HEK293T cell lines were transfected with plasmids described in the present invention and listed in Table 1.

For determination of effector expression and the system's state the cell medium was changed 2 hours post transfection and again 2 days post transfection, after which cells were incubated an additional 2 days. At the first medium change, an inducer for state 1 or state 2, tetracycline or ecdysone, was added. The expression of the effectors, blue fluorescent protein and yellow fluorescent protein, was monitored on a flow cytometer. A laser with a wavelength of 405 nm (blue fluorescent protein - BFP) and 488 nm (yellow fluorescent protein - mCit) was used. Emission was measured in the 430-480 nm (BFP) and 520-550 nm (mCit) range.

Results shown in figure 6A and 6B demonstrate that the switches function as described in the present invention. Results shown in figure 6A prove the switch enters state 1 or state 2 accordingly in the presence of plasmids PCMV-TALA: VP 16 (column 1), PCMV-TALB:VP16 (column 2), p _CM v-TALA:VP16 and pc _M v- ALB : KRAB (column 3), p _CM v-TALB:VP 16 and PCMV-TALA :KRAB (column 4). All cells were, in addition to the plasmids listed earlier, also transfected with plasmids SEQ ID NO. 1, 2, 4, 5 of the switch as described in the present invention.

Results shown in figure 6B prove the switch enters state 1 or state 2 accordingly with the addition of inducers for state 1 or 2 and switching of the switch. All cell were in addition to the plasmids listed earlier also transfected with plasmids SEQ ID NO. 1,2,4,5,7,8,9,10 of the switch by the invention. Without the addition of an inducer the cells enter either state 1 or state 2. Inducer 1 causes the cells to enter state 1 exclusively, inducer 2 causes the cells to enter state 2 exclusively.

Example 3. TAL-repressors

A repressor was prepared by fusing a TAL DNA-binding domain and a KRAB repressor domain, where the KRAB repressor domain was fused to the C- and N-terminal end of the TAL DNA-binding domain. Two different DNA-binding domains were used: TALA and TALB. Operons with operators containing DNA binding elements specific for either TALA or TALB DNA-binding domain were prepared (labeled [A] and [B] respectively). The operator was followed by a CMV constitutive promoter and an effector, which is a luciferase gene in this example. HEK293 cells, whose culturing and transfection are described in implementation example 2, were transfected with repressors TAL:KRAB, KRAB:TAL or KRAB:TAL:KRAB and reporter, either pcMV-[A]-effector (luciferase gene) or PCMV-[B]- effector (luciferase gene). After 72 hours of culturing luciferase expression was assayed. Reduced luciferase activity is a direct proof of TAL-repressor function (figure 7). Results showed that the TAL-repressor function is independent of the KRAB repressor domain position, since the effector (luciferase) expression is sufficiently repressed.

Example 4. DNA binding elements for TAL DNA-binding domains We prepared operons with an effector (luciferase) whose operons contained one, two, four or seven DNA binding elements lx[D], 2x[D], 4x[D], 7x[D] for the TALD DNA-binding domain fused to a KRAB repressor domain. The operons also contained a CMV constitutive promoter, placed between the operator and structural genes. HEK293 cells, whose culturing and transfection are described in implementation example 2, were transfected with the TALD: KRAB repressor and a reporter, either pcMV-lx[D]-effector (luciferase gene), PCMV 2x[D]-effector (luciferase gene), PCMV-4X[D] -effector (luciferase gene) or PCMV-7X[D]- effector (luciferase gene). After 72 hours of culturing luciferase expression was assayed. Reduced luciferase activity is a direct proof of the TAL-repressor function (figure 8) on reporters with a different number of DNA-binding elements. Results showed that the function of a TAL-repressor is effective and dependent on the number of DNA-binding elements in the operons operator.

Example 5. TAL-activators

Activators were prepared by fusing a TAL DNA-binding domain and a VP 16 activation domain, where the VP 16 activation domain was fused to the C-terminal end of a TAL DNA- binding domain. Here we used two different DNA-binding domains TALA and TALB. We prepared operons with operators, which contained DNA binding elements specific for either TALA or TALB DNA-binding domain (labeled [A] and [B] respectively). The operator was followed by CMV or minimal constitutive promoter and an effector, which is a luciferase gene in this example. HEK293 cells, whose culturing and transfection are described in implementation example 2, were transfected with a TAL:VP16 activator (different amounts) and a reporter, either pcMv-[A]-effector (luciferase gene) or PC V-[B] -effector (luciferase gene). After 72 hours of culturing luciferase expression was assayed. Luciferase activity is a direct proof of the TAL-activator function, which depends on the dose of the activator (figure 9)·