Light-guided in vivo evolution of dynamic, multi-state and computational protein functionalities

Abstract

The present invention relates to a method for the evolution of a protein of interest by applying light-based selection pressure to a yeast cell, (i) wherein in the yeast cell the expression of one or more gene(s) that encode a protein controlling the progression of the yeast cell through the cell cycle is / are optogenetically controlled by an optogenetic gene expression system that is activated by light pulses of a specific wavelength, and wherein the protein of interest is the protein or one of the proteins constituting the optogenetic gene expression system; or (ii) wherein in the yeast cell the expression of one or more gene(s) that encode a protein controlling the progression of the yeast cell through the cell cycle is / are controlled by the protein of interest or a signal transduction cascade that comprises the protein of interest, and wherein the expression of the protein of interest is optogenetically controlled by an optogenetic gene expression system that is activated by light pulses of a specific wavelength, wherein the method comprises (a) exposing the yeast cell to light pulses that differ from the light pulses that optimally activate gene expression but that still in part activate gene expression; (b) culturing the yeast cell contemporaneously under conditions where it divides and wherein random mutations are introduced for a time whereby a light-based selection pressure towards yeast cells being capable of optimally dividing under the light of (a) occurs, and (c) optionally isolating one or more yeast cells being capable of optimally dividing under the light pulses of (a) and preferably identifying the one or more mutation(s) in the gene encoding the protein of interest.

Description

[0001] Light-guided in vivo evolution of dynamic, multi-state and computational protein functionalities

[0002] The present invention relates to a method for the evolution of a protein of interest by applying lightbased selection pressure to a yeast cell, (i) wherein in the yeast cell the expression of one or more gene(s) that encode a protein controlling the progression of the yeast cell through the cell cycle is / are optogenetically controlled by an optogenetic gene expression system that is activated by light pulses of a specific wavelength, and wherein the protein of interest is the protein or one of the proteins constituting the optogenetic gene expression system; or (ii) wherein in the yeast cell the expression of one or more gene(s) that encode a protein controlling the progression of the yeast cell through the cell cycle is / are controlled by the protein of interest or a signal transduction cascade that comprises the protein of interest, and wherein the expression of the protein of interest is optogenetically controlled by an optogenetic gene expression system that is activated by light pulses of a specific wavelength, wherein the method comprises (a) exposing the yeast cell to light pulses that differ from the light pulses that optimally activate gene expression but that still in part activate gene expression; (b) culturing the yeast cell contemporaneously under conditions where it divides and wherein random mutations are introduced for a time whereby a light-based selection pressure towards yeast cells being capable of optimally dividing under the light of (a) occurs, and (c) optionally isolating one or more yeast cells being capable of optimally dividing under the light pulses of (a) and preferably identifying the one or more mutation(s) in the gene encoding the protein of interest.

[0003] In this specification, a number of documents including patent applications and manufacturer’s manuals are cited. The disclosure of these documents, while not considered relevant for the patentability of this invention, is herewith incorporated by reference in its entirety. More specifically, all referenced documents are incorporated by reference to the same extent as if each individual document was specifically and individually indicated to be incorporated by reference.

[0004] Directed evolution is a key technique in biological engineering and synthetic biology. It has been used to improve and diversify enzymes1-3, fluorescent proteins45, and antibodies67, producing numerous of the modern molecular tools used in research and industry. The method relies on mutations, either from standing variation or mutagenesis, to generate variability in genetically encoded molecules. These are then screened for a desired function. The mutation-screen cycle is often repeated to direct the molecules’ functionalities closer to the target in each round.

[0005] A major innovation has been to couple a molecule of interest’s functionality to organismal proliferation and fitness, so that continuous in vivo self-selection obviates labor-intensive screening89. An orthogonal set of major advances in the field has been in elevating mutation rates only for to-be-evolved genes10- Despite the importance of directed evolution for biological engineering, the properties that can be practically evolved with existing methods are limited to steady-state functionalities. However, much of natural and synthetic biology requires molecules that switch between different states. Specifically, all biological signal transduction, control, and computation, require changes between states, usually, on (1) or off (0). Among many such examples, a large fraction of transcription factors and kinases are controlled by one or more upstream signals. Evolving molecules with such properties requires, in principle, that selection pressure is applied to all of the computations, that is, all input-output relationships, simultaneously. Otherwise, if only one state is evolved, e.g., on-state enzymatic activity, the other states or the ability to switch to them can be corrupted. For example, directed evolution for stronger transcriptional activity of rtTA led to loss of controllability, which was desired in those experiments13 14. To cope with this limitation, researchers have resorted to alternating directed evolution campaigns for one state at a time, which is both laborious and, as shown herein, does not easily succeed15.

[0006] Herein, a novel paradigm for directed evolution is provided, ‘optovolution’, which is germane for evolving all input-output relationships of a protein of interest (POI) as well as switches among them at the same time.

[0007] In addition, optovolution is applied to proteins that are needed to switch ‘on’ or ‘off’ based on ‘on’ or ‘off’ input: a LOV transcription factor. LOV domains make up the core of most non-ion-channel optogenetic systems16-21, presumably because they are small and reliable and do not need exogenous chromophores. Interest in LOV domains is reflected in multiple diversification efforts22-24. The EI222 transcription factor, in particular, is already broadly used25. However, different applications require it to be less leaky or more light-sensitive, for example, to control expression of genotoxic genes or reduce phototoxicity, respectively. Further, all existing LOV systems are activated by blue light, which causes phototoxicity. The single currently available excitation light spectrum prevents LOV-based multiplexing with different colors, which would take advantage of the aforementioned highly desirable properties of LOV domains. (Ideally, after building and validating an optogenetically controlled system, one could shift the excitation spectrum by mutating specific residues in order to be able to combine it with others and address each separately. Otherwise, the same system cannot be straightforwardly re-used in combination with others.) Here, more than 30 new variants of EI222 were evolved, including mutants responsive to light of color other than blue that were thought to be particularly challenging to obtain. The green-shifted mutants are expected to be especially useful to ameliorate the long-standing problem of phototoxicity when using LOV domains.

[0008] The present invention relates in a first aspect to a method for the evolution of a protein of interest by applying light-based selection pressure to a yeast cell, (i) wherein in the yeast cell the expression of one or more gene(s) that encode a protein controlling the progression of the yeast cell through the cell cycle is / are optogenetically controlled by an optogenetic gene expression system that is activated by light pulses of a specific wavelength, and wherein the protein of interest is the protein or one of the proteins constituting the optogenetic gene expression system; or (ii) wherein in the yeast cell the expression of one or more gene(s) that encode a protein controlling the progression of the yeast cell through the cell cycle is / are controlled by the protein of interest or a signal transduction cascade that comprises the protein of interest, and wherein the expression of the protein of interest is optogenetically controlled by an optogenetic gene expression system that is activated by light pulses of a specific wavelength, wherein the method comprises (a) exposing the yeast cell to light pulses that differ from the light pulses that optimally activate gene expression but that still in part activate gene expression; (b) culturing the yeast cell contemporaneously under conditions where it divides and wherein random mutations are introduced for a time whereby a light-based selection pressure towards yeast cells being capable of optimally dividing under the light of (a) occurs, and (c) optionally isolating one or more yeast cells being capable of optimally dividing under the light pulses of (a) and preferably identifying the one or more mutation(s) in the gene encoding the protein of interest.

[0009] In a related aspect the present invention relates to a method for the evolution of a protein of interest by applying light-based selection pressure to a eukaryotic cell, (i) wherein in the eukaryotic cell the expression of one or more gene(s) that encode a protein controlling the progression of the eukaryotic cell through the cell cycle is / are optogenetically controlled by an optogenetic gene expression system that is activated by light pulses of a specific wavelength, and wherein the protein of interest is the protein or one of the proteins constituting the optogenetic gene expression system; or (ii) wherein in the eukaryotic cell the expression of one or more gene(s) that encode a protein controlling the progression of the eukaryotic cell through the cell cycle is / are controlled by the protein of interest or a signal transduction cascade that comprises the protein of interest, and wherein the expression of the protein of interest is optogenetically controlled by an optogenetic gene expression system that is activated by light pulses of a specific wavelength, wherein the method comprises (a) exposing the eukaryotic cell to light pulses that differ from the light pulses that optimally activate gene expression but that still in part activate gene expression; (b) culturing the eukaryotic cell contemporaneously under conditions where it divides and wherein random mutations are introduced for a time whereby a light-based selection pressure towards eukaryotic cells being capable of optimally dividing under the light of (a) occurs, and (c) optionally isolating one or more eukaryotic cells being capable of optimally dividing under the light pulses of (a) and preferably identifying the one or more mutation(s) in the gene encoding the protein of interest.

[0010] In the related aspect the eukaryotic cell is preferably a fungal cell, an animal cell or a plant cell. With these options the fungal cell is preferred. The animal cell is preferably a vertebrate cell, more preferably a mammalian cell and most preferably a human cell. The fungal cell is preferably selected from S. cerevisiae, Pichia pastoris, A. niger, A. oryzae, Trichoderma reesei and Thermothelomyces thermophilus (all being used in biotechnology, for example, for the production of enzymes).

[0011] The term “evolution of a protein” refers to the changes in the amino acid sequence (i.e. mutations in nucleotides that code for the amino-acids of the protein) of the protein from cell cycle to cell cycle that lead to the evolution of protein functions. Hence, protein evolution requires two steps: the mutation of nucleotides that code for amino acids, and the fixation of new variants in the population of cells. Selection pressure can generally be defined as any cause that reduces or increases reproductive success in a portion of a population of cells or organisms. In accordance with the present invention the cause is light (in particular light pulses).

[0012] Yeasts are eukaryotic, single-celled microorganisms classified as members of the fungus kingdom.

[0013] The proteins controlling the progression of the yeast cell through the cell cycle are known for different types of yeast cells as well as for higher eukaryotes, including humans; see Table 1. These proteins comprise cyclins and cyclin-dependent kinase enzymes. Cyclins are a family of proteins that control the progression of a cell through the cell cycle by activating cyclin-dependent kinase (CDK) enzymes or group of enzymes required for synthesis of cell cycle.

[0014] Table 1 : Proteins controlling the progression of the yeast cell through the cell cycle (see, for example, David O. Morgan, The Cell Cycle: Principles of Control, ISBN-10 0199206104)

[0015] Cell cycle progression is a highly regulated process orchestrated by numerous checks and balances to ensure that division and proliferation is a favored outcome. Progression is mediated by cyclin-dependent kinases (CDKs) and their partner cyclins. Each pair is associated with a certain phase of cell cycle.

[0016] In the yeast cell or eukaryotic cell (i) the expression of one or more gene(s) that encode a protein controlling the progression of the cell through the cell cycle is / are optogenetically controlled by an optogenetic gene expression system that is activated by light pulses of a specific wavelength, or (ii) the expression of the protein of interest is optogenetically controlled by an optogenetic gene expression system that is activated by light pulses of a specific wavelength, whereby then the expression of one or more gene(s) that encode a protein controlling the progression of the cell through the cell cycle is / are then controlled by the protein of interest or a signal transduction cascade that comprises the protein of interest.

[0017] It is to be understood that both options (i) and (ii) have the technical effect that the one or more gene(s) that encode a protein controlling the progression of the cell through the cell cycle is / are only expressed if suitable light (in particular suitable light pulses) is applied that induces the expression.

[0018] In this respect it is of note that optogenetic control (or optogenetics) is the genetic approach for controlling cellular processes with light. It generally can provide spatiotemporal, quantitative and reversible control over biological signaling and metabolic processes, overcoming limitations of chemically inducible systems.

[0019] In the case of (i) the expression of the one or more gene(s) that encode a protein controlling the progression of the cell through the cell cycle is / are directly optogenetically controlled whereas in the case of (ii) the expression of the one or more gene(s) that encode a protein controlling the progression of the cell through the cell cycle is / are directly controlled by the protein of interest or a signal transduction cascade that comprises the protein of interest and the protein of interest or a signal transduction cascade that comprises the protein of interest and the expression of the protein of interest is optogenetically controlled. Hence, in the case of (ii) there is an “indirect optogenetic control” of the expression of the one or more gene(s) that encode a protein controlling the progression of the cell through the cell cycle via the optogenetic control of the expression of the protein of interest. The expression of the protein of interest then induces the expression of the one or more gene(s) that encode a protein controlling the progression of the cell through the cell cycle; directly or via a signal transduction cascade that comprises the protein of interest.

[0020] It is furthermore to be understood that both options (i) and (ii) have the technical effect that the cell can only effectively progress through the cell cycle if the one or more gene(s) that encode a protein controlling the progression of the cell through the cell cycle is / are expressed. If no light is applied that induces the expression, then the cell cannot effectively progress through the cell cycle. It follows that in accordance with the invention the division and proliferation of the cell can be controlled by light.

[0021] In this respect it is preferred with increasing preference that the progression through the cell cycle of the cells is inhibited by at least 50%, at least 60%, at least 70%, at least 80%, at least 90%, and at least 95% if no light is applied that induces the expression as compared to the same cells if light is applied that induces the expression and / or as compared to corresponding cells, wherein the progression of the cell through the cell cycle is not optogenetically controlled.

[0022] In this respect it is of note that a number of optogenetic gene expression control systems are available in the prior art.

[0023] For a review on light-mediated control of gene expression systems that can be used in connection with the present invention reference is made to Yamada et al. (2020), Neuroscience Research, Volume 152, March 2020, pages 66-77, the data base OptoBase at https: / / www.optobase.org / switches / and the following Table 2 that has been taken from Yamada et al. (2020).

[0024] Table 2: Light-mediated control of gene expression systems

[0025]

[0026] Table 2 provides an overview of optogenetic switches, including EL222 as used in the appended examples.

[0027] For each of the known optogenetic switches the light pulses that optimally activate gene expression are known and also for further optogenetic switches that might be developed the light pulses that optimally activate gene expression can be determined by routine means.

[0028] According to step a) the cell is exposed to light pulses that differ from the light pulses that optimally activate gene expression but that still in part activate gene expression is some cells.

[0029] The light pulses that differ from the light pulses that optimally activate gene expression but that still in part activate gene expression may differ, for example, from the light pulses that optimally activate gene expression by the color (wavelength or composition of colors) of the light, the light intensity, the duration of the light pulses, the frequency of the light pulses or a combination thereof.

[0030] By this means the method of the invention applies an evolutionary pressure towards cells that optimally express or that can at all express the one or more gene(s) that encode a protein controlling the progression of the cell through the cell cycle and, this, can optionally divide and proliferate exposed to light pulses that differ from the light pulses that optimally activate gene expression but that still in part activate gene expression.

[0031] In this respect it is to be understood that the light pulses that differ from the light pulses that optimally activate gene expression may still in part and to a limited extend activate gene expression in cells with the wild-type or unmutated gene expression system I protein of interest, or may not or essentially may not activate gene expression in cells with the wild-type or unmutated gene expression system I protein of interest but may only activate gene expression in part in the cells, wherein the gene expression system I protein of interest becomes or has become mutated, so that gene expression is again in part activated in these mutated cells of the entire cell population.

[0032] In the latter case the selection pressure is so strong, that after shifting the light pulses from optimal to selective, most of the cells in the cell population die. Only the few cells that might proliferate will be the ones with mutations. Sicu a strong selection pressure is illustrated by the appended examples.

[0033] In any case the above described evolutionary pressure results in the selection of the protein functionality of interest, because the evolutionary pressure results in selecting changes in the amino acid sequence (i.e. amino acid mutations) of the protein of interest that change the protein function such that the cells can optimally divide and proliferate under the light pulses of step a).

[0034] Accordingly, step b) of the method of the invention requires culturing the cell contemporaneously under conditions where it divides and wherein random mutations are introduced for a time whereby a lightbased selection pressure towards yeast cells being capable of optimally dividing under the light of (a) occurs.

[0035] Step c) of the method is optional and requires isolating one or more cells being capable of optimally dividing under the light pulses of a) and preferably also requires identifying the one or more mutation(s) in the gene encoding the protein of interest. One or more cells being capable of optimally dividing under the light pulses of a) can be selected, for example, on solid media, in which case the cells are cultured on solid media until single colonies appear. The single colonies can then be picked. An alternative in liquid culture is letting cells grow over time and after sufficient time sequencing the gene of interest. Yet further alternatively, cells may evolved in liquid media and then plated onto the solid plates to estimate the rate of portion of mutated genes.

[0036] The one or more mutation(s) in the gene encoding the protein of interest can also be determined by routine means and methods, for example, the sequencing of the gene encoding the protein of interest.

[0037] In accordance with a preferred embodiment of the first aspect (i) the yeast cell expresses an optogenetic gene expression system comprising a light-sensing module and a transcription factor module, wherein the light-sensing module is capable of sensing light pulses of a specific wavelength, and the transcription factor module comprises a DNA binding domain (DBD), a transcription activation domain and a nuclear localization sequence (NLS), and wherein the optogenetic gene expression system forms an active transcription complex when being exposed to the light pulses of a specific wavelength, wherein the protein of interest is the protein or one of the proteins constituting the optogenetic gene expression system, and in the yeast cell one or more gene(s) that encode a protein controlling the progression of the yeast cell through the cell cycle is / are only expressed under the control of the optogenetic gene expression system, wherein the promoter region(s) of the one or more gene(s) that encode a protein controlling the progression of a cell through the cell cycle comprised) a binding site of the active transcription complex, preferably next to the minimal promoter(s) and upstream of the gene(s), and wherein the one or more gene(s) that encode(s) a protein controlling the progression of a cell through the cell cycle is / are expressed and the yeast cell starts to progress through the cell cycle and divide if the active transcription complex binds to the binding site of the active transcription complex; or (ii) the yeast cell expresses an optogenetic gene expression system comprising a light-sensing module and a transcription factor module, wherein the light-sensing module is capable of sensing light pulses of a specific wavelength, and the transcription factor module comprises a DNA binding domain (DBD), a transcription activation domain and a nuclear localization sequence (NLS), and wherein the optogenetic gene expression system forms an active transcription complex when being exposed to the light pulses of a specific wavelength; in the yeast cell the gene encoding the protein of interest is under the control of the optogenetic gene expression system, wherein the promoter region of the gene encoding the protein of interest comprises a binding site of the active transcription complex, preferably next to the minimal promoter and upstream of the gene, and wherein the gene that encodes the protein of interest is expressed if the active transcription complex binds to the binding site of the active transcription complex, and in the yeast cell one or more gene(s) that encode(s) a protein controlling the progression of the yeast cell through the cell cycle is / are only expressed under the control of the protein of interest or a signal transduction cascade that comprises the protein of interest, wherein the one or more gene(s) that encode(s) a protein controlling the progression of a cell through the cell cycle is / are expressed and the yeast cell starts to progress through the cell cycle and divides if the protein of interest is expressed, and the method comprises (a) exposing the yeast cell to light pulses that differ from the light pulses that optimally activate the formation of the active transcription complex, but still in part activate the formation of an active transcription complex; (b) culturing the yeast cell contemporaneously under conditions where it divides and wherein random mutations are introduced for a time whereby a light-based selection pressure towards yeast cells being capable of optimally dividing under the light of (a) occurs, and (c) optionally isolating one or more yeast cells being capable of optimally dividing under the light pulses of (a) and preferably identifying the one or more mutation(s) in the gene encoding the protein of interest.

[0038] Also in accordance with this preferred embodiment the yeast cell can be any eukaryotic cell as described herein above in connection with the related aspect of the present invention.

[0039] As mentioned above, the EL222 optogenetic switch is used in the appended examples.

[0040] EL222 is a light-activated DNA-binding protein from Erythrobacter litoralis. Has reversible, lightdependent DNA-binding activity. Upon illumination an internal FMN-protein adduct is formed which changes the protein conformation so that the previously sequestered DNA-binding domain is free to bind DNA. Binds to sequences within in its own promoter when illuminated but not when it has been incubated in the dark.

[0041] The minimal elements needed for light-dependent transcriptional activation EL222 comprise (1) A photosensory Light-Oxygen-Voltage (LOV) domain as light-sensing module and (2) a Helix-Turn-Helix (HTH) as DNA-binding domain. In the dark, the LOV domain binds the HTH domain, covering the HTH 4a helix essential to dimerization and DNA binding. Blue light illumination (450 nm) triggers the photochemical formation of a protein / flavin adduct within the LOV domain, disrupting inhibitory LOV / HTH interactions and allowing EL222 to dimerize and bind DNA. These structural changes spontaneously reverse in the dark, rapidly inactivating EL222 (T ~11 s at 37°C); see Motta-Mena, Nat Chem Biol. 2014 Mar; 10(3): 196-202.

[0042] This is reflected by the above preferred embodiment which refers to a light-sensing module that is capable of sensing light pulses of a specific wavelength, and the transcription factor module comprises a DNA binding domain (DBD). The light-sensing module is capable of sensing light pulses of a specific wavelength is therefore preferably photosensory LOV domain and / or the DBD preferably comprises a HTH domain.

[0043] Two N-terminal additions adapt EL222 for eukaryotic applications: a VP16 transcriptional activation domain (AD) and a nuclear localization signal (NLS) sequence. Also this is reflected by the above preferred embodiment which refers to a transcription activation domain and an NLS. The transcription activation domain is preferably a VP16 transcriptional activation domain from Herpes simplex.

[0044] The optogenetic gene expression system of the above preferred embodiment forms an active transcription complex when being exposed to the light pulses of a specific wavelength and the promoter region(s) of the one or more gene(s) that encode a protein controlling the progression of a cell through the cell cycle comprise(s) a binding site of the active transcription complex, preferably next to the minimal promoter(s) and upstream of the gene(s). A minimal eukaryotic promoter includes the transcription start site (TSS) and as elements directly upstream a binding site for RNA polymerase and general transcription factor binding sites, e.g. TATA box, B recognition element. A minimal promoter preferably comprises a TATA box.

[0045] The present invention relates in a second aspect to a yeast cell for the evolution of a protein of interest by applying a light-based selection pressure to the yeast cell, (i) wherein in the yeast cell the expression of one or more gene(s) that encode a protein controlling the progression of the yeast cell through the cell cycle is / are optogenetically controlled by an optogenetic gene expression system that is activated by light pulses of a specific wavelength, and wherein the protein of interest is the protein or one of the proteins constituting the optogenetic gene expression system; or (ii) wherein in the yeast cell the expression of one or more gene(s) that encode a protein controlling the progression of the yeast cell through the cell cycle is / are controlled by the protein of interest or a signal transduction cascade that comprises the protein of interest, and wherein the expression of the protein of interest is optogenetically controlled by an optogenetic gene expression system that is activated by light pulses of a specific wavelength.

[0046] In accordance with a preferred embodiment of the second aspect (i) in the yeast cell one or more gene(s) that encode a protein controlling the progression of the yeast cell through the cell cycle is / are only expressed under the control of an optogenetic gene expression system, wherein the light-sensing module is capable of sensing light pulses of a specific wavelength comprising a light-sensing module and a transcription factor module, wherein the transcription factor module comprises a DNA binding domain (DBD), a transcription activation domain and a nuclear localization sequence (NLS), and wherein the optogenetic gene expression system forms an active transcription complex when being exposed to the light pulses of a specific wavelength, wherein the promoter region(s) of the one or more gene(s) that encode a protein controlling the progression of a cell through the cell cycle comprised) a binding site of the active transcription complex, preferably next to the minimal promoter(s) and upstream of the gene(s), and wherein the one or more gene(s) that encode(s) a protein controlling the progression of a cell through the cell cycle is / are expressed and the yeast cell starts to progress through the cell cycle and divide if the active transcription complex binds to the binding site of the active transcription complex; or (ii) the yeast cell expresses an optogenetic gene expression system comprising a lightsensing module and a transcription factor module, wherein the light-sensing module is capable of sensing light pulses of a specific wavelength, and the transcription factor module comprises a DNA binding domain (DBD), a transcription activation domain and a nuclear localization sequence (NLS), and wherein the optogenetic gene expression system forms an active transcription complex when being exposed to the light pulses of a specific wavelength, wherein the protein of interest is or is a part of the optogenetic gene expression system, and wherein in the yeast cell one or more gene(s) that encode a protein controlling the progression of the yeast cell through the cell cycle is / are only expressed under the control of the optogenetic gene expression system, wherein the promoter region(s) of the one or more gene(s) that encode a protein controlling the progression of a cell through the cell cycle comprise(s) a binding site of the active transcription complex, preferably next to the minimal promoter(s) and upstream of the gene(s), and wherein the one or more gene(s) that encode(s) a protein controlling the progression of a cell through the cell cycle is / are expressed and the yeast cell starts to progress through the cell cycle and divides if the active transcription complex binds to the binding site of the active transcription complex; or (iii) the yeast cell expresses an optogenetic gene expression system comprising a light-sensing module and a transcription factor module, wherein the light-sensing module is capable of sensing light pulses of a specific wavelength, and the transcription factor module comprises a DNA binding domain (DBD), a transcription activation domain and a nuclear localization sequence (NLS), and wherein the optogenetic gene expression system forms an active transcription complex when being exposed to the light pulses of a specific wavelength; wherein in the yeast cell the gene encoding the protein of interest is under the control of the optogenetic gene expression system, wherein the promoter region of the gene encoding the protein of interest comprises a binding site of the active transcription complex, preferably next to the minimal promoter and upstream of the gene, and wherein the gene that encodes the protein of interest is expressed if the active transcription complex binds to the binding site of the active transcription complex, wherein in the yeast cell one or more gene(s) that encode(s) a protein controlling the progression of the yeast cell through the cell cycle is / are only expressed under the control of the protein of interest or a signal transduction cascade that comprises the protein of interest, wherein the one or more gene(s) that encode(s) a protein controlling the progression of a cell through the cell cycle is / are expressed and the yeast cell starts to progress through the cell cycle and divides if the protein of interest is expressed.

[0047] The preferred embodiments and definitions as described herein above in connection with the first aspect apply mutatis mutandis to the second aspect.

[0048] Also in accordance with the second aspect and the preferred embodiments thereof the yeast cell can also be in broader aspect any eukaryotic cell as described herein above in connection with the related aspect of the present invention.

[0049] It is to be understood that the second aspect of the invention is directed to genetically engineered cells that can be used for the evolution of a protein of interest by applying light-based selection pressure to a yeast cell. Accordingly, the invention is also directed to the use of cells of the second aspect for the evolution of a protein of interest by applying light-based selection pressure to a yeast cell.

[0050] The present invention relates in a third aspect to a kit for the evolution of a protein of interest by applying a light-based selection pressure to a yeast cell, wherein the kit comprises at least one yeast cell of the second aspect of the invention.

[0051] The preferred embodiments and definitions as described herein above in connection with the first aspect apply mutatis mutandis to the third aspect.

[0052] Also in accordance with the third aspect and the preferred embodiments thereof the yeast cell can also be in a broader aspect any eukaryotic cell as described herein above in connection with the related aspect of the present invention.

[0053] The kit preferably comprises in addition instructions how to us the kit, in particular for the evolution of a protein of interest by applying a light-based selection pressure to a cell.

[0054] The at least one cell can be packaged into a suitable vessel, such as a vial, tube or flask. The cells may be in culture medium or may be frozen. In the latter case the cells have to be thawed before use.

[0055] In accordance with a preferred embodiment of the second and third aspect the cell comprises a vector encoding in expressible from the one or more gene(s) that encode a protein controlling the progression of the yeast cell through the cell cycle that are - without the vector - only expressed from if suitable light (in particular a suitable light pulses) is applied that induces the expression of the one or more gene(s) that encode a protein controlling the progression of the yeast cell through the cell cycle.

[0056] By this vector it is possible to expand the cell without the suitable light. Once the cell has been sufficiently expanded the extrachromosomal vector can be removed from the expanded cells and the cells can be used for the evolution of a protein of interest by applying a light-based selection pressure to the cells, in particular by the method of the first aspect.

[0057] In this respect it is of note that cells tend to automatically lose an extrachromosomal vector after a few cell cycles.

[0058] In order to make sure that the method of the invention starts with a population of cells without the vector, the cells can optionally be grown in so-called counter-selective media which exists for a number of selection markers that can be integrated into vectors. Counter-selective media are media that select for cells without the vector. In particular and as illustrated by the appended examples, the URA3 gene which can be counter-selected for in media with 5-FOA (5-fluoroorotic acid) can be used. Hence, although cells tend to lose further measures can be taken to make sure that all cells lost the vector, preferably by isolating single cells on 5-FOA. For this the procedure is preferably as follows:

[0059] 1) growing cells under blue light pulsing and regular media (non-selective, non-counter-selective) to enable the proliferation of cells that lose the plasmid; and

[0060] 2) selecting a single colony on 5-FOA plate grown under pulsing blue light

[0061] In accordance with a preferred embodiment of all three aspects of the invention, the yeast cell is Saccharomyces cerevisiae.

[0062] In this respect is of note that the use of a Saccharomyces cerevisiae cell for the evolution of a protein of interest by applying a light-based selection pressure to the cell is illustrated by the appended examples.

[0063] In accordance with a more preferred embodiment of all three aspects of the invention, the one or more gene(s) that encode a protein controlling the progression of the yeast cell are one or more (or at least one) selected from CLN1, CLN2, CLN3, CLB1, CLB2, CLB3, CLB4, CLB5 and CLB6.

[0064] All of the genes CLN1, CLN2, CLN3, CLB3, CLB4, CLB5 and CLB6 encode cyclins of a Saccharomyces cerevisiae cell. G1-specific cyclins are encoded by CLN1, CLN2, and CLN3 and are required for the entry into the cell cycle (Start) and thereby for the S phase. G2-specific B-type cyclins encoded are by CLB1, CLB2, CLB3, and CLB4 are required for mitosis. CLB5 and CLB6 encode B-type cyclins that appear in late G1 along with those of CLN1 and CLN2 and are required for DNA replication (Schwon and Nasmyth, Genes Dev, 7(7A):1160-75).

[0065] The one or more selected from CLN1, CLN2, CLN3, CLB3, CLB4, CLB5 and CLB6 are preferably at least one of CLN1, CLN2 and CLN3 and at least one of CLB3, CLB4, CLB5 and CLB6, and are more preferably CLN2 and / or CLB5. At least one of CLN1, CLN2 and CLN3 are preferably all of CLN1, CLN2 and CLN3 and / or at least one of CLB3, CLB4, CLB5 and CLB6 are preferably at least three of the four and most preferably all four of CLB3, CLB4, CLB5 and CLB6.

[0066] One or more of the other cyclines that are not placed under the optogentic control are preferably deleted.

[0067] In this case it is of note that in the examples the preparation of a yeast stain is illustrated, wherein the expression of CLN2 and CLB5 is optogentically controlled and in the other CLN1,2,3 as well CLB3,4,5,6 are deleted (LIP-CLN2, LIP-CLB5 and cln1,2,3A c!b3,4,5,6 ), as well as cyclin-CDK inhibitor SIC1 is deleted. This strain is called Dajbog and used in the appended examples.

[0068] In accordance with a more preferred embodiment of all three aspects of the invention, the yeast cell in addition comprises a mutated polymerase instead of a wildtype polymerase having an increased mutation rate as compared to the wildtype polymerase, wherein preferably the mutated polymerase is the Saccharomyces cerevisiae POL3-L523D and the wildtype polymerase is Saccharomyces cerevisiae POL3.

[0069] The use of a mutated polymerase having an increased mutation rate has the advantage of accelerating the evolution of the protein of interest because random mutations in the protein of interest are introduced more often which increases the chances of the generation of mutant proteins and yeast cells with such a mutant protein being capable of optimally dividing under the light of (a).

[0070] The generation of mutant polymerases having an increased mutation rate is a method or routine and is, for example, described in Loh et al., Proc Natl Acad Sci U S A. 2010 Jan 19; 107(3): 1154-1159.

[0071] The mutant polymerase having an increased mutation rate is most preferably Saccharomyces cerevisiae POL3-L523D. The amino acid sequence of Saccharomyces cerevisiae POL3-L523D is SEQ ID NO: 4 and the nucleic acid sequence is SEQ ID NO: 5. The mutated polymerase therefore also preferably comprises or consists of an amino acid sequence of SEQ ID NO: 4 or a sequence being at least 80% identical thereto, or is encoded by a nucleic acid sequence of SEQ ID NO: 5 or a sequence being at least 80% identical thereto.

[0072] Mutant of POL3 that increases global mutation rate (L523D): SEQ ID NO: 4

[0073] MSEKRSLPMVDVKIDDEDTPQLEKKIKRQSIDHGVGSEPVSTIEIIPSDSFRKYNSQGFKAKDTDLMGT

[0074] QLESTFEQELSQMEHDMADQEEHDLSSFERKKLPTDFDPSLYDISFQQIDAEQSVLNGIKDENTSTVV

[0075] RFFGVTSEGHSVLCNVTGFKNYLYVPAPNSSDANDQEQINKFVHYLNETFDHAIDSIEVVSKQSIWGY

[0076] SGDTKLPFWKIYVTYPHMVNKLRTAFERGHLSFNSWFSNGTTTYDNIAYTLRLMVDCGIVGMSWITLP

[0077] KGKYSMIEPNNRVSSCQLEVSINYRNLIAHPAEGDWSHTAPLRIMSFDIECAGRIGVFPEPEYDPVIQIA

[0078] NVVSIAGAKKPFIRNVFTLNTCSPITGSMIFSHATEEEMLSNWRNFIIKVDPDVIIGYNTTNFDIPYLLNRA

[0079] KALKVNDFPYFGRLKTVKQEIKESVFSSKAYGTRETKNVNIDGRLQLDLLQFIQREYKLRSYTLNAVSA

[0080] HFLGEQKEDVHYSIISDLQNGDSETRRRLAVYCLKDAYDPLRLMEKLMALVNYTEMARVTGVPFSYLL

[0081] ARGQQIKVVSQLFRKCLEIDTVIPNMQSQASDDQYEGATVIEPIRGYYDVPIATLDFNSLYPSIMMAHNL

[0082] CYTTLCNKATVERLNLKIDEDYVITPNGDYFVTTKRRRGILPIILDELISARKRAKKDLRDEKDPFKRDVL

[0083] NGRQLALKISANSVYGFTGATVGKLPCLAISSSVTAYGRTMILKTKTAVQEKYCIKNGYKHDAVVVYGD

[0084] TDSVMVKFGTTDLKEAMDLGTEAAKYVSTLFKHPINLEFEKAYFPYLLINKKRYAGLFWTNPDKFDKLD

[0085] QKGLASVRRDSCSLVSIVMNKVLKKILIERNVDGALAFVRETINDILHNRVDISKLIISKTLAPNYTNPQP

[0086] HAVLAERMKRREGVGPNVGDRVDYVIIGGNDKLYNRAEDPLFVLENNIQVDSRYYLTNQLQNPIISIVA

[0087] PIIGDKQANGMFVVKSIKINTGSQKGGLMSFIKKVEACKSCKGPLRKGEGPLCSNCLARSGELYIKALY

[0088] DVRDLEEKYSRLWTQCQRCAGNLHSEVLCSNKNCDIFYMRVKVKKELQEKVEQLSKW

[0089] Mutant of POL3 that increases global mutation rate (L523D): SEQ OD NO: 5

[0090] ATGAGTGAAAAAAGATCCCTTCCCATGGTTGATGTGAAGATCGATGACGAGGATACTCCCCAGTT

[0091] GGAAAAGAAAATCAAACGGCAATCAATAGATCATGGTGTTGGAAGTGAACCTGTTTCAACAATAG

[0092] AGATTATTCCGAGTGATTCTTTTCGAAAATATAATAGTCAAGGCTTCAAAGCAAAGGATACAGATTT

[0093] AATGGGTACGCAATTAGAGTCTACTTTTGAACAAGAGCTATCGCAAATGGAACATGATATGGCCG

[0094] ACCAAGAAGAGCATGACCTGTCATCATTCGAGCGTAAGAAACTTCCAACCGATTTTGACCCAAGT

[0095] TTGTATGATATTTCTTTCCAACAAATTGATGCGGAACAGAGCGTACTGAATGGTATCAAAGATGAA

[0096] AATACATCTACCGTGGTAAGGTTTTTTGGTGTCACTAGTGAAGGACACTCTGTACTTTGTAATGTT

[0097] ACAGGGTTCAAGAACTATCTTTACGTCCCAGCGCCCAATTCTTCCGACGCTAACGATCAGGAGCA

[0098] AATCAACAAGTTTGTGCACTATTTAAACGAAACATTTGACCACGCTATTGATTCGATTGAAGTTGTA

[0099] TCTAAACAGTCTATCTGGGGTTATTCCGGAGATACCAAATTACCATTCTGGAAAATATACGTCACC

[0100] TATCCGCATATGGTCAACAAACTGCGTACTGCGTTTGAAAGAGGTCATCTTTCATTCAACTCGTGG

[0101] TTTTCTAACGGCACGACTACTTATGATAACATTGCCTACACTTTAAGGTTAATGGTAGATTGTGGA

[0102] ATTGTCGGTATGTCCTGGATAACATTACCAAAAGGAAAGTATTCGATGATTGAGCCTAATAACAGA

[0103] GTTTCCTCTTGTCAGTTGGAAGTTTCAATTAATTATCGTAACCTAATAGCACATCCTGCTGAGGGT

[0104] GATTGGTCTCATACAGCTCCATTGCGTATCATGTCCTTTGATATCGAGTGTGCTGGTAGGATTGG

[0105] CGTCTTTCCGGAACCTGAATACGATCCCGTCATCCAAATTGCCAACGTTGTGAGTATTGCTGGCG

[0106] CTAAGAAACCATTCATTCGTAATGTGTTTACTCTGAATACATGCTCACCCATAACAGGTTCAATGAT

[0107] >

[0108] TTTTTCCCACGCCACTGAAGAGGAAATGTTGAGCAATTGGCGTAACTTTATCATCAAAGTTGATCC TGATGTTATCATTGGTTATAATACTACAAATTTTGATATCCCTTATCTTTTAAACCGTGCAAAGGCG CTAAAGGTGAATGATTTCCCATATTTTGGAAGGTTAAAAACCGTTAAGCAAGAAATTAAAGAGTCT GTGTTCTCTTCGAAGGCTTATGGTACAAGAGAAACCAAAAATGTCAATATTGACGGCCGATTACA GTTGGATCTTTTGCAATTTATTCAGCGTGAGTATAAACTAAGATCCTACACGTTGAATGCAGTCTC TGCGCACTTTTTAGGTGAACAGAAGGAGGATGTACATTATAGCATCATTTCTGATCTACAAAATGG CGATAGTGAAACAAGAAGAAGGTTGGCCGTTTACTGTTTGAAAGACGCCTACGACCCTTTAAGGC TTATGGAAAAACTAATGGCGTTAGTTAACTATACAGAAATGGCTCGTGTTACAGGTGTGCCATTTT CATATTTACTAGCTCGTGGTCAACAAATTAAAGTTGTTTCTCAACTATTTCGAAAGTGCCTGGAGA TTGATACTGTGATACCTAACATGCAATCTCAGGCCTCTGATGACCAATATGAGGGTGCCACTGTTA TTGAGCCTATTCGTGGTTATTACGATGTACCGATTGCAACTTTGGATTTCAATTCTTTATATCCAAG TATTATGATGGCGCACAACCTATGTTATACAACACTTTGTAACAAAGCTACTGTAGAGAGATTGAA TCTTAAAATTGACGAAGACTACGTCATAACACCTAATGGAGATTATTTTGTTACCACAAAAAGAAG GCGTGGTATATTACCAATTATTCTGGATGAATTAATAAGTGCTAGAAAACGCGCTAAAAAAGATCT GAGAGATGAGAAGGATCCATTCAAAAGAGATGTTTTAAATGGTAGACAATTGGCTTTGAAGATTTC AGCTAACTCTGTCTATGGTTTTACAGGAGCGACGGTGGGTAAATTGCCATGTTTAGCCATTTCTTC ATCTGTTACTGCTTATGGTCGTACCATGATTTTAAAAACTAAAACCGCAGTCCAAGAAAAATATTGT ATAAAGAATGGTTATAAGCACGATGCCGTTGTGGTTTACGGTGACACTGATTCCGTTATGGTAAA GTTTGGTACAACAGATTTAAAGGAAGCTATGGATCTTGGTACCGAAGCTGCCAAATATGTCTCCA CTCTATTCAAACATCCGATTAACTTAGAATTTGAAAAAGCATACTTCCCTTACCTTTTGATAAATAA AAAGCGTTATGCAGGTTTATTCTGGACTAATCCTGACAAGTTTGACAAGTTGGACCAAAAAGGCCT TGCTTCTGTCCGTCGTGATTCCTGTTCCTTGGTTTCTATTGTTATGAATAAAGTTTTAAAGAAAATT TTAATTGAAAGAAATGTAGATGGTGCTTTAGCTTTTGTCAGAGAAACTATCAATGATATTCTGCATA ATAGAGTAGATATTTCAAAGTTGATTATATCAAAGACGTTAGCCCCAAATTACACAAATCCACAGC CGCACGCCGTTTTGGCTGAACGTATGAAGAGGAGAGAGGGCGTTGGTCCAAATGTTGGTGATCG TGTGGACTATGTCATTATCGGTGGTAATGATAAACTTTACAATAGAGCAGAAGATCCATTATTTGT ACTAGAAAACAATATTCAAGTGGATTCGCGCTATTATTTAACTAATCAATTACAAAATCCAATCATT AGTATTGTTGCACCTATTATTGGCGACAAACAGGCGAACGGTATGTTCGTTGTGAAATCCATTAAA ATTAACACAGGCTCTCAAAAAGGAGGCTTGATGAGCTTTATTAAAAAAGTTGAGGCTTGTAAAAGT TGTAAAGGTCCGTTGAGGAAAGGTGAAGGCCCTCTTTGTTCAAACTGTCTAGCAAGGTCTGGAGA ATTATACATAAAGGCATTATACGATGTCAGAGATTTAGAGGAAAAATACTCAAGATTATGGACACA ATGCCAAAGGTGCGCTGGTAACTTACATAGTGAAGTTTTGTGTTCAAATAAGAACTGTGACATTTT TTATATGCGGGTTAAGGTTAAAAAAGAGCTGCAGGAGAAAGTAGAACAATTAAGCAAATGGTAA

[0109] It is noted that many ways to elevate the mutation rates are available in the art; for example, using a chemical that induces mutations (e.g. alkylating agents such as ethylmethane sulfonate and N-methyl- N-nitrosourea that induce point mutations in DNA), proofreading mutants (noting that the extent of proofreading in DNA replication can determine the mutation rate), or orthogonal replication systems (see, for example, Tian et al., Nature Chemical Biology (2023), https: / / doi.org / 10.1038 / s41589-023- 01387-2).

[0110] In accordance with a preferred embodiment of all three aspects of the invention, the optogenetic gene expression system comprises as light-sensing module the light-sensing module of EL222, as DNA binding domain (DBD) the DBD of EL222, as transcription activation domain VP16, and / or as NLS the sequence PKKKRKV (SEQ ID NO: 6).

[0111] The entire stretch comprising the EL222 light-sensing module, DNA binding domain (DBD) of EL222, VP16, and the NLS of the optogenetic gene expression system preferably comprises or consists of the amino acid sequence of SEQ ID NO: 1 or a sequence being at least 80% identical thereto, or is encoded by SEQ: 2 or a sequence being at least 80% identical thereto.

[0112] SEQ ID NO: 1

[0113] MASPKKKRKVAPPTDVSLGDELHLDGEDVAMAHADALDDFDLDMLGDGDSPGPGFTPHDSAPYGALD

[0114] MADFEFEQ.MFTDALG\DEYGGEFGADDTRVEVQPPAQWVLDLIEASPIASVVSDPRLADNPLIAINQAFT

[0115] DLTGYSEEECVGRNCRFLAGSGTEPWLTDKIRQGVREHKPVLVEILNYKKDGTPFRNAVLVAPIYDDDDEL

[0116] LYFLGSQV'EV'DDDQPNMGMARRERAAEMLKTLSPRQLEVTTLVASGLRNKEVAARLGLSEKTVKMHR

[0117] GLVMEKLNLKTSADLVRIAVEAGI*

[0118] SEQ ID NO: 2 atggctagcccgaaaaaaaagagaaaagtagccccaccaacggacgtttcacttggtgacgaattacatttagacggtgaggatgt gaatcggagcagatgatactagagttgaagtccaacctccagcccaatgggtattggatttgatcgaggcctcacccatagca agtgttgttagtgatcccagattggctgataatcctttgattgctatcaaccaggcctttaccgatctaactgggtattctgagga agaatgtgttggtcgtaattgtagatttttggcaggatctggtactgaaccctggctaacagataagatccgtcaaggtgttcgt gagcataaacccgtgttggttgagattttgaattacaaaaaggatggtactccttttcgtaatgcagttttggtggcaccaatct atgatgacgatgatgaattactatacttccttggtagccaagtggaagtggatgacgatcaaccgaacatgggcatggcccgta gagaacgtgctgcggagatgttaaaaactctttcaccacgtcaattagaagtcacaactctggttgcatcaggcctaagaaataaa gaagtggctgccagactgggtctttcagaaaaaactgtcaaaatgcacagaggtttggtaatggaaaaattaaacttaaaaacg agtgcagatttagttagaattgccgtagaagcaggtatttaa

[0119] DNA binding domain is colored in bold

[0120] LOV (light sensing) domain is colored in italic

[0121] VP16 activation domain underlined

[0122] SV40 NLS is double-underlined Accordingly the light-sensing module of EL222 preferably comprises or consists of the part of the amino acid sequence of SEQ ID NO: 1 in italic or a sequence being at least 80% identical thereto, or is encoded by the part of SEQ: 2 in italic or a sequence being at least 80% identical thereto.

[0123] The DNA binding domain (DBD) of EL222 preferably comprises or consists of the part of the amino acid sequence of SEQ ID NO: 1 in bold or a sequence being at least 80% identical thereto, or is encoded by the part of SEQ: 2 in bold or a sequence being at least 80% identical thereto.

[0124] The transcription activation domain VP16 preferably comprises or consists of the part of the amino acid sequence of SEQ ID NO: 1 being underlined or a sequence being at least 80% identical thereto, or is encoded by the part of SEQ: 2 being underlined or a sequence being at least 80% identical thereto.

[0125] The NLS preferably comprises or consists of the part of the amino acid sequence of SEQ ID NO: 1 being double-underlined or a sequence being at least 80% identical thereto, or is encoded by the part of SEQ: 2 being double-underlined or a sequence being at least 80% identical thereto.

[0126] The sequence identity of at least 80% herein is with increasing preference and for each occurrence independently at least 85%, at least 90%, at least 95%, at least 97.5% and at least 99%. For determining the sequence identity preferably the BLAST (Basic Local Alignment Search Tool) program is used. BLAST is an algorithm and program for comparing primary biological sequence information, such as the amino-acid sequences of proteins or the nucleotides of DNA and / or RNA sequences. BLAST is one of the most widely used bioinformatics programs for sequence searching.

[0127] In the claimed method the promoter region preferably comprises or consists of the light inducible promoter sequence of SEQ ID NO: 3 or a sequence being at least 80% identical thereto:

[0128] SEQ ID NO: 3 tacgtgagttcgccagcttcgagtaggtagcctttagtccatgcgttataggtagcctttagtccatgcgttataggtagcctttagtc catgcgttataggtagcctttagtccatgcgttataggtagcctttagtccatgcttaagagacactagagggtatataatggaag ctcgacttccagcttggcaatccggtactgttggtaaagccaccgcggccgctaaaatc pC120; 5 x EI222 binding site in bold; lx EI222 binding site underlined and minimal promoter in italic

[0129] Accordingly the minimal promoter preferably comprises or consists of the part of the nucleic acid sequence of SEQ ID NO: 3 in italic. In accordance with a preferred embodiment of all three aspects of the invention, in the yeast cell in addition a gene that encodes a cyclin-CDK inhibitor is deleted, preferably wherein the yeast cell is Saccharomyces cerevisiae and the deleted gene that encodes a cyclin-CDK inhibitor is SIC1 .

[0130] A cyclin-dependent kinase inhibitor protein (cyclin-CDK inhibitor) is a protein which inhibits the cyclin / cyclin-dependent kinase complex (cyclin-CDK).

[0131] Sic1 , a protein, is a stoichiometric inhibitor of Cdk1-Clb (B-type cyclins) complexes in Saccharomyces cerevisiae. Because B-type cyclin-Cdk1 complexes are the drivers of S-phase initiation, Sic1 prevents premature S-phase entry.

[0132] As discussed in the appended examples, the deletion of a gene that encodes a cyclin-CDK inhibitor is advantageous because it favors the evolution of the protein of interest because it renders cells more sensitive to the light pulses that activate the expression of one or more genes that regulate the cell cycle.

[0133] In accordance with a preferred embodiment of all three aspects of the invention, the light pulses that differ from the light pulses that optimally activate expression I the formation of an active transcription complex differ from the light pulses that optimally activate the formation of an active transcription complex by their wavelength, light intension, pulse length and / or pulse frequency.

[0134] The light pulses that optimally activate expression I the formation of an active transcription complex preferably comprise blue light and a pulse length of 5x about 25 sec, about every 100 min. The term “about” is with increasing preference ±20%, ±10% and ±5%.

[0135] Accordingly in a more preferred embodiment of all three aspects of the invention, the light pulses that optimally activate expression I the formation of an active transcription complex comprise blue light and a pulse length of 5x about 25 sec about every 100 min, wherein the evolution towards more lightsensitive mutants and less sensitive light-mutants is achieved by increasing or decreasing the amount of light compared to the optimal duration of light, respectively; wherein the evolution towards mutants sensitive to different wavelengths be achieved by administering or providing the light of different wavelength.

[0136] In accordance with a preferred embodiment of all three aspects of the invention, the binding site of the active transcription complex is pC120.

[0137] The binding site of the active transcription complex pC120 preferably comprises or consists of the part of the nucleic acid sequence of SEQ ID NO: 3 in bold and under-lined in one to ten copies and preferably 4 to 6 copies, and more preferably comprises or consists of the part of the nucleic acid sequence of SEQ ID NO: 3 in bold.

[0138] In accordance with a preferred embodiment of all three aspects of the invention, the yeast cell is Saccharomyces cerevisiae, the genes CLN1, CLN2, CLN3, CLB3, CLB4, CLB5 and CLB6 are deleted in the yeast cell, and the binding site of the active transcription complex is pC120.

[0139] As discussed above, strain Dajbog is used in the appended examples. Dajbog is a Saccharomyces cerevisiae stain wherein the expression of CLN2 and CLB5 is optogenetically controlled and in the other CLN1,2,3 as well as CLB3,4,5,6 are deleted (LIP-CLN2, LIP-CLB5 and cln1,2,3A clb3,4,5,6A'). In addition the cyclin-CDK inhibitor SIC1 is deleted. The Dajbog cells also comprise a binding site of the active transcription complex pC120.

[0140] Also in this case the binding site of the active transcription complex pC120 preferably comprises or consists of the part of the nucleic acid sequence of SEQ ID NO: 3 in bold and under-lined in one to ten copies and preferably 4 to 6 copies, and more preferably comprises or consists of the part of the nucleic acid sequence of SEQ ID NO: 3 in bold.

[0141] The present invention relates in a fourth aspect to a mutated EI222 protein comprising or consisting of (a) the amino acid sequence of SEQ ID NO: 7 with one of the following amino acid substitutions Asp18Tyr, Val139Leu, Val41 Phe, Phe59Tyr, Cys75Arg, Gly80Arg, Gly80Glu, Gly82Val, Thr83Ala, Thr83Ser, Glu84Lys, Glu84Ala, Glu84Gly, Leu87lle, Leu87Pro, Thr88Pro, Asp89Tyr, Lys90Glu, Asp89Tyr, Lys99Glu, Val103lle, Val103Ph, Glu104Gly, He105Val, Asn107Asp, Tyr108His, Pro114Arg, Pro114Leu, Val121 Leu, Val121 Met, Leu132Pro, Phe115Ser, Gly136Ser, Ala157Val, Pro165Ser and Val201 Leu, or (b) an amino acid sequence which is at least 80% identical to the amino acid sequence of (a) provided that the one amino acid substitution is retained.

[0142] Also described herein are a nucleic acid molecule encoding the mutated EI222 protein of the seventh aspect, a vector (preferably an expression vector) comprising the nucleic acid molecule encoding the mutated EI222 protein of the seventh aspect and a host cell (preferably a yeast cell as described above) comprising said nucleic acid molecule or said vector.

[0143] Wild-type EI222 amino acid sequence: SEQ ID NO: 7

[0144] MGQDRPIDGSGAPGADDTRVEVQPPAQWVLDLIEASPIASVVSDPRLADNPLIAINQAFTDLTGYSEE ECVGRNCRFLAGSGTEPWLTDKIRQGVREHKPVLVEILNYKKDGTPFRNAVLVAPIYDDDDELLYFLG SQVEVDDDQPNMGMARRERAAEMLKTLSPRQLEVTTLVASGLRNKEVAARLGLSEKTVKMHRGLVM EKLNLKTSADLVRIAVEAGI

[0145] The above mutated EI222 proteins were obtained in the appended examples of the application by the method of the invention. The below Table 3 summarizes the EI222 mutants and provides information on the light pulses that are different from the light pulses optimally activating gene expression that were used to obtain these EI222 mutants:

[0146] Table 3: Overview on mutated EI222 proteins as obtained in the appended examples

[0147] Also in connection with the fourth aspect the sequence identity of at least 80% is with increasing preference at least 85%, at least 90%, at least 95%, al least 97.5% and at least 99%. For determining the sequence identity again preferably the BLAST (Basic Local Alignment Search Tool) program is used.

[0148] As regards the embodiments characterized in this specification, in particular in the claims, it is intended that each embodiment mentioned in a dependent claim is combined with each embodiment of each claim (independent or dependent) said dependent claim depends from. For example, in case of an independent claim 1 reciting 3 alternatives A, B and C, a dependent claim 2 reciting 3 alternatives D, E and F and a claim 3 depending from claims 1 and 2 and reciting 3 alternatives G, H and I, it is to be understood that the specification unambiguously discloses embodiments corresponding to combinations A, D, G; A, D, H; A, D, I; A, E, G; A, E, H; A, E, I; A, F, G; A, F, H; A, F, I; B, D, G; B, D, H; B, D, I; B, E, G; B, E, H; B, E, I; B, F, G; B, F, H; B, F, I; C, D, G; C, D, H; C, D, I; C, E, G; C, E, H; C, E, I; C, F, G; C, F, H; C, F, I, unless specifically mentioned otherwise.

[0149] Similarly, and also in those cases where independent and / or dependent claims do not recite alternatives, it is understood that if dependent claims refer back to a plurality of preceding claims, any combination of subject-matter covered thereby is considered to be explicitly disclosed. For example, in case of an independent claim 1 , a dependent claim 2 referring back to claim 1 , and a dependent claim 3 referring back to both claims 2 and 1 , it follows that the combination of the subject-matter of claims 3 and 1 is clearly and unambiguously disclosed as is the combination of the subject-matter of claims 3, 2 and 1 . In case a further dependent claim 4 is present which refers to any one of claims 1 to 3, it follows that the combination of the subject-matter of claims 4 and 1 , of claims 4, 2 and 1 , of claims 4, 3 and 1 , as well as of claims 4, 3, 2 and 1 is clearly and unambiguously disclosed.

[0150] The figures show.

[0151] Figure 1 . Dynamic, multi-state protein of interest (POI) can be evolved by coupling its activity to the cellcycle oscillator. In the cells designed for this purpose, the output of POI is transmitted to CLB5, one of the S-phase cyclins, while the endogenous copies of cyclins have been deleted to make Clb5 oscillations essential. CLB5 overexpression is known to trigger the passage through Start, lead cells through S phase, and induce the expression of mitotic cyclins CLB1 and CLB2. On the other hand, strong continuous CLB5 overexpression leads to cell cycle arrest, possibly due to inhibition of origin licensing. This makes Clb5 both necessary for proliferation, and toxic if continuously overexpressed, ensuring that during proliferation of cells, POI will not evolve to be continuously active. In such a system, if the POI switches between repeated on-and-off outputs cells will proliferate maximally. Any error, either in switching speed or state, will cause skipped cell cycles. This gives an advantage to mutants with better fidelity to the desired input-output relationships.

[0152] Figure 2. To compare the ability of different cyclins driven by the optogenetic EL222 / LIP system to trigger passage through the Start point of the cell cycle, the percentage of arrested cells that responded to an impulse of light by budding was quantified. Each bar represents a different experiment in which the portion of arrested cells that bud was scored. The genotype of the strain and the length of the light pulse is given below the bar. Error bars present standard deviation. LIP-CLB5 can be used to override cln1,2,3A (G1-) arrest, however with a much lower efficiency compared to LIP-CLN2. Increasing the duration of the light pulse does not increase the percentage of budded cells. Although LIP-CLB5 can initiate the cycle in the cln1 ,2,3A clb3,4,6A cells, simultaneously inducing LIP-CLN2 increases the portion of the cells that commit to the cycle from around 20% to around 70%. Hence, to ensure reliable control of the cell cycle using optogenetics during the evolution, CLN2 and CLB5 cyclins were simultaneously expressed in a pulsatile manner. Error bar represent 90% confidence interval.

[0153] Figure 3. Pulses of simultaneous Cln2 and Clb5 expression drive cells through consecutive division cycles. Cells arrested in G1 phase of the cycle respond to a pulse of light by budding. Later during the cycle, they proceed to mitosis, divide the nuclei (fluorescently labeled histones inside the cells marks the cell nuclei) and complete the cell cycle. The process is repeated by giving cells a new pulse of light. The genomic background of cells shown here is cln1,2,3A clb3,4,5,6A POL3- L523D HTB2-mCherry PGK1pr-EL222 LIP-CLN2 LIP-CLB5. The time with respect to the first light pulse is given in minutes on each of the figures. The scale bar is 5 pm.

[0154] Figure 4. To evolve different properties of the artificial cell cycle control system, cells grown under optimal pulses of light (top panel) are placed under regimes of pulsing light that select for a particular type of mutant proteins. 1 : Shorter-than-optimal pulses of light select for cells that have increased sensitivity to light. 2: Oppositely, one can decrease the sensitivity to light by placing cells under longer pulses of light and / or with light regime with light applied also between the main pulses. 3: Finally, different colors of the input light can be used to select for mutants that can be activated by colors of light other than usual.

[0155] Figure 5. Two in vivo continuous approaches for evolution of dynamic, multi-conformational proteins. A: Evolution on solid plates enables the screening of different mutations that allow cells to survive under selective light regimes. First, cells are grown to high-enough quantities under optimal light-environment. Cells are then transferred to a regime of light that selects for the desired property. Colonies that have grown are then transferred onto a plate with fresh nutrients and grown further. Finally, DNA is extracted from cells and the region of interest is sequenced. B: Alternatively, proteins are evolved in a well-mixed environment such as liquid culture in an Erlenmayer flask. This allows different mutants that arise during proliferation to compete against each other and the best one among them to win. Cells are kept in a exponentially growing phase by diluting them in fresh media periodically. After enough time, DNA is extracted from cells and the region of interest is sequenced.

[0156] Figure 6. The system for continuous directed evolution of dynamic, multi-conformation proteins is used to discover new versions of the EL222 transcription factor that have increased or decreased sensitivity to blue light. The variants discovered in evolution experiments were cloned and transformed into strains that contain the fluorescent reporter driven from light-inducible promoter (LIP). To characterize the effect of mutations, the strains that contain different EL222 mutants were placed under darkness; low levels of blue light (one pulse of 10 s every 5 min); or high levels of blue light (30 s pulse every minute). After 6 h of induction, the transcriptional output of the discovered optogenetic mutants was measured. For comparison of the discovered mutants, the existing sensitized mutants of EL222 were also included (strongLO I / 26and AQTrip22). Rows: mutations in EL222. Columns: Light-regimes used for inducing optogenetic proteins. Values: Level of expression of LIP-ymScarletl fluorescent reporter in cells with the EL222 variant subtracted by the value of wild-type cells containing only the fluorescent reporter (arbitrary units).

[0157] Figure 7. Variants of EL222 obtained in evolutionary experiments designed to select for lower basal activity show no detectable activity in darkness (basal activity, leakiness). The violin plots show histograms of values of fluorescence measured in cells kept in darkness that contained the specified version of EL222 and LIP-yEVenus as a reporter. The cells with desensitized versions of EL222 show fluorescence levels (violins 4 and 5) comparable to cells that do not contain transcription factor (violin 6). a.u. - arbitrary units.

[0158] Figure 8. The system for continuous directed evolution of dynamic, multi-conformation proteins is used to discover new versions of the EL222 transcription factor that have increased sensitivity to light of colors other than blue. The variants discovered in evolution experiments were cloned and transformed into strains that contain the fluorescent reporter driven from light-inducible promoter (LIP). To characterize the effect of mutations, the strains that contain different EL222 mutants were placed under darkness; or light with wavelengths higher than blue (with peaks of emission at 525 nm, 555 nm, 570 nm, 605 nm, 648 nm). After 6 h of induction, the transcriptional output of the discovered optogenetic mutants was measured. Rows: mutations in EL222. Columns: Light-regimes used for inducing optogenetic proteins. Values: Level of expression of LIP-ymScarletl fluorescent reporter in cells with the EL222 variant subtracted by the value of wild-type cells containing only the fluorescent reporter (arbitrary units).

[0159] The examples illustrate the invention.

[0160] Example 1 - Results

[0161] Conceptual requirements for evolving dynamic, multi-state, and computational properties As started a general theoretical analysis was conducted. It was considered how to evolve a POI with one or more inputs and two possible outputs, 0 or 1. As in electrical engineering, 0 and 1 are useful abstractions, here, representing molecular activity. They can be used to describe a wide range of directed evolution goals. These may include increasing, lowering, or changing the nature of the activity of either output state, or modifying the kinetics of switching between them. 0 or 1 need not refer to absolute zero or the maximally attainable activity (see Discussion; Example 2).

[0162] There can be, in principle, a single input / to the POI, for example, the transcriptional induction of the POI, or a complex input / =( / ),... , / „) such as specific phosphorylation patterns at n different sites of the POI. If each input channel is binary, there are 2ninput-output relations. In practice, the inputs are ideally controlled with optogenetics because light administration, even for weeks, is precise, dosable, cheap, and straightforward. Light input to the POI can be direct or through multiple intermediate steps. For example, optogenetic transcription factors, kinases, and degrons can mediate the input to the POI. Although light control was involved in a single screening step in past work (the general power of optogenetics for directing evolution has hitherto not been recognized (Discussion; Example 2). Directed evolution could in principle be similarly performed with chemical inputs to the POI instead of light input. However, the need for repeatedly changing the cell culture medium makes directed evolution with optogenetic control of the POI inputs, where feasible, much more attractive. As output, the POI may, for example, transcriptionally activate, phosphorylate, or degrade its targets.

[0163] With the inputs and outputs thus defined, the main conceptual challenge was to find a system which performs selection on each input-output relationship of the POI and the kinetics of the transitions. It was suppose that the simple or complex inputs are labeled h, fe, h, / , ... and correspond to desired POI outputs 1 , 0, 1 , 0, ... It was realized that a cell cycle oscillator can perform directed evolution on this system if it is made fully dependent on the POI’s output. In such a system, if the POI switches between 1-0-1-0-... outputs driven by repeated cycles of inputs 11-12-13-14-... cells will proliferate maximally. Any error, either in switching speed or state, e.g., 1-1 -1-0-... , will cause skipped cell cycles. This incurs a large fitness penalty, which gives an advantage to mutants with better fidelity to the desired input-output relationships. There is freedom in devising protocols for the A- / 2- / 3- / 4-... inputs. How to sequence the inputs optimally represents an interesting mathematical problem, which was solved analytically.

[0164] Tunability of the selection pressure is an important aspect of directed evolution methods. Because in the system that is conceptualized herein, the input strengths and timing are controlled by light, the selection pressure on each state and the kinetics of the transitions can be varied. So, the input-output relationships can, for example, be challenged or corrupted by light input that is too weak or too strong to be restored by mutational adaptations of the POI toward a desired goal. Also, different variants of the re-engineered cell cycle oscillator and the different conditions under which it can be run allow control over selection pressures on POI states and transitions (Discussion; Example 2).

[0165] Technical requirements To implement the above considerations, it was critical to make the cell cycle fully depend on the POI’s 1-0-1-0-... output. The key idea which was pursued was to find an element of the cell cycle, which would be required for one part of the cell cycle but poisonous for another. When engineered to be downstream of the POI, this element would drive one cell cycle for each 1-0 output oscillation of the POI.

[0166] The overall system needed to satisfy stringent requirements:

[0167] 1 . Tight coupling: Cell cycle transitions and arrest must be tightly dependent on the switching of the POI. Mutations that break the POI so that it would be in the 0 or 1 output state constantly should not lead to cell proliferation since this would circumvent the selection pressure.

[0168] 2. Mutational robustness: No (known) mutations in cell cycle genes should break the tight coupling either. Such mutations would otherwise allow bypassing the selection pressure.

[0169] 3. Reliability: A brief period of the POI’s 1 output should trigger one cell cycle transition and switching to 0 should allow the completion of the rest of the cell cycle in almost all cells. Repetitive 1-0 switches with appropriate timing should drive the cells through consecutive cycles.

[0170] 4. Tolerance to poor initial performance: POIs may work too poorly to drive the cell cycle initially. Genetic constructs, which are called ‘training wheel plasmids’ herein, should support cell proliferation initially and should be cleanly removable early in the directed evolution campaign.

[0171] 5. Usability: Any POI should be easily insertable in a plug-and-evolve fashion.

[0172] 6. Stability: Over months of directed evolution, cells should maintain a stable genome and not polyploidize, which makes genetic analyses difficult.

[0173] Implementation in budding yeast

[0174] The system was engineered in budding yeast because its cell cycle is relatively fast (=90 min period in glucose medium) and because it was possible to leverage decades of research in yeast cell cycle control. A short generation time is important since it sets the maximum speed with which selection pressure can be exerted on POI switching. Even faster 1 -0-1-0... POI output oscillations would put unwanted selection pressure on cell physiology.

[0175] In search of a cell cycle element to put under the control of POIs, one might consider cell cycle control regulators, which oscillate once per cell cycle. However, it is not obvious which to choose since the cell cycle control system is remarkably robust. For example, constitutive expression of CLN2 and CLB2 in the absence of all other Cdk1 cyclins (cln1-3 clb1-6 ) suffices for cell cycles27. So, even in a highly stripped-down cell cycle control system, the two required Cdk1 cyclins did not need to oscillate transcriptionally.

[0176] However, past experiments showing that constant transcriptional induction suffices for cell proliferation were often performed with relatively weak constitutive promoters such as GALL26. So, a strategy was pursued, in which a particularly potent element of cell cycle control was searched for that is needed for one part of the cell cycle but poisonous to another when expressed strongly. Once found, one would engineer cells to be more sensitive both when the cell cycle regulator was needed in the cell cycle and to when it had to be absent. For the search, the relatively strong, blue light-sensitive LOV transcription factor EI22225was used. EI222 controls the light-inducible promoter (LIP), in which the protein binding sites (5xC720)are upstream of a TATA box28. LIP was placed upstream (5’) to the ORF of the cell cycle control regulator gene whose transcription it was wished to control (Table 1 , column 1). Endogenous genes were knocked out conditionally as needed to make the cell cycle dependent on the candidate cell cycle control regulator and for the ability to proliferate cells before the tests (Table 1 , column 2).

[0177] Table 1 : Tests to find cell cycle regulators which are needed for one part of the cell cycle but toxic for another in specifically engineered genetic backgrounds. During tests, the chemical-responsive promoters were turned off MET3pr or GALL. or testing the LIP-CLB5 construct, the clnA background was used. For optovolution, the more robust clnA clb3-6A sic1A LIP-CLB5 LIP-CLN2 background was developed (see main text).

[0178] It was started with the key components of the core cyclin-Cdk / APC negative feedback loop of the cell cycle. First, optogenetic control of CLB2 was tested, the major M cyclin in budding yeast29in a clb1 ,2A background. However, continuous expression of CLB2 with EI222 / L / P did not prevent the cells from cycling. Since Clb2 activity has to be reduced in order to allow the cell cycle exit30, it was next tested whether CLB2Akd, an undegradable version of the Clb2 cyclin30was more suitable. Clb2kd indeed continued to be needed to enter mitosis in a clb1 ,2A background. However, a regime of pulsing light under which cells performed multiple cell cycles was not found. After the first mitosis, Clb2kd blocked mitotic exit nearly permanently because presumably Sic1 inhibition and dilution did not suffice to reduce Clb2kd concentrations. Furthermore, controlling the cyclin-CDK inhibitor CDC20, whose deletion leads to metaphase arrest, was similarly not suitable since its overexpression did not arrest cells.

[0179] The cell cycle regulators discussed thus far control mitotic transitions, where arrest and release often induces aneuploidy. The focus was next shifted to cyclins which control the G1 to Start transition, which is less perilous for genomic instability. Again, constitutive expression of the G1 / S cyclin CLN2 from EI222 / L / P in a clnA background did not kill cells. Thus, it was turned to the S-phase cyclin CLB5. LIP- CLB5 was first tested in a clnA strain since CLB5 expression is known to override a clnA G1 arrest, although with lower efficiency than G1 / S cyclin CLN23In darkness, cells were arrested in G1. Light pulses with a 90 min period led to repeated cell cycles. Surprisingly, it was found that constitutive CLB5 expression by EI222 / LIP was lethal. (In contrast, GAL1 pr-CLB5 overexpression in galactose medium, as opposed to glucose medium, where cell cycles are substantially slower, is tolerated32.) Thus, CLB5 was a promising candidate for creating a tight coupling between the POI and the cell cycle. The system fulfilled requirement 1 .

[0180] Next, the system was engineered for mutational robustness (requirement 2). To further increase the cells’ dependence on LIP-CLB5 expression, the four endogenous CLB3,4,5,6 copies were deleted, which would be functionally redundant with the light-inducible CLB5.2933No mutations are known that can by-pass the requirement for Clb5 in a cln1,2,3A clb3,4,5,6A background34.

[0181] By deleting CLB3-6 to make the strain more strongly depend on LIP-CLB5 induction, the strain was made less tolerant to continuous LIP-CLB5 expression. This would further require Clb5 to oscillate on and off for proliferation. This was accomplished by deleting SIC1 , a B-type cyclin inhibitor. SIC1 is part of a circuit that can oscillate when cyclin levels are constant, thus rescuing such cell cycles27. When SIC1 is deleted, the clnA arrest becomes leaky. However, this depends on at least one of CLB3-6 being present, which is not the case here. Cln1,2,3A clb3,4,56A sic1A cells continued to be tightly blocked in darkness.

[0182] The reliability of cell cycle initiation in response to LIP-CLB5 pulsing was low in this strain, violating requirement 3. Already, only 30% of clnA cells budded in response to a LIP-CLB5 pulse. In clnA clb3,4,6A cells, light-induced LIP-CLB5 only led to budding in 15% of dark-arrested cells. This portion was much higher (>90%) for clnA cells with LIP-CLN2. Increasing the amount of administered light did not change the fraction of cells that committed to the cell cycle. Guided by these results, it was sought to increase the reliability of the response to a light pulse by adding a LIP-CLN2 construct, which does not undermine the dependence on LIP-CLB5 for S phase and triggering entry into mitosis in the clnA clb3-6A background. Simultaneous expression of LIP-CLN2 and LIP-CLB5 in the clnA clb3,4,6A background did indeed lead to >70% of cells budding in response to a 20-min pulse of light, satisfying requirement 3.

[0183] To enable routine propagation of these cells prior to directed evolution campaigns, a ‘training-wheel plasmid’ was created (requirement 4). This plasmid allowed cells to grow without exogenously induced Clb5 oscillations, and the directed evolution begins with counter-selection against this plasmid, a URA3- marked CEN / ARS plasmid containing a CLB5 copy and MET3pr-CLN2 was used, which can be selected against with 5-FOA. Finally, to increase the mutation rate for directed evolution, the POL3-L523D mutation was introduced that increases the global mutation rate about hundred fold35.

[0184] In summary, tight coupling of the cell cycle machinery to externally applied periodic on-off light signals using Clb5 in a clnA clb3-6A sic1A background was achieved. Clb5 induced Start, S phase, and triggered mitotic entry. Simultaneously induced CLN2 substantially increased the fraction of cells that initiated a cell cycle in response to the light pulses. Once Clb5 and Cln2 are no longer induced, they are degraded and diluted, allowing completion of each cycle and resetting the cell state for a new cycle. In darkness and low light, due to the lack of G1 and G1 / S cyclins, cells arrest in G1 . Under constant light, cells could not proliferate as expected due to Clb5 blocking origin of replication relicensing, especially enhanced due to the SIC1 deletion. The mutation rate was further increased globally and allowed the strain to proliferate without Clb5 oscillations using a training wheel plasmid until the start of the directed evolution campaign. In principle, any POI controlling Clb5 activity can be inserted for directed evolution (requirement 5). In our tests, strains were never found to polyploidize, allowing straightforward genetic analysis (requirement 6). The strain fulfilled the requirements for the evolution of dynamic proteins. This strain was named Dajbog, because external control over periodic cycles was necessary for the cells’ proliferation possible.

[0185] Independently of the use for directed evolution, researchers have previously aimed to entrain the cell cycle-clock to external pulses of a chemical or light36-38. To our knowledge, this is the first case where cell cycle is enslaved by an external oscillator in an evolutionary robust manner.

[0186] Directed evolution of a LOV-transcription factor

[0187] The Dajbog strain can be generally used for evolving dynamic, multi-state, and computational functionalities. The method was applied to optogenetic systems, which are relay systems for an external on or off light state to on or off cellular activity. They have at least two input-output relationships. Optogenetic systems have been challenging to evolve, requiring repeated, sequential campaigns for the on and off states since mutations improving one state may destroy the other. Given that EI2222539is itself widely used and that LOV domains are of broad interest, it was chosen to evolve EI222 toward three different goals, i) more light sensitivity for stronger output with less blue light, ii) less leakiness, for example, for controlling genotoxic genes, or iii) changing the excitation spectrum for multiplexing and reduced phototoxicity.

[0188] The Dajbog strain was used with EI222 directly controlling CLN2 and CLB5 transcription, that is, in the same configuration as was used to develop the method. Cells initially proliferated under a permissive light pulse regime (5 times 25 sec on, 35 sec off every 100 min). Then, the light pulse regime was changed to apply specific selection pressures, in which cells with wild-type EI222 could not proliferate. With cells spread on agar plates, colonies emerged after about two days.

[0189] For characterizing the evolved EL222 variants, they were cloned and introduced as a single copy in a wild-type strain where LIP drives the transcription of a bright, yeast-optimized ymScarletl fluorescent protein gene. Thus, mutations elsewhere in the genome could not affect our characterizations. EL222 variants in these strains were driven from a constitutive PGKIpr.

[0190] First, it was decided to enhance the 1 -output (on) state of EI222. To evolve more light-sensitive mutants, cells were exposed to very brief light pulses (1 sec on every 100 min). In the mutant colonies that grew under these conditions, 12 functional mutants were found. Thus, the yield of interesting mutants was substantial. Under low light conditions (20% of maximal light intensity), these EI222 variants responded faster and to a higher final level than wild type. Under strong light induction (80% of maximal light intensity), the sensitized mutant and wild-type EI222 reached comparable levels of output, presumably corresponding to saturation.

[0191] Next, it was sought to reduce the leakiness of EI222 / L / P, that is, suppress activity in the O-output (off) state. Cells were exposed to very long duration pulses of light (40 times 30 sec on, 30 sec off every 100 min) in addition to continuous, dim background light. The mutants that grew under these conditions showed substantially reduced leakiness, comparable to transcription from LIP in the absence of EI222. They were also overall less strongly induced by light.

[0192] Finally, it was sought to modify the nature of the 1 -output state. In order to evolve EI222 to respond to other colors of light, cells were grown under green and red light (40 times 40 sec on, 20 sec off every 100 min). The ability to change this property is particularly interesting because LOV domains are thought to be hard to shift in terms of excitation light, due to rigidity of the interactions between the protein and the flavin chromophore39.

[0193] In total, 35 mutations were discovered that mostly clustered around the LOV photoreceptive core, 1 in the Ja helix and 2 in the DNA binding domain.

[0194] LOV domain proteins such as EI222 are ubiquitous across the tree of life and control a variety of cellular processes40-46. Most approaches to altering LOV-optogenetic tools have been by introducing mutations from known slow- or fast-cycling natural LOV domains22 2647. The novelty of the mutants was checked by searching a catalog of 7000 known and putative LOV-domains48. It was found that the majority of the mutations occur rarely (<10%). in natural proteins with LOV domains. These findings demonstrate yet again the power of directed evolution for identifying variants that are not frequent or at all present in nature. The found mutations are, although valuable, unlikely to be characterized otherwise.

[0195] Long-term optovolution of sensitized LOV domains

[0196] The directed evolution campaigns described thus far were designed to be unbiased: Any EI222 mutant which emerged and which could switch on and off repeatedly for =3 days under the respective selection pressure would generate a colony and be detected by us. Next, it was tested whether optovolution could be used for directed evolution over longer periods of time in liquid culture, where strains with different mutations could compete against one another. The fittest mutant would take over the cell culture at the end.

[0197] It was again sought to boost the 1 -output activity of EI222, this time over one week (Methods; Example 3). The Dajbog strain was evolved under 5 s of light, administered every 100 min. Two mutations were observed that won in two different long-term experiments, Glu84Asp and Val121 Met, which were already discovered on agar plates.

[0198] Comparison with the URA3 / 5-FOA system

[0199] It was decided to compare optovolution to other approaches that have been pursued for evolving multistate proteins, which rely on sequential directed evolution campaigns for each of the output states. Note that in optovolution, alternative selection for the 0- and 1 -outputs as well as for O-to-1 and 1-to-0 switches occurs once per cell cycle, i.e., every =90 min. The URA3 / 5-FOA system represents a reasonable comparison for optovolution since it requires input-dependent proliferation under two opposite selection pressures: growth needs induced expression of URA3 in -Uracil medium and suppression of URA3 in 5-FOA medium. This method was used in the directed evolution of the CRY2-C / B15optogenetic system. For this comparison, a budding yeast strain was constructed in which URA3 is transcribed by the EI222 / L / P system in a genetic background where the endogenous URA3 ORF was entirely deleted. The LIP-URA3 construct was validated by confirming that the cells were dead in -Uracil medium in darkness and viable in -Uracil medium under 30 sec on, 30 sec off pulsing blue light. However, cells were also dead in 5-FOA medium in darkness, presumably due to the leakiness of EI222ZL / P. (Note that leakiness was already minimized by inserting the LIP-URA3 construct and EL222 gene as single copies.) URA3 codes for orotidine 5'-phosphate decarboxylase, one of the most proficient enzymes49, potentially explaining how even low basal transcription may not be tolerated in 5-FOA medium.

[0200] Thus, it was decided to not work to boost the 1 -output (on) state of EI222 because the counter-selection step for low O-output (off) activity should not yield any surviving mutants. Instead, it was sought to leverage this sensitivity to O-output activity to evolve less leaky EI222. An experiment in which LIP-URA3 PGK1pr-EL222 ura3A0 cells that grew on 5-FOA in darkness are selected was performed. However, after sequencing 200 strains, no mutations in EL222 were found that were resistant to 5-FOA. Instead, as expected, some strains had mutations in the URA3 gene.

[0201] Example 2 - Discussion

[0202] Directed evolution is a powerful tool for biological engineering. Given appropriate selection pressures, directed evolution can solve challenging engineering problems and uncover unexpected solutions -- such as EI222 mutants that respond to light other than blue. Here, the first method that is germane for directed evolution of dynamic, multi-state, and computational protein functionalities is presented. Such capabilities are fundamental to signal transduction - and thus to life. While much of cellular biochemistry is regulated, it had been unclear how proteins with such properties could be evolved in the laboratory. Apart from the fundamental conceptual puzzle, this problem needed to be solved for protein engineering, where increasingly sophisticated computational systems are being designed, which require switchable output50.

[0203] The limitation being sought to overcome is that current approaches evolve only one functionality at a time. Properties that are not under selection pressure can consequently degrade during directed evolution experiments. Not only does the existing solution to this problem, to run directed evolution campaigns for each functionality sequentially, require repeated interventions and thus labor, it is also inherently challenging. The leakiness of EI222 prevented us from evolving the 1 -output state using the URA3 / 5-FOA method. Then, in attempting to reduce the leakiness of EI222, no mutations in EI222 in 200 5-FOA- resista nt colonies were found. It was thus substantially more likely to evolve non-functional mutants than leakiness-reducing mutations. This cannot happen with optovolution.

[0204] In contrast to directed evolution approaches that are not continuous, our method by-passes manual screening for each round. In optovolution, both 0-1 and 1-0 switching of the POI are screened every =90 min by the cell cycle control machinery in vivo. The coupling of the desired POI properties to organismal fitness automatically ensures that the evolved variants are functional inside cells. Additional screening in vivo is not needed. Light control is powerful for directed evolution because it allows communicating with cells extremely quickly, for extended periods, and conveniently. Thus, new kinds of selection pressures can be created, including the need for a POI to output 1 -0-1-0- as was exploited here. As far as the inventors know, this is the first time that optogenetics has been leveraged for continuous, self-selecting directed evolution.

[0205] Furthermore, light control allows tuning the fitness landscape of the POI and thus of the selection pressure, enabling different evolutionary outcomes depending on the light protocol. The same strain was used, where EI222 directly controlled CLB5 and CLN2, to enhance the 1 -output activity of EI222, to suppress the O-output activity, or to change the excitation color.

[0206] There is a multitude of possibilities for adjusting the system for different directed evolution campaigns. Decades-long research in cell cycle control and a vast repertoire of tools from yeast genetics can be leveraged:

[0207] 1 . Restoring SIC1 , which made the system highly sensitive to residual Clb5, allows POIs to be evolved that are initially more leaky.

[0208] 2. The carbon source can be varied to slow down the cell cycle to match POIs that switch poorly.

[0209] 3. To vary the strength of the POI initially, the POI gene can be introduced and integrated in the yeast genome in exactly one copy, specific higher numbers of copies, or on 2u plasmids in about 50 copies.

[0210] 4. Similarly, to force evolution of the POI instead of the optogenetic system mediating input to the POI, the optogenetic system can similarly be inserted in multiple copies, minimizing undesired mutations.

[0211] 5. A plethora of degrons can be added to reduce the activity of a POI.

[0212] Using the Dajbog strain, mutations in the POI gene were easily identified by sampling evolved strains. Nevertheless, the system can in principle be further optimized to block mutations not in the POI gene. Should by-pass mutations occur in non-essential genes, one can in principle block this avenue by deleting the gene in case of gain-of-function mutations. Alternatively, one can in principle insert additional gene copies in case of loss-of-function mutations for non-essential and essential genes. The deep literature on cell cycle control is of advantage for improving optovolution further.

[0213] A general challenge with directed evolution is that a POI may be so poor with respect to a desirable function that the directed evolution campaign cannot even be initiated. This is mitigated in two key ways in our method: Crucially, the selection pressure can be straightforwardly externally tuned by changing the light pulse regime, e.g., shorter, longer, dimmer, or brighter light pulses. This is not feasible to implement in other directed evolution methods. Furthermore, the training wheel plasmid allows proliferation in the absence of any POI function, and directed evolution can be initiated with one round of selection for functional mutants upon induced loss of the training wheel plasmid on 5-FOA.

[0214] In principle, any protein that can be coupled to Clb5 is evolvable using optovolution. While it was focused on transcription factors here (EI222), Clb5 protein levels could in principle be modulated by other means, including by localization and degradation. (The use of destruction-box deficient Clb5db opens up additional possibilities.) While opening up new possibilities for evolution by maintaining the dynamic responsiveness of the evolved molecules, optovolution does not preclude the evolution of more classical phenotypes of interest. Altered ligand-dependencies or substrate-specificities can be targets of optovolution.

[0215] Mutation rates are a key element of directed evolution. Our work has been orthogonal to advances in increasing mutation rates specifically for genes of interest. These are needed when mutations elsewhere in the genome can bypass the desired POI functionality. However, Dajbog was a tight genetic system, which did not need focusing mutation rates to achieve our directed evolution goals. The partial lack of proofreading using POL3-L523D increased the genome-wide mutation rate to about 0.5x10-7per basepair per generation3551. This means that in ~15 ml of a yeast population at OD=0.1 , all single mutants of the POI will be present. Straightforward ways to increasing the mutation rates globally are the use of DNA damaging radiation and chemicals, although not all nucleotide mutations can be attained in this manner. Another straightforward method for very high rates of focused mutations is to introduce the POI on mutagenized plasmid libraries. Nevertheless, our method is extendible to using targeted high- mutation rate systems such as OrthoRep10, MutaT712and the system for plasmid recombination52, which generate the mutational diversity at a higher rate.

[0216] Optovolution is particularly timely for the revolution that is currently taking place in Al-based protein design. There is a substantial interest in proteins with dynamic, multi-state (switching), and computational properties. Prototypes that have been computationally designed can be screened and improved in a high-throughput mannerwith optovolution. There is no similarly germane method available for these protein functionalities currently. Additionally, machine-learning guided methods will keep relying on experimentally validated training sets, which can be further enriched through directed evolution53. Optovolution is expected to provide solutions to previously otherwise intractable engineering problems and pave the way for more sophisticated engineered proteins.

[0217] Example 3 - Methods Plasmid construction

[0218] Cloning was performed using restriction enzyme digestion and T4 ligase ligation (New England Biolabs) or using Gibson assembly (New England Biolabs). Plasmid propagation was done in DH5alpha E. Coli strain. The strains were routinely propagated in LB medium supplemented with ampicillin.

[0219] Strain construction

[0220] The clnA clb3-6A strain was obtained through several routine crosses between strains that contained some of the deletions. The main strain used for evolution was constructed from a clnA clb3-6A strain that contained the URA3-marked training wheel plasmid (MET3pr-CLN2, CLB5pr-CLB5, CEN / ARS). The WT POL3 gene was first replaced with a variant that increases the mutation rate (POL3-L523D). After that, it was selected for the loss of the training wheel plasmid. First, the strain was grew on plates containing methionine, which shut offs MET3pr, for a day under pulsing light (25 s ON every minute, 5 times every 100 min). After that, it was transferred to 5-FOA plates containing methionine, under the same pulsing light regime. Isolated colonies were picked in which the loss of the training wheel plasmid was confirmed by the lack of growth in darkness on plates lacking methionine, or lacking uracil. From there onwards, the strain was propagated under pulsing light. The used light-regime was five pulses every 100 min, where each pulse was 25 s ON, 35 s OFF. To delete the SIC1 gene, the strain was transformed with a plasmid that was designed to remove the genomic SIC1.

[0221] To measure the activity of the evolved EL222 variants, the strain was constructed that contained a single-copy LIP-ymScarletl or LIP-ymScarletl-PEST integrated in the genome as a reporter. This strain was then transformed with variants of evolved EL222 by digesting the plasmids with Pmel. Integration of transcription factors in single copies was confirmed by PCR in all transformants.

[0222] Transformations of the yeast strains were done using the high-efficiency LiAc / DNA carrier / PEG (polyethylene glycol) protocol54. All used strains were W303 derived.

[0223] Media

[0224] 2% glucose synthetic complete media (SCD) supplemented with essential nutrients was used55. For preparing solid SCD plates, 2% agar was used. For the experiments with 5-FOA (‘Comparison with the URA / 5-FOA system for evolving dynamic proteins’ section) the standard recipe with 0.1 % 5- FOA was used56’57.

[0225] Light conditions

[0226] Throughout the experimental procedures, blue-light sensitive cells were manipulated under red light and they were otherwise kept in aluminum foil.

[0227] For administering blue light that induced EL222, commercial blue light panels were used (HQRP, USA). Panels were composed of 15x15 square array of diodes, where each diode had light intensity of 0.35 mW. Diode panels were placed inside incubators, at a distance of 20 cm above the agar plates with yeast.

[0228] For evolving EL222 variants that are responsive to light other than blue, diode arrays built of diodes with emission peaks at 525 nm, 555 nm, 570 nm, 605 nm, 648 nm were used.

[0229] To adjust the pulsing regime of the diodes, Nl USB 6501 (Texas Instruments, USA) controller was used. The controller was continuously run by a Matlab script. The regime of pulsing light could be adjusted with a resolution of up to one second.

[0230] “One-shot” continuous optovolution

[0231] Prior to evolution, cells were propagated on agar plates containing SCD, under five 25 s ON / 35 s OFF pulses of blue light every 100 min, until growth saturation. To further enrich the diversity of the population, cells were then mutagenized by exposing them for 10 s to UV light with 254 nm emission peak (model G30T8, Sankyo Denki, Japan). Cells were scraped from the plate and resuspended in 1000 pL of warm liquid SCD. After that, on each fresh SCD agar plate, 100-200 pL of cells mixture were plated and adjusted the selection pressure by updating the regime of light (for details on used light regime see main text). After2 days of growth, the evolving populations were replica plated to provide fresh nutrients. After another 2 days of growth, isolated colonies were picked by restreaking them on fresh SCD plates. Such strain patches were then grown under the same light regime as the ones they evolved on for additional two days.

[0232] After the selection of variants, the test was performed in darkness that enabled us to filter out the cells with mutations that bypass the need for light control (e.g. due to leaky variants of EL222). For this, two consecutive replica plates were performed, to reduce the starting number of cells. The absence of growth was confirmed by comparing evolved cells with WT cells propagated under the same conditions after two days. The phenotype of cells was evaluated by observing G1-arrested unbudded cells under the microscope.

[0233] Finally, once having the strains of interest, it was proceeded with the identification of mutations by sequencing. The whole process took about 10 days per screen.

[0234] Long-term continuous optovolution

[0235] Same as for the “one-shot” continuous evolution, cells were propagated on SCD plates and UV- mutagenized right before the evolution. Cells were scrapped from the plate and resuspended them in 1000 pL of liquid SCD. Initial cultures were obtained by placing 200 pL of resuspended cells into 100 mL of warm liquid SCD. For liquid culture evolution, transparent 250 mL Erlenmayer flasks were used (VWR, Switzerland). Cells were grown in shaking incubator (Infors HT, Switzerland) with rotation of the stage set to 250 rotations per minute and temperature to 30 C. For administering pulsing light, the same blue-light panels as for the “one-shot” evolutionary experiments were used, mounted on a custom-made holder inside the incubator.

[0236] To prevent the cultures from growing to saturation, which might prevent homogeneous light penetration, they were diluted every 12 h during the experiment. Cells were diluted into 100 mL of warm SCD media to a final OD (660 nm) of 0.04. The experiment was performed for a week, after which it was proceeded to sequencing using the genomic DNA from the bulk culture as a template.

[0237] Mutation identification

[0238] To screen for mutations Sanger sequencing (Microsynth AG, Switzerland) was performed using primers that span the coding sequence of interest. The identified mutations were confirmed by sequencing the plasmids after cloning the gene variants.

[0239] Microscopy and image analysis

[0240] Images were recorded using a Nikon Ti2-E (Nikon, Switzerland) microscope equipped with a 60x objective and a Hamamatsu Orca-Flash 4.0 camera (Hamamatsu, Japan). The microscope was operated using NIS-Elements software and the objective’s axial position was controlled by Nikon Perfect Focus System. To not induce EI222 while observing the cells under the microscope, a high-pass green interference filter (Nikon, Switzerland) placed between the diascopic light and the sample was used. When performing timelapse microscopy to monitor the dynamic properties of EI222, to induce the optogenetic protein diascopic light was used, as previously established26.

[0241] Image analysis was performed using YeaZ, a Python-based tool for yeast cell segmentation58. To calculate the level of the fluorescent reporter, the autofluorescence of unlabeled strain was subtracted from the fluorescence of the strain harboring the reporter.

[0242] Quantifying the robustness of the optogenetic cell-cycle control

[0243] To score the portion of cells that bud in response to a pulse of light, either strains with clnA MET3pr- CLN2; or clnA clb3-6A MET3pr-CLN2 CLB5pr-CLB5 genetic background were used. After loading cells into the microfluidic chip and growing them for 4-5 h, methionine was supplied which shut off MET3pr. Two hours after methionine has been added, it was verified that cells are large and unbudded which corresponds to G1 arrest. Different optogenetic constructs (LIP-CLB5 and LIP-CLN2) were pulsed by using diascopic light from the microscope set to 80% of maximal intensity for 20 min or 40 min, depending on the experiment. Cells were imaged every 10 min. References

[0244] 1 . Joo, H., Lin, Z. & Arnold, F. H. Laboratory evolution of peroxide-mediated cytochrome P450 hydroxylation. Nature 399, 670-673 (1999).

[0245] 2. Moore, J. C. & Arnold, F. H. Directed evolution of a para-nitrobenzyl esterase for aqueous-organic solvents. Nat Biotechnol 14, 458-467 (1996).

[0246] 3. Kuchner, O. & Arnold, F. H. Directed evolution of enzyme catalysts. Trends in Biotechnology 15, 523-530 (1997).

[0247] 4. Crameri, A., Whitehorn, E. A., Tate, E. & Stemmer, W. P. C. Improved Green Fluorescent Protein by Molecular Evolution Using DNA Shuffling. Nat Biotechnol 14, 315-319 (1996).

[0248] 5. Campbell, R. E. et al. A monomeric red fluorescent protein. Proc. Natl. Acad. Sci. U.S.A. 99, 7877-7882 (2002).

[0249] 6. Li, Y. et al. Directed evolution of human T-cell receptors with picomolar affinities by phage display. Nat Biotechnol 23, 349-354 (2005).

[0250] 7. Xu, L. et al. Directed Evolution of High-Affinity Antibody Mimics Using mRNA Display. Chemistry & Biology 9, 933-942 (2002).

[0251] 8. Esvelt, K. M., Carlson, J. C. & Liu, D. R. A system for the continuous directed evolution of biomolecules. Nature 472, 499-503 (2011).

[0252] 9. Crook, N. et al. In vivo continuous evolution of genes and pathways in yeast. Nat Commun 7, 13051 (2016).

[0253] 10. Ravikumar, A., Arzumanyan, G. A., Obadi, M. K. A., Javanpour, A. A. & Liu, C. C. Scalable, Continuous Evolution of Genes at Mutation Rates above Genomic Error Thresholds. Ce / / 175, 1946-1957. e13 (2018).

[0254] 11. Yi, X., Khey, J., Kazlauskas, R. J. & Travisano, M. Plasmid hypermutation using a targeted artificial DNA replisome. Sci. Adv. 7, eabg8712 (2021).

[0255] 12. Moore, C. L., Papa, L. J. & Shoulders, M. D. A Processive Protein Chimera Introduces Mutations across Defined DNA Regions In Vivo. J. Am. Chem. Soc. 140, 11560-11564 (2018).

[0256] 13. Berman, C. M. et al. An Adaptable Platform for Directed Evolution in Human Cells. J. Am. Chem. Soc. 140, 18093-18103 (2018).

[0257] 14. English, J. G. et al. VEGAS as a Platform for Facile Directed Evolution in Mammalian Cells. Cell 178, 748-761 ,e17 (2019).

[0258] 15. Taslimi, A. et al. Optimized second-generation CRY2-CIB dimerizers and photoactivatable Cre recombinase. Nat Chem Biol 12, 425-430 (2016).

[0259] 16. Zhao, E. M. et al. Optogenetic regulation of engineered cellular metabolism for microbial chemical production. Nature 555, 683-687 (2018).

[0260] 17. Reade, A. et al. TAEL: A zebrafish-optimized optogenetic gene expression system with fine spatial and temporal control. Development dev.139238 (2016) doi:10.1242 / dev.139238.

[0261] 18. Ding, Q. et al. Light-powered Escherichia coli cell division for chemical production. Nat Commun 11 , 2262 (2020).

[0262] 19. Ochoa-Fernandez, R. et al. Optogenetic control of gene expression in plants in the presence of ambient white light. Nat Methods 17, 717-725 (2020).

[0263] 20. Aditya, C., Bertaux, F., Batt, G. & Ruess, J. A light tunable differentiation system for the creation and control of consortia in yeast. Nat Commun 12, 5829 (2021).

[0264] 21. Benzinger, D. & Khammash, M. Pulsatile inputs achieve tunable attenuation of gene expression variability and graded multi-gene regulation. Nat Commun 9, 3521 (2018).

[0265] 22. Zoltowski, B. D., Motta-Mena, L. B. & Gardner, K. H. Blue light-induced dimerization of a bacterial LOV-HTH DNA-binding protein. Biochemistry 52, 6653-6661 (2013).

[0266] 23. Baumschlager, A., Weber, Y., Canovas, D., Dionisi, S. & Khammash, M. Enhancing the performance of Magnets photosensors through directed evolution. http: / / biorxiv.org / lookup / doi / 10.1101 / 2022.11.14.516313 (2022) doi:10.1101 / 2022.11 .14.516313.

[0267] 24. Zoltowski, B. D., Vaccaro, B. & Crane, B. R. Mechanism-based tuning of a LOV domain photoreceptor. Nat Chem Biol 5, 827-834 (2009).

[0268] 25. Motta-Mena, L. B. et al. An optogenetic gene expression system with rapid activation and deactivation kinetics. Nat Chem Biol 10, 196-202 (2014).

[0269] 26. Gligorovski, V., Sadeghi, A. & Rahi, S. J. Multidimensional characterization of inducible promoters and a highly light-sensitive LOV-transcription factor. Nat Commun 14, 3810 (2023).

[0270] 27. Rahi, S. J., Pecani, K., Ondracka, A., Oikonomou, C. & Cross, F. R. The CDK-APC / C oscillator predominantly entrains periodic cell-cycle transcription. Cell 165, 475-487 (2016).

[0271] 28. Rivera-Cancel, G., Motta-Mena, L. B. & Gardner, K. H. Identification of Natural and Artificial DNA Substrates for Light-Activated LOV-HTH Transcription Factor EL222. Biochemistry 51 , 10024- 10034 (2012). 29. Morgan, D. O. The cell cycle: Principles of control. (Published by New Science Press in association with Oxford University Press ; Distributed inside North America by Sinauer Associates, Publishers, 2007).

[0272] 30. Wasch, R. & Cross, F. R. APC-dependent proteolysis of the mitotic cyclin Clb2 is essential for mitotic exit. Nature 418, 556-562 (2002).

[0273] 31 . Epstein, C. B. & Cross, F. R. CLB5: a novel B cyclin from budding yeast with a role in S phase. Genes Dev. 6, 1695-1706 (1992).

[0274] 32. Jacobson, M. D., Gray, S., Yuste-Rojas, M. & Cross, F. R. Testing cyclin specificity in the exit from mitosis. Mol Cell Biol 20, 4483-4493 (2000).

[0275] 33. Loog, M. & Morgan, D. O. Cyclin specificity in the phosphorylation of cyclin-dependent kinase substrates. Nature 434, 104-108 (2005).

[0276] 34. Cross, F. R. & Levine, K. Molecular Evolution Allows Bypass of the Requirement for Activation Loop Phosphorylation of the Cdc28 Cyclin-Dependent Kinase. Molecular and Cellular Biology 18, 2923-2931 (1998).

[0277] 35. Jin, Y. H. et al. The Multiple Biological Roles of the 3'— >5' Exonuclease of Saccharomyces cerevisiae DNA Polymerase 6 Require Switching between the Polymerase and Exonuclease Domains. Molecular and Cellular Biology 25, 461-471 (2005).

[0278] 36. Charvin, G., Cross, F. R. & Siggia, E. D. Forced periodic expression of G 1 cyclins phase-locks the budding yeast cell cycle. Proc. Natl. Acad. Sci. U.S.A. 106, 6632-6637 (2009).

[0279] 37. Allard, C. A. H., Decker, F., Weiner, O. D., Toettcher, J. E. & Graziano, B. R. A size-invariant budduration timer enables robustness in yeast cell size control. PLoS ONE 13, e0209301 (2018).

[0280] 38. Yang, X., Jost, A. P.-T., Weiner, O. D. & Tang, C. A light-inducible organelle-targeting system for dynamically activating and inactivating signaling in budding yeast. MBoC 24, 2419-2430 (2013).

[0281] 39. Losi, A., Gardner, K. H. & Mbglich, A. Blue-Light Receptors for Optogenetics. Chem. Rev. 118, 10659-10709 (2018).

[0282] 40. Briggs, W. R., Christie, J. M. & Salomon, M. Phototropins: A New Family of Flavin-Binding Blue Light Receptors in Plants. Antioxidants & Redox Signaling 3, 775-788 (2001).

[0283] 41 . Avila-Perez, M., Hellingwerf, K. J. & Kort, R. Blue Light Activates the oB-Dependent Stress Response of Bacillus subtilis via YtvA. J Bacteriol 188, 6411-6414 (2006).

[0284] 42. Brunner, M. & Kaldi, K. Interlocked feedback loops of the circadian clock of Neurospora crassa: Circadian clock of Neurospora crassa. Molecular Microbiology 68, 255-262 (2008).

[0285] 43. Chen, C.-H., DeMay, B. S., Gladfelter, A. S., Dunlap, J. C. & Loros, J. J. Physical interaction between VIVID and white collar complex regulates photoadaptation in Neurospora. Proc. Natl. Acad. Sci. U.S.A. 107, 16715-16720 (2010).

[0286] 44. Dietier, J. et al. A Light-Oxygen-Voltage Receptor Integrates Light and Temperature. Journal of Molecular Biology 433, 167107 (2021).

[0287] 45. Endres, S. et al. Structure and function of a short LOV protein from the marine phototrophic bacterium Dinoroseobacter shibae. BMC Microbiol 15, 30 (2015).

[0288] 46. Hill, S., Austin, S., Eydmann, T., Jones, T. & Dixon, R. Azotobacter vinelandii NIFL is a flavoprotein that modulates transcriptional activation of nitrogen-fixation genes via a redoxsensitive switch. Proc. Natl. Acad. Sci. U.S.A. 93, 2143-2148 (1996).

[0289] 47. Pudasaini, A., El-Arab, K. K. & Zoltowski, B. D. LOV-based optogenetic devices: light-driven modules to impart photoregulated control of cellular signaling. Front. Mol. Biosci. 2, (2015).

[0290] 48. Glantz, S. T. et al. Functional and topological diversity of LOV domain photoreceptors. Proc. Natl. Acad. Sci. U.S.A. 113, (2016).

[0291] 49. Miller, B. G. & Wolfenden, R. Catalytic Proficiency: The Unusual Case of OMP Decarboxylase. Annu. Rev. Biochem. 71 , 847-885 (2002).

[0292] 50. Langan, R. A. et al. De novo design of bioactive protein switches. Nature 572, 205-210 (2019).

[0293] 51 . Lang, G. I. & Murray, A. W. Estimating the Per-Base-Pair Mutation Rate in the Yeast Saccharomyces cerevisiae. Genetics 178, 67-82 (2008).

[0294] 52. Romanini, D. W., Peralta-Yahya, P., Mondol, V. & Cornish, V. W. A Heritable Recombination System for Synthetic Darwinian Evolution in Yeast. ACS Synth. Biol. 1, 602-609 (2012).

[0295] 53. Biswas, S., Khimulya, G., Alley, E. C., Esvelt, K. M. & Church, G. M. Low-N protein engineering with data-efficient deep learning. Nat Methods 18, 389-396 (2021).

[0296] 54. Gietz, R. D. & Schiestl, R. H. High-efficiency yeast transformation using the LiAc / SS carrier DNA / PEG method. Nat Protoc 2, 31-34 (2007).

[0297] 55. Synthetic complete (SC) medium. Cold Spring Harb Protoc 2016, pdb.rec090589 (2016).

[0298] 56. 5-Fluoroorotic Acid (5-FOA) Plates. Cold Spring Harb Protoc 2016, pdb.rec086637 (2016).

[0299] 57. Boeke, J. D., Trueheart, J., Natsoulis, G. & Fink, G. R.

[0010] 5-Fluoroorotic acid as a selective agent in yeast molecular genetics, in Methods in Enzymology vol. 154 164-175 (Elsevier, 1987). 58. Dietier, N. et al. A convolutional neural network segments yeast microscopy images with high accuracy. Nat Commun 11 , 5723 (2020).

Claims

CLAIMS1 . A method for the evolution of a protein of interest by applying light-based selection pressure to a yeast cell,(i) wherein in the yeast cell the expression of one or more gene(s) that encode a protein controlling the progression of the yeast cell through the cell cycle is / are optogenetically controlled by an optogenetic gene expression system that is activated by light pulses of a specific wavelength, and wherein the protein of interest is the protein or one of the proteins constituting the optogenetic gene expression system; or(ii) wherein in the yeast cell the expression of one or more gene(s) that encode a protein controlling the progression of the yeast cell through the cell cycle is / are controlled by the protein of interest or a signal transduction cascade that comprises the protein of interest, and wherein the expression of the protein of interest is optogenetically controlled by an optogenetic gene expression system that is activated by light pulses of a specific wavelength wherein the method comprises(a) exposing the yeast cell to light pulses that differ from the light pulses that optimally activate gene expression but that still in part activate gene expression;(b) culturing the yeast cell contemporaneously under conditions where it divides and wherein random mutations are introduced for a time whereby a light-based selection pressure towards yeast cells being capable of optimally dividing under the light of (a) occurs, and(c) optionally isolating one or more yeast cells being capable of optimally dividing under the light pulses of (a) and preferably identifying the one or more mutation(s) in the gene encoding the protein of interest.

2. The method of claim 1 ,(i) wherein the yeast cell expresses an optogenetic gene expression system comprising a lightsensing module and a transcription factor module, wherein the light-sensing module is capable of sensing light pulses of a specific wavelength, and the transcription factor module comprises a DNA binding domain (DBD), a transcription activation domain and a nuclear localization sequence (NLS), and wherein the optogenetic gene expression system forms an active transcription complex when being exposed to the light pulses of a specific wavelength, wherein the protein of interest is the protein or one of the proteins constituting the optogenetic gene expression system, andwherein in the yeast cell one or more gene(s) that encode a protein controlling the progression of the yeast cell through the cell cycle is / are only expressed under the control of the optogenetic gene expression system, wherein the promoter region(s) of the one or more gene(s) that encode a protein controlling the progression of a cell through the cell cycle comprise(s) a binding site of the active transcription complex, preferably next to the minimal promoter(s) and upstream of the gene(s), and wherein the one or more gene(s) that encode(s) a protein controlling the progression of a cell through the cell cycle is / are expressed and the yeast cell starts to progress through the cell cycle and divide if the active transcription complex binds to the binding site of the active transcription complex; or(ii) wherein the yeast cell expresses an optogenetic gene expression system comprising a lightsensing module and a transcription factor module, wherein the light-sensing module is capable of sensing light pulses of a specific wavelength, and the transcription factor module comprises a DNA binding domain (DBD), a transcription activation domain and a nuclear localization sequence (NLS), and wherein the optogenetic gene expression system forms an active transcription complex when being exposed to the light pulses of a specific wavelength; wherein in the yeast cell the gene encoding the protein of interest is under the control of the optogenetic gene expression system, wherein the promoter region of the gene encoding the protein of interest comprises a binding site of the active transcription complex, preferably next to the minimal promoter and upstream of the gene, and wherein the gene that encodes the protein of interest is expressed if the active transcription complex binds to the binding site of the active transcription complex, and wherein in the yeast cell one or more gene(s) that encode(s) a protein controlling the progression of the yeast cell through the cell cycle is / are only expressed under the control of the protein of interest or a signal transduction cascade that comprises the protein of interest, wherein the one or more gene(s) that encode(s) a protein controlling the progression of a cell through the cell cycle is / are expressed and the yeast cell starts to progress through the cell cycle and divides if the protein of interest is expressed, wherein the method comprises(a) exposing the yeast cell to light pulses that differ from the light pulses that optimally activate the formation of the active transcription complex, but still in part activate the formation an active transcription complex;(b) culturing the yeast cell contemporaneously under conditions where it divides and wherein random mutations are introduced for a time whereby a light-based selection pressure towards yeast cells being capable of optimally dividing under the light of (a) occurs, and(c) optionally isolating one or more yeast cells being capable of optimally dividing under the light pulses of (a) and preferably identifying the one or more mutation(s) in the gene encoding the protein of interest.

3. A yeast cell for the evolution of a protein of interest by applying a light-based selection pressure to the yeast cell,(i) wherein in the yeast cell the expression of one or more gene(s) that encode a protein controlling the progression of the yeast cell through the cell cycle is / are optogenetically controlled by an optogenetic gene expression system that is activated by light pulses of a specific wavelength, and wherein the protein of interest is the protein or one of the proteins constituting the optogenetic gene expression system; or(ii) wherein in the yeast cell the expression of one or more gene(s) that encode a protein controlling the progression of the yeast cell through the cell cycle is / are controlled by the protein of interest or a signal transduction cascade that comprises the protein of interest, and wherein the expression of the protein of interest is optogenetically controlled by an optogenetic gene expression system that is activated by light pulses of a specific wavelength.

4. The yeast cell of claim 3,(i) wherein in the yeast cell one or more gene(s) that encode a protein controlling the progression of the yeast cell through the cell cycle is / are only expressed under the control of an optogenetic gene expression system, wherein the light-sensing module is capable of sensing light pulses of a specific wavelength comprising a light-sensing module and a transcription factor module, wherein the transcription factor module comprises a DNA binding domain (DBD), a transcription activation domain and a nuclear localization sequence (NLS), and wherein the optogenetic gene expression system forms an active transcription complex when being exposed to the light pulses of a specific wavelength, wherein the promoter region(s) of the one or more gene(s) that encode a protein controlling the progression of a cell through the cell cycle comprise(s) a binding site of the active transcription complex, preferably next to the minimal promoter(s) and upstream of the gene(s), and wherein the one or more gene(s) that encode(s) a protein controlling the progression of a cell through the cell cycle is / are expressed and the yeast cell starts to progress through the cell cycle and divide if the active transcription complex binds to the binding site of the active transcription complex; or(ii) wherein the yeast cell expresses an optogenetic gene expression system comprising a lightsensing module and a transcription factor module, wherein the light-sensing module is capable of sensing light pulses of a specific wavelength, and the transcription factor module comprises a DNA binding domain (DBD), a transcription activation domain and a nuclear localization sequence (NLS), and wherein the optogenetic gene expression system forms an active transcription complex when being exposed to the light pulses of a specific wavelength, wherein the protein of interest is or is a part of the optogenetic gene expression system, and wherein in the yeast cell one or more gene(s) that encode a protein controlling the progression of the yeast cell through the cell cycle is / are only expressed under the control of the optogenetic gene expression system, wherein the promoter region(s) of the one or more gene(s) that encode a protein controlling the progression of a cell through the cell cycle comprise(s) a binding site of the active transcription complex, preferably next to the minimal promoter(s) and upstream of the gene(s), and wherein the one or more gene(s) that encode(s) a protein controlling the progression of a cell through the cell cycle is / are expressed and the yeast cell starts to progress through the cell cycle and divides if the active transcription complex binds to the binding site of the active transcription complex; or(iii) wherein the yeast cell expresses an optogenetic gene expression system comprising a lightsensing module and a transcription factor module, wherein the light-sensing module is capable of sensing light pulses of a specific wavelength, and the transcription factor module comprises a DNA binding domain (DBD), a transcription activation domain and a nuclear localization sequence (NLS), and wherein the optogenetic gene expression system forms an active transcription complex when being exposed to the light pulses of a specific wavelength; wherein in the yeast cell the gene encoding the protein of interest is under the control of the optogenetic gene expression system, wherein the promoter region of the gene encoding the protein of interest comprises a binding site of the active transcription complex, preferably next to the minimal promoter and upstream of the gene, and wherein the gene that encodes the protein of interest is expressed if the active transcription complex binds to the binding site of the active transcription complex, wherein in the yeast cell one or more gene(s) that encode(s) a protein controlling the progression of the yeast cell through the cell cycle is / are only expressed under the control of the protein ofinterest or a signal transduction cascade that comprises the protein of interest, wherein the one or more gene(s) that encode(s) a protein controlling the progression of a cell through the cell cycle is / are expressed and the yeast cell starts to progress through the cell cycle and divides if the protein of interest is expressed.

5. A kit for the evolution of a protein of interest by applying a light-based selection pressure to a yeast cell, wherein the kit comprises at least one yeast cell of claim 3 or 4.

6. The method of claim 1 or 2, the yeast cell of clams 3 or 4 or kit of claim 5, wherein the yeast cell is Saccharomyces cerevisiae.

7. The method, yeast cell or kit of claim 6, wherein the one or more gene(s) that encode a protein controlling the progression of the yeast cell are one or more selected from CLN1, CLN2, CLN3, CLB3, CLB4, CLB5 and CLB6.

8. The method, yeast cell or kit of any one of claims 1 to 7, wherein the yeast cell in addition comprises a mutated polymerase instead of a wildtype polymerase having an increased mutation rate as compared to the wildtype polymerase, wherein preferably the mutated polymerase is the Saccharomyces cerevisiae POL3-L523D and the wildtype polymerase is Saccharomyces cerevisiae POL3.

9. The method, yeast cell or kit of any one of claims 1 to 8, wherein the optogenetic gene expression system comprises as light-sensing module the light-sensing module of EI222, as DBD the DBD of EI222, as transcription activation domain VP16, and / or as NLS the sequence PKKKRKV (SEQ ID NO: 6).

10. The method, yeast cell or kit of any one of claims 1 to 9, wherein in the yeast cell in addition a gene that encodes a cyclin-CDK inhibitor is deleted, preferably wherein the yeast cell is Saccharomyces cerevisiae and the deleted gene that encodes a cyclin-CDK inhibitor is SIC1.11 . The method of anyone claims 1 , 2 and 6 to 10, wherein the light pulses that differ from the light pulses that optimally activate expression I the formation of an active transcription complex differ from the light pulses that optimally activate the formation of an active transcription complex by their wavelength, light intension, pulse length and / or pulse frequency.

12. The method, yeast cell or kit of any one of claims 1 to 11 , wherein the binding site of the active transcription complex is pC120.

13. The method, yeast cell or kit of any one of claims 1 to 12, wherein the yeast cell is Saccharomyces cerevisiae, the genes CLN1, CLN2, CLN3, CLB3, CLB4, CLB5 and CLB6 are deleted in the yeast cell, and the binding site of the active transcription complex is pC120.

14. A mutated EL222 protein comprising or consisting of(a) the amino acid sequence of SEQ ID NO: 7 with one of the following amino acid substitutions Asp18Tyr, Val139Leu, Val41 Phe, Phe59Tyr, Cys75Arg, Gly80Arg, Gly80Glu, Gly82Val, Thr83Ala, Thr83Ser, Glu84Lys, Glu84Ala, Glu84Gly, Leu87lle, Leu87Pro, Thr88Pro, Asp89Tyr, Lys90Glu, Asp89Tyr, Lys99Glu, Val103lle, Val103Ph, Glu104Gly, He105Val, Asn107Asp, Tyr108His, Pro114Arg, Pro114Leu, Val121 Leu, Val121 Met, Leu132Pro, Phe115Ser, Gly136Ser, Ala157Val, Pro165Ser and Val201 Leu, or(b) an amino acid sequence which is at least 80% identical to the amino acid sequence of (a) provided that the one amino acid substitution is retained.

Images