Development of Int-Plex@ binary memory switch system: plant genome modulation driven by large serine-integrases

17 The comprehension of virus -host interactions has allowed numerous advances in 18 developing biotechnological methodologies for plant genome editions, constituting a 19 promising path for plant genetic engineering. Among these advancements, phage -20 encoded large serine -integrases have emerged as noteworthy tools to modulate plant 21 metabolic pathways by inserting, excising, or inverting DNA stretches in a reversible 22 and specific way. The present work shows the f oundation of the Int-Plex@ (INTegrase 23 PLant EXpression) binary memory switch system, which consists of the application of 24 four distinct orthogonal prophage large serine-integrases (Int) (BxB1, phiC31, Int13, 25 and Int9) as an input trigger mechanism for the inversion or excision of genomic DNA. 26 The memory genetic switch is divided into the excision module and the inversion 27 module. The excision module is activated by Bx B1 or phiC31 enzymes (input). In this 28 case, the DNA sequence flanked by its attachment sites is excised from the genome 29 (output). The inversion module is activated by Int9 or Int13 (input). The inverted 30 mgf gene sequence is flipped to its functional coding sequence, and the switch output is 31 mGFP. Moreover, prokaryotic-based cell-free in vitro transcription-translation reactions 32 (TxTl) w ere used as a fast platform for testing Ints attB/P in tandem site activity . 33 Furthermore different plasmid delivery strategies for plant cell Int heterologous 34 expression were tested: leaf tissue agroinfiltration of Agrobacterium 35 tumefaciens transformed with binary plasmids and a biolistic system. After each 36 treatment, the edited genomic DNA sequences were amplified and verified by Sanger 37 and Nanopore sequencing. Despite the challenges of using Ints, the potential benefits 38 are significant and deserve deeper exploration and development. The Int-Plex@ binary 39 genome memory switch system can be applied to produce genetic circuits combined 40 with omic s tools and sgRNAs to engineer and modulate plant metabolic pathways 41 temporally and reversibly. 42 43

orthogonal phage serine integrase , genome editing , nanopore sequencing, 44 plant biotechnology, synthetic biology. 45 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted January 11, 2024. ; https://doi.org/10.1101/2024.01.11.575089doi: bioRxiv preprint 1. INTRODUCTION 46 47 Over the last several decades, major advances in plant synthetic biology have emerged 48 from significant discoveries based on the understanding of the co -evolutionary arms 49 race between viruses and their hosts. These ongoing battles have led to the evolution of 50 intricate and complex molecular interactions that were exploited to create novel 51 biotechnological tools based on the bottom -up approach for plant genome modulation. 52 Examples include (I) bacteria restriction endonucleases and phage DNA ligases, which 53 allowed the revolution of recombinant DNA technology and are used in 54 BioBrickTM assembly and modern one -step DNA assembly techniques 1-7; (II) viral 55 vectors for gene delivery 8,9; (III) virus -induced gene silencing (VIGS) 10,11; (IV) the 56 RNA interference (RNAi) technology, which is widely used in gene down -regulation 57 studies12-14; and (V) CRISPR -Cas systems, which has recently been widely applied in 58 crop genome editing technology 12,15-18. However, these last two technologies are 59 irreversible and permanent; once the plant gene is silenced or edited, it cannot return to 60 its original state. Therefore, to overcome these limitations, a promising approach, 61 originating from bacteriophage -prokaryotic interactions, is the application of 62 filamentous phage -encoded large serine integrases (Ints), which are multifunctional 63 recombinases with the potential to edit DNA sequences and also serve as efficient in 64 vitro DNA assemblers 19-22. Therefore, considering the development of a dynamic and 65 reversible genome modulation system in plant cells, it is possible to include the 66 orthogonal Ints in modular synthetic genetic devices based on Boolean logic gates 23-29. 67 Ints catalyze directly and unidirectionally a site -specific insertion, deletion or 180° 68 inversion of DNA sequence flanked by the two attachment sites ( att), attB (bacteria) 69 and attP (phage) depending on the design of their orientation relative to each 70 other, which results in the formation of two new post -recombination attachment 71 sites, attL (left) and attR (right)19,20,25-28,30. Bx B1 and phiC31 Ints have been 72 successfully used in plant engineering and gene stacking29,31-34. 73 Furthermore, using a bioinformatics approach, Yang et al. prospected 11 new Ints with 74 their respective functional att sites20. Subsequently, Gomide et al . showed that those 75 Ints function in eukaryotic cells and ranked Int9 and Int13 among the m as the most 76 effective Ints for plasmid flipping in the plant cell context 25. Thus, the IntPlex@ binary 77 memory switch takes Bx B1, phiC31, Int9, and Int13 as input signals to control an 78 output (genome edition) in a user -defined manner (site -specific excision or inversion). 79 Lastly, the possibility of using distinct plasmid delivery as a general method for 80 integrase expression was examined. 81 82 2. RESULTS 83 84 2.1 Int-Plex@ binary memory switch system design 85 We have assembled a multifunctional genetic switch system based on att site pairs from 86 4 Ints, namely BxB1, phiC31, Int9, and Int13. The cassette constructs consist of the 87 reverse complement sequence of green fluorescent protein reporter gene gfp (mgfp or 88 degfp) flanked by in tandem attB or attP sites from the Ints selected (Figure 1A) . 89 Promoter and terminator sequences to complete the transcription unit varied according 90 to the in vitro or in vivo experimental model. Effector Ints were delivered individually 91 in a separate plasmid under transcriptional control of a constitutive promoter 25,35. This 92 configuration allows the formation of a binary system with an Excision Module 93 controlled by BxB1 or phiC31 recombinases designed as a NOR logic gate, in which the 94 presence of either Int (input) prevents reporter expression due to its deletion (output) 95 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted January 11, 2024. ; https://doi.org/10.1101/2024.01.11.575089doi: bioRxiv preprint and an Inversion Module controlled by Int9 or Int13 with the structure of an XOR gate 96 (Figure 1B). To achieve varying outcomes in DNA rearrangement depending on the Int 97 acting on the system, we inserted the att sites in the construction either in the same or 98 opposing orientation relative to their respective site counterpart. While for BxB1 and 99 phiC31, attB and attP sites are in sense orientation, with their recombination resulting in 100 excision of the DNA sequence flanked by them from the construction (Figure 1, C and 101 D), Int9 and Int13 att sites are in reverse complement orientation, which means that 102 upon recombination, the target DNA sequence will be 180° flipped (output), modulating 103 the gfp expression (Figure 1 , E and F). Although both Int9 or Int13 are capable of 104 unidirectionally flipping the reporter gene coding sequence and activating its 105 transcription, an essential feature of this type of logic gate is the gene set/reset, that 106 upon concurrent or sequential introduction of both enzymes, a second inversion event 107 will result in the return of the target DNA orientation to the original OFF state, silencing 108 reporter expression (Figure 1, G and H) . In this double -editing event, one Int will 109 recognize its attB/P sites and catalyze the inversion of the reporter from its OFF state to 110 ON state, forming the intermediate plasmid with attL/R sites which results in the GFP 111 production. Instantaneously, this intermediate plasmid is the substrate for the action of 112 the second Int, which, upon recombination of its attB/P sites, will result in the formation 113 of the final plasmid with both attL/R sites for Int9 and Int13 with consequent recovery 114 from the ON state to initial OFF state, silencing gfp expression (Figure 1, G and H). 115 116 2.2 In vitro function of Int-Plex@ binary memory switch system 117 Our first goal was to understand the behavior of this new binary memory system and 118 ensure that the presence of att sites in tandem would not interfere with expected Int 119 activity. For the functional evaluation, we chose cell -free in vitro transcription-120 translation reactions (TxTl), given its valuable application as a fast platform for testing 121 Int activity, DNA editing tools and genetic circuits35,36. The switch was assembled using 122 TxTl optimized parts, including the degfp gene as a reporter , Ribosome Binding Site 123 (RBS) and the T7Max transcription system 35 (Figure 2A) . Reactions containing the 124 reporter construction alone or with plasmids for Int expression were incubated overnight 125 and analyzed for DNA rearrangement and deGFP fluorescence emission. This enabled 126 us to test the efficacy of each reaction independently and transiently in vitro for the first 127 time before integrating the memory switch into the plant genome . Besides deGFP 128 expression, DNA excisions by BxB1 or phiC31 were detected, as predicted, by the PCR 129 amplifications (Figure 2, B and C ). The excised circular DNA molecule was confirmed 130 by Sanger sequencing, where post -recombination attR site for BxB1 is present (Figure 131 3, D and E). Sanger sequencing of phiC31 treatments is in progress. All sequences are 132 available for consultation at the link provided in the Data availability section . In 133 addition, the introduction of Int9 or Int13 could invert the DNA and allowed deGFP 134 production in vitro (Figure 3, A and B). Although the presence of both Int9 and Int13 in 135 the same reaction must result in the re -establishment of the initial OFF state with degfp 136 silencing, the closed nature of the reaction environment and relatively short incubation 137 time preserved the protein produced while the single -edited intermediate was present, 138 maintaining a high level of fluorescence. The double -editing event was confirmed by 139 Sanger sequencing of the reporter construct, where both post -140 recombination attL/attR site pairs for Int9 and Int13 are present in the same molecule 141 (Figure 3, C-E). 142 143 2.3 Int-Plex@ binary memory switch system design for plant genome modulation 144 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted January 11, 2024. ; https://doi.org/10.1101/2024.01.11.575089doi: bioRxiv preprint To demonstrate the applicability of this tool in DNA modulation, we also verified the 145 functionality of the switch stably integrated into the plant genome. Here, we assembled 146 the reverse complement sequence of mgfp flanked by four in tandem Bx B1, phiC31, 147 Int9, and Int13 attB and attP sites in a synthetic genetic switch, denoted IntPlex@ 148 binary memory switch and inserted into the Nicotiana benthamiana genome. Each 149 serine integrase was plant codon optimized and inserted in a different set of plasmids 150 for two systems of gene delivery: agroinfiltration by Agrobacterium tumefaciens and 151 biolistic. The IntPlex@ binary memory switch takes these four serine integrase input 152 signals to control an output (genome edition) in a user -defined manner (excision or 153 inversion). 154 155 2.3.1 Plant genome excision triggered by phiC31 or BxB1 156 BxB1 or phiC31 catalyzed the excision of mgfp from the genome. Upon delivery of the 157 effector plasmids containing the integrases by agroinfiltration, molecular analyses of 158 plant DNA within five d.p.i (days post infiltration) showed successful deletion, with 159 genome sequencing confirmation of proper formation of respective attL sites for each 160 Int (Figure 4A and Figure 5) and loss of DNA sequence flanked by its attachment sites 161 (output) (Figure 4, B and C) . Sanger sequencing of phiC31 treatments is in progress. 162 Moreover, Nanopore sequencing was utilized to identify the excised circular DNA 163 molecule containing mgfp, BxB1 attR site and remaining attB/P sites from the other 164 integrases (Figure 6). The sequencing run lasted four hours, yielding roughly 800,000 165 reads. As depicted in Figure 7, the basecalled read lengths exhibited a bimodal 166 distribution, with the median read length positioned within the distribution's upper 167 mode. Quality scores ranged along the x -axis, predominantly showing mid to high 168 values indicative of reliable base calling accuracy. This pattern reflects a diverse set of 169 read lengths, with shorter reads tending to have higher quality scores. For subsequent 170 analysis, only reads with quality scores of 10 or above were maintained (383,245 reads). 171 However, the detection of phiC31 output and stability and processing time for the 172 circular DNA excised degradation remains to be evaluated. Regarding the delivery of 173 the Int effector by biolistic, neither genome editing, nor the output molecule were 174 detected (data not shown) . Nanopore sequencing will be applied to detect the output 175 molecules and analyze the low detection rate. All sequences are available for 176 consultation at the link provided in the Data availability section. 177 178 2.3.2 Plant genome inversion triggered by Int 9 or Int 13 179 This switch operates as an ON switch, where MGFP is expressed in the ON state and 180 not in the OFF state. Upon delivery of the effector plasmids containing the integrases by 181 agroinfiltration, molecular analyses of plant DNA within five d.p.i showed successful 182 DNA inversion. Int9 recognized its attB/P sites and catalyzed the inversion 183 of mgfp from its OFF state to ON state, resulting in the formation of the intermediate 184 genome with attL/Int9 and attR/Int9 sites and the output MGFP after int delivery via 185 agroinfiltration (Figure 8A). Likewise, Int13 recognized its attB/P sites and catalyzed 186 the inversion of mgfp from its OFF state to ON state, resulting in the formation of the 187 intermediate genome with attL/Int13 and attR/Int13 sites and the output MGFP (Figure 188 8B). As for effector Int delivery by biolistic, phenotypical analysis showed no increase 189 in mGFP fluorescence (data not shown). However, fluorescence could be detected in the 190 positive control of plants bombarded with microparticles coated with a plasmid 191 constitutively expressing eGFP 37. Despite the lack of signal, DNA inversion could be 192 detected for both Ints by PCR amplification, with confirmation of inversion and 193 recombined attL/R sites by DNA sequencing. 194 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted January 11, 2024. ; https://doi.org/10.1101/2024.01.11.575089doi: bioRxiv preprint 2.3.3 Plant genome modulation: Turning DNA ON and OFF 195 While a one -way memory switch is functional, reversibly turning a gene on and off 196 would allow interesting bidirectional regulation. Thus, Int9 and Int13 were 197 agroinfiltrated into the plant five days apart to turn mgfp on and off in the same cell. 198 Initially, Int 9 recognized its attB/P sites and catalyzed the inversion of mgfp from its 199 OFF state to ON state, resulting in the formation of the intermediate genome with 200 attL/Int9 and attR/Int9 sites and the output MGFP. Five days after, this intermediate 201 genome was the substrate for the action of In t13 that recognized its attB/P sites and 202 catalyzed the inversion of mgfp from its ON state to OFF state, resulting in the 203 formation of the final genome with both attL/Int9 and attR/Int9 and attL/Int13 and 204 attR/Int13 sites. Thus, the system presents the MGFP until the ON state mRNA 205 molecule can no longer be translated according to the MGFP half-life time. The Sanger 206 sequencing of Int9 and Int13 treatments are in progress. All sequences are available for 207 consultation at the link provided in the Data availability section. 208 209 210 3. DISCUSSION 211 212 In this study, our primary goal was to broaden the plant genome modulation toolbox by 213 creating a binary memory switch controlled by serine integrases. Currently, the 214 available array of tools for constructing synthetic memory systems in plants is quite 215 limited, mainly relying on phiC31 and BxB129,31,32,38,39. Efforts for the systematic 216 identification of new Ints and their att sites, including metagenomic studies, 217 development of new bioinformatic approaches, and functional characterization, 218 contribute to expanding the availability of such tools and enhance the possibilities for 219 assembly of more complex multicomponent genetic circuits 40. In that context, we 220 applied phiC31, BxB1, and two more recent Ints (Int9 and Int13) previously 221 characterized in a plant model to create a bimodular switch for DNA excision and 222 bidirectional inversion 25. All four Int successfully rearranged the target DNA inserted 223 into the genome of N. benthamiana as programmed, confirming the functionality of 224 both modules in our system. 225 More than assembling a new switch system for gene expression regulation, this proof -226 of-concept demonstration paves the way for applying new Int candidates other than 227 phiC31 and BxB1 in plants. Beyond orthogonality, which allows for various effectors in 228 the same circuit, different integrases can show a range of efficiency degrees beneficial 229 for implementing flow control nodes in a synthetic regulatory pathway. As 230 demonstrated by Gomide et al. (2020), for instance, Int13 and Int5 both could invert a 231 target DNA sequence in Arabidopsis thaliana protoplasts, although resulting in different 232 levels of reporter expression25. 233 Another essential advantage point of IntPlex@ is the use of independent Ints to control 234 inversion directionality. After the first inversion event by Int9 or Int13, restoration to 235 the initial OFF state depends exclusively on introducing the second enzyme, with no 236 chance of interference. In contrast, the RESET system published by Bernabé -Orts et al. 237 (2020) applies phiC31 with its correspondent RDF protein to revert unidirectionality 29. 238 Although aiming at somewhat different outcomes, it is noteworthy that with RDF, they 239 observed an increase in system instability. Recombination of attL/R sites requires direct 240 interaction of Int-RDF, but the presence of free phiC31 led to new recombination of the 241 attB/P sites just restored, resulting in mixed states of the switch at the same time29. 242 The Sanger Sequencing method is widely used for identifying plant genome editions. 243 Nevertheless, it comes with some limitations, including the prerequisite for pre -existing 244 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted January 11, 2024. ; https://doi.org/10.1101/2024.01.11.575089doi: bioRxiv preprint knowledge of the DNA sequence and a constrained ability to simultaneously detect 245 various DNA fragments. On the other hand, t he utilization of Nanopore DNA 246 sequencing has enabled determination of total DNA products of the plant cells genome 247 edition reactions in a short time and with high -throughput data analysis in real time. 248 Nanopore sequencing holds great promise for detecting edited DNA and when 249 combined with stoichiometry analysis, this technology provides a powerful toolset to 250 access genomic diversity for studying edited DNA at a quantitative level. The 251 integration of nanopore sequencing and stoichiometry analysis not only enhances the 252 ability to detect edits accurately but also facilitates a deeper understanding of the 253 dynamics and efficiency of serine integrase editing reactions. 254 It is worth mentioning that when a system is imported from the prokaryotic to the 255 eukaryotic cell, as is the case for bacteriophage -derived recombinases, it is necessary to 256 consider the presence of the nucleus as a physical barrier to the performance of enzymes 257 that act on DNA. Nevertheless, all four integrases tested in this work were able to 258 overcome the nuclear barrier and interact with chromatin without the need to add NLS 259 or other protein engineering. It is possible that the passage through the nuclear pore 260 occurred due to the presence of intrinsic NLS in its native structure after exposure of 261 positively charged amino acids during protein folding 25. Notably, the system functioned 262 consistently over two plant generations. Employing large serine -integrases in modular 263 switch devices exemplifies a sophisticated and controlled approach, affording precision 264 in genetic alterations with the potential for intricate modulation of plant physiological 265 processes. Consequently, the integration of large serine -integrases into the 266 biotechnological toolkit heralds a promising trajectory for refining and expanding 267 strategies in plant genetic engineering. However, further research is needed to fully 268 understand the potential of this system in plant biotechnology, the potential for 269 unintended off-target effects, and to optimize its use. 270 271 4. CONCLUSIONS 272 273 PhiC31, BxB1, Int9, and Int13 promoted efficient directional DNA site -specific 274 recombination of complex chromatin of the plant cell, and thus have major potential 275 applications in genome engineering and metabolic pathway engineering. This memory 276 genetic switch system is dynamic and may be combined with other technologies like 277 inducible promoters and advanced sequencing tools. For example, implementing the 278 sequence barcoding with Nanopore sequencing amplifies the capabilities of genomic 279 analysis and allows for comprehensive profiling of the entire edited genome, capturing 280 structural variations or ensuring the fidelity of the editing process. Together, this 281 combined approach not only validates the accuracy of serine integrase -mediated edits at 282 a detailed level but also offers a holistic view of the entire genome modulation, 283 fostering a deeper understanding of the genomic landscape post -editing. Lastly, it is 284 more than evident that we are living in an era marked by the development of tools that 285 allow crop genomes to be edited to provide resistance to environmental challenges, 286 improve nutritional content and optimize defenses against pathogens. However, the lack 287 of efficacy of plasmid delivery systems has hampered advancements in the precision of 288 gene editing. This is a crucial moment that requires the convergence of efforts to 289 develop studies that aim to refine plasmid or gene delivery systems, ensuring that 290 genetic modifications reach the maximum number of cells efficiently. 291 292 5. MATERIAL AND METHODS 293 294 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted January 11, 2024. ; https://doi.org/10.1101/2024.01.11.575089doi: bioRxiv preprint 5.1 mGFP genetic switch plasmid design 295 In pLSB_pCAMBIA2300_IntX or pUC_IntX plasmids, both actin2 gene promoter and 296 NOS terminator drive (a) plant codon -optimized serine integrases transcription for 297 heterologous expression or (b) serine integrases fused with NLS signal. In 298 pLSB_PVX_GW_IntX plasmids, both subgenomic CP promoter and NOS terminator 299 drive serine integrase transcription for heterologous expression with or without the NLS 300 signal. 301 302 5.2 Serine-Integrase Plasmid Design 303 The Int9, Int13, phiC31, and BxB1 genes were retrieved from pUC57 25 and cloned into 304 PVX-GW binary plasmid37 by Epoch Life Science Inc. These plasmids resulted in a set 305 of serine integrase expression plasmids, individually called pLSB_PVX_GW_X (X = 306 Int9, Int13, phiC31, or BxB1). The Int9, Int13, phiC31, and BxB1 genes controlled by 307 the actin2 gene promoter and NOS terminator were retrieved from pUC57 and cloned in 308 pCAMBIA2300 plasmid by Epoch Life Science Inc 25. These plasmids resulted in a set 309 of serine integrase expression plasmids, individually called pLSB_pCAMBIA2300_X 310 (X = Int9, Int13, phiC31, or BxB1). 311 312 5.3 Plant transformation 313 Transgenic plants of Nicotiana benthamiana were produced at University of California 314 Riverside´s Department of Botany and Plant Science performed the plant 315 transformation. pART27 plasmid containing the genes BXB1_PHIC31_9_13_mGFP(rc) 316 and nptII were delivered into GV3101 Agrobacterium tumefaciens via electroporation. 317 Antibiotic-resistant colonies were selected and confirmed by colony PCR for glycerol 318 stocks. The transformed A. tumefaciens strain was used to inoculate the leaf tissue of 319 six-week-old plants. Callus induction and shoot induction media containing MS salts, 320 MS vitamins, 0.56 0.56mM Myo, 8.84 8.84 μM BAP, 0.54 0.54 μM NAA, 3% sucrose, 321 0.21 0.21 mM kanamycin, 0.55 0.55 mM Cefatoxime and 1.321.32 mM Carbenecillin 322 were used to induce calli and shoots. Thirty and ten shoots transformed with 323 BXB1_PHIC31_9_13_mGFP(rc) and pART27-nptII (empty plasmid) respectively were 324 transferred to rooting media containing ½ MS salts, MS vitamins, 0.56 0.56 mM Myo- 325 inositol, 0.98 0.98 μM IBA, 1.5% sucrose, 0.34 0.34 mM vancomycin, 0.55 0.55 mM 326 cefotaxime, and 1.32 mM Carbenicillin. A total of 40 N. benthamiana plants were 327 analyzed by end-PCR using specific primers for each construct. Twenty-one plants 328 amplified an expected band of 717 bp with the specific primers FwmGFP/RvmGFP, and 329 nine plants transformed with pART27 (empty plasmid) amplified a band of 699 bp for 330 the nptII gene. Nicotiana benthamiana transgenic seeds germinated in vermiculite, and 331 plantlets were grown in a greenhouse. Eight-week-old plants were used for the 332 Agrobacterium infiltration or biolistic procedure. 333 334 5.4 In vitro Transcription/Translation Reactions 335 In vitro , Transcription/Translation Reactions, or TxTl reactions, were performed with 336 myTxTl T7 Expression Kit (Daicel Arbor Biosciences) following manufacturer 337 instructions. Briefly, 24 μL reactions were assembled in 1.5 mL centrifugation tubes 338 with 18 μL Sigma 70 Master Mix, 0.1 nM P70a-T7rnap HP, and 10 nM of each plasmid 339 (reporter + effector Int). Reactions were incubated at 29°C for 16 h. After incubation, 340 GFP production was visually assessed with an LED blue light Transilluminator 341 (KASVI), and 2 μL were used as a template for PCR amplification. 342 343 5.5 Delivery Systems 344 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted January 11, 2024. ; https://doi.org/10.1101/2024.01.11.575089doi: bioRxiv preprint 5.5.1 Agroinfiltration 345 The binary plasmids were electroporated into Agrobacterium tumefaciens strain 346 GV3101 and selected on LB agar supplemented supplemented with 0.1 mM kanamycin 347 and 0.1 mM gentamicin. . After two days of growth, several colonies were carefully 348 harvested from the plate and suspended in LB medium (2 mL). The bacteria were grown 349 overnight on an orbital shaker at 28 °C at 200 rpm. The cells were collected by 350 centrifugation for 15 min at 4000 g, and the precipitate was resuspended in the 1x 351 MMA infiltration buffer (10 mM MES, pH = 5.6, 10 mM MgCl 2, 180 µM 352 acetosyringone) to OD600 = 0.3. Agrobacteria harboring plasmids with each integrase 353 and 35S: p19 gene were mixed in equal volumes. Leaves of 8 -week-old N. benthamiana 354 plants were injected with agrobacteria suspension using a syringe without a needle. 355 Agroinfiltration assays were assessed with and without co -infiltration with 35S: p19 356 with no changes in results. The infiltrations were performed on the same plant, and 357 duplicate infiltrations were performed on at least three plants within a single 358 experimental assay. 359 360 5.5.2 Biolistic 361 Genetic modified N. benthamiana plant leaves carrying inverted mgfp gene flanked by 362 integrases attachment sites were directly shot with gold (1.5 -3.0 µm) and tungsten 363 (approx. 1µm) microparticles as previously described 41. The microparticles were coated 364 with pUC27 plasmids carrying Int9, Int13, phiC31, and BxB1 integrase genes 365 individually25. As a positive control of gene delivery, a plasmid carrying a constitutive 366 promoter controlling GFP gene was shot under the same conditions at wild -type N. 367 benthamiana leaves. Microparticles mixed with 8 μL DNA (1μg.μL-1) were accelerated 368 at 650 psi into plant leaves (2,5 -3,5 cm length) at a 2 cm distance. Eight shots of each 369 DNA construction were bombarded in 8 -week-old plants. 72h after foreign DNA was 370 delivered, shooted leaves were analyzed on Confocal laser scanning microscope (Leica 371 TCS SP8, GER) to visualize mgfp expression. PCR screening has been performed to 372 detect mgfp and the attL and attR post-recombination sites. 373 374 5.6 Plant total DNA extraction and PCR 375 Total plant DNA isolation was based on the modified CTAB Method Protocol described 376 by Lacorte et al. (2010) with adaptations 39. Firstly, plants were screened by PCR using 377 primer pairs DEP_mGFP_FW and DEP_mGFP_RV - designed to detect mgfp – and 378 nptII_Fw and npt_II_Rv - for the amplification of nptII, a kanamycin resistance marker 379 – to identify positive transformants. Following the integrase delivery method and 380 incubation period, we collected slices of approximately 20mg from leaves of treated and 381 control plants and proceeded with DNA extraction, as mentioned above. DNA samples 382 were then used as templates in PCR reactions for detection of either attL and attR sites 383 resulting from Int9 or Int13 recombination or the deletion of mgfp gene flanked by 384 phiC31 and BxB1 att sites upon introduction of these integrases in the system. For attL 385 and attR sites amplification, we have used the primer combinations 35S_282F + 386 attR_IntX_R and attL_IntX_F + IDP_OCSt_349_Rv, respectively, with “IntX” on the 387 primer name denoting the use of specific primers for each Int. As for deletion 388 confirmation, primer pair 35S_282F + IDP_OCSt_349_Rv was used to amplify the 389 whole cassette between the 35S promoter and OCS terminator, with a band shift of 390 about 1kb indicating mgfp excision. The circular molecule containing the mgfp gene 391 excised from the system by phiC31 or BxB1 was amplified using the primer pair 392 NbDel_mGFP_493_Fw + NbDel_mGFP_300_Rv. Two PCR reactions were necessary 393 for the biolistic method to amplify attL and attR sites. The primer pair 35S_282F + 394 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted January 11, 2024. ; https://doi.org/10.1101/2024.01.11.575089doi: bioRxiv preprint IDP_OCST_349_RV performed the first round, which amplified the cassette. The first 395 reaction was then used as a template for the specific reaction with primers attached on 396 the attL and attR sites, paired with terminator and promoter primers, respectively. That 397 way, the second round of amplification for detection of the Int9 attL site used primers 398 35S_282F + OnOff_attR9_core_Rv, for Int9 attR site primers OnOff_attL9_core_Fw + 399 IDP_OCST_349_RV, for Int13 attL site primers 35S_282F + OnOff_attR13_core_Rv 400 and to amplify Int13 attR site primers OnOff_attL13_core_Fw + IDP_OCST_349_RV. 401 As for the primers used for molecular detection in TxTl reactions, inversion of the 402 reporter was amplified with primer pair 2041GbA_ck_5p_FW + deGFP_Hd_Fw while 403 excision was detected with primers 2041GbA_ck_5p_FW and 2041_ck_3p_RV1. 404 Primers used in this study are listed in Table 1. 405 406 Table 1. 407 PRIMER NAME PRIMER SEQUENCE DEP_MGFP_FW atgagtaaaggagaagaact DEP_MGFP_RV ttatttgtatagttcatccatg NPT_II_FW atggcaattaccttatccgcaacttc NPT_II_RV cagaagaactcgtcaagaaggcg 35S_282F attgatgtgatatctccactgacgtaagggatgacgcac ATTR_INT9_R tggaagtgtgtatcaggtaactggatacctcatc ATTR_INT13_R gtagaacttgaccagttggtcctgtaaatataagcaatcc ATTL_INT9_F ataattggcgaacgaggtatctgcatagttattccgaac ATTL_INT13_F tccagatccagttgttttagtaacataaataca IDP_OCST_349_RV taggtttgaccggttctgcc ONOFF_ATTL9_CORE_FW ttggcgaacgaggtatctgc ONOFF_ATTR9_CORE_RV gaagtgtgtatcaggtaactgg ONOFF_ATTL13_CORE_FW tgtagaacttgaccagttggt ONOFF_ATTR13_CORE_RV gtttttccagatccagttgtt NBDEL_MGFP_493_FW tcaagacccgccacaacatcg NBDEL_MGFP_300_RV gaagaagatggtcctctcctgc 2041GBA_CK_5P_FW atgaccaaaatcccttaacgtgag DEGFP HD FW cccgggaagcttatggagcttttcactggcgttg 2041_CK_3P_RV1 tggcttaactatgcggcatcag 408 Amplification reactions were performed in a Veriti ™ 96-Well Fast Thermal 409 Cycler (Applied Biosystems, USA). Reaction set -up included 2.5ul of 10X PCR Buffer 410 (Thermo Scientific, USA) , 1.5 mM MgCl2, 0.4 mM dNTP mix, 1 U Platinum ™ Taq 411 DNA Polymerase (5 U/ μL) (Thermo Scientific, USA), 0.2 μM each primer and 1 μL of 412 purified DNA in a 25 μL final volume. PCR cycling consisted of one initial incubation 413 at 94°C for 2 minutes, followed by 40 cycles of denaturation at 94°C, 30s; Annealing 414 TEMP for 30s; Extension at 72°C for 1 min/kb, with final incubation at 4°C. Given that 415 primer pair 35S_282F + IDP_OCSt_349_Rv will amplify both intact and disrupted 416 cassette resulting from mgfp gene deletion by phiC31 or BxB1, an extension time of just 417 30 seconds were used in this case to favor amplification of the 1kb smaller amplicon. 418 All PCR products were detected in 1% agarose gel electrophoresis stained with 419 SYBR™ Safe DNA Gel Stain (Invitrogen, USA). Gel slices containing the specific 420 bands were excised under LED Blue Light transluminator (KASVI, Taiwan) to prevent 421 nicking of the DNA and its purification was carried out using the ReliaPrep ™ DNA 422 Clean-Up and Concentration System (Promega, USA) according to manufacturer 423 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted January 11, 2024. ; https://doi.org/10.1101/2024.01.11.575089doi: bioRxiv preprint specifications. Isolated amplicons were cloned in pGEM -T-Easy (Promega, USA) and 424 sequenced (Macrogen, USA) after DNA propagation in Escherichia coli DH10β strain. 425 426 5.7 Confocal laser scanning Microscopy 427 Leaf samples were analyzed using the Confocal laser scanning microscope (Leica TCS 428 SP8, GER) after 5 d.p.i . Argon laser line excitation wavelength and emission bandpass 429 filter wave lengths for MGFP were 48 4 nm and 500 -550 nm. Chlorophyll 430 autofluorescence was detected, in parallel with MGFP acquisition, using a 650 –710 nm 431 bandpass filter. Image acquisition parameters (e.g. laser power, pinhole, detector gain, 432 etc.) and sampling time post -infiltration were held constant within an experiment (i.e. 433 within each figure). Raw data were processed in the LAS X (Leica Application Suite , 434 GER) software (www.leica-microsystems.com, GER). 435 436 5.8 Nanopore sequencing and Data analysis 437 Library preparation was carried out using the ligation sequencing kit (Oxford Nanopore 438 Technologies, UK) SQK -NBD112-24, according to the manufacturer's instructions. 439 Briefly, all amplicons were purified, and approximately 200 fmol per sample were 440 submitted to end -prep and native barcode ligation, using NEBNext Ultra II End 441 repair/dA-tailing Module (New England Biolabs, USA) and NEB Blunt/TA Ligase 442 Master Mix (New England Biolabs, USA) respectively. Next, native adapter ligation 443 and cleaning steps were performed using NEBNext Quick Ligation Module (New 444 England Biolabs, USA) kit and Agencourt AMPure XP beads (Beckman Coulter, USA), 445 respectively. The library was sequenced on a R10.4 flow cell (FLO -MIN112) (Oxford 446 Nanopore Technologies, UK) on a MinION Mk1b sequencer with MinKNOW 447 (23.04.6). Basecalling, demultiplexing and adapter/barcode tr imming were performed 448 using Guppy [v6.5.7]42, with a Super accurate (SUP) basecalling model. The reads with 449 barcodes on both ends and quality scores ≥ 10 were mapped against all the in silico 450 predicted genome -edited sequences, including the predicted circular molecules 451 containing the excised mgfp gene, using minimap243. 452 453 Competing interests 454 The authors declare no conflict of interest. 455 456

457 We acknowledge funding support from Embrapa Genetic Resources and 458 Biotechnology/National Institute of Science and Technology in Synthetic Biology, 459 National Council for Scientific and Technological Development/ Ministry of 460 Agriculture Livestock and Supply (465603/2014 -9; 400145/2023-5), Research Support 461 Foundation of the Federal District (0193.001.262/2017), and Coordination for the 462 Improvement of Higher Education Personnel. We thank Plant Transformation Research 463 Center, University of California -Riverside, USA, for producing the transgenic plants of 464 N. benthamiana; Plant Germplasm Quarantine Station, Embrapa Genetic Resources and 465 Biotechnology, Brazil, for regulating the importation of N. benthamiana 466 (21016.000015/2020-21; 21016.004393/2020-83). 467 468 Author contributions 469 M.A.O.: conceptualization, investigation, data curation, and writing - original draft. 470 R.N.L.: conceptualization, investigation, data curation, and writing - original draft. 471 L.H.F.: investigation, data curation, and analysis. M.M.S.A.: investigation, data 472 curation, and analysis. F.L.M.: investigation, data curation, and analysis. R.V.B.: 473 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted January 11, 2024. ; https://doi.org/10.1101/2024.01.11.575089doi: bioRxiv preprint investigation, data curation, and analysis. C.L.: investigation, data curation, and 474 analysis. E.L.R.: funding acquisition, supervision, project administration, and writing - 475 review & editing. All authors revised the manuscript and approved the final version. 476 477 Data availability 478 The sequencing data obtained in this study are available in: https://github.com/Rech-479 PBSyn/Serine_Integrases 480 481 Additional information 482 Additional Supporting Information may be found in the online version of this article at 483 the publisher’s website or under request. Correspondence and requests for materials 484 should be addressed to Elibio Rech ([email protected]). 485 486

487 488 489 490 1 Ortega, C., Abreu, C., Oppezzo, P. & Correa, A. Overview of High-Throughput Cloning 491

for the Post-genomic Era. Methods in molecular biology (Clifton, N.J.) 2025, 492 3-32, doi:10.1007/978-1-4939-9624-7_1 (2019). 493 2 Casini, A., Storch, M., Baldwin, G. S. & Ellis, T. Bricks and blueprints: methods and 494 standards for DNA assembly. Nature reviews. Molecular cell biology 16, 568-576, 495 doi:10.1038/nrm4014 (2015). 496 3 Gibson, D. G. et al. Enzymatic assembly of DNA molecules up to several hundred 497 kilobases. Nature methods 6, 343-345, doi:10.1038/nmeth.1318 (2009). 498 4 Gibson, D. G. Enzymatic assembly of overlapping DNA fragments. Methods in 499 enzymology 498, 349-361, doi:10.1016/b978-0-12-385120-8.00015-2 (2011). 500 5 Sleight, S. C., Bartley, B. A., Lieviant, J. A. & Sauro, H. M. In-Fusion BioBrick assembly 501 and re-engineering. Nucleic acids research 38, 2624-2636, doi:10.1093/nar/gkq179 502 (2010). 503 6 Vazquez-Vilar, M. et al. GB3.0: a platform for plant bio-design that connects functional 504 DNA elements with associated biological data. Nucleic acids research 45, 2196-2209, 505 doi:10.1093/nar/gkw1326 (2017). 506 7 Arber, W. Polylysogeny for bacteriophage lambda. Virology 11, 250-272, 507 doi:https://doi.org/10.1016/0042-6822(60)90065-9 (1960). 508 8 Mahmood, M. A., Naqvi, R. Z., Rahman, S. U., Amin, I. & Mansoor, S. Plant Virus-509 Derived Vectors for Plant Genome Engineering. Viruses 15, doi:10.3390/v15020531 510 (2023). 511 9 Khakhar, A. & Voytas, D. F. RNA Viral Vectors for Accelerating Plant Synthetic Biology. 512 Frontiers in plant science 12, 668580, doi:10.3389/fpls.2021.668580 (2021). 513 10 Zulfiqar, S. et al. Virus-Induced Gene Silencing (VIGS): A Powerful Tool for Crop 514 Improvement and Its Advancement towards Epigenetics. International journal of 515 molecular sciences 24, doi:10.3390/ijms24065608 (2023). 516 11 Jagram, N. & Dasgupta, I. Principles and practice of virus induced gene silencing for 517 functional genomics in plants. Virus genes 59, 173-187, doi:10.1007/s11262-022-518 01941-5 (2023). 519 12 Rajput, M. et al. RNA Interference and CRISPR/Cas Gene Editing for Crop 520 Improvement: Paradigm Shift towards Sustainable Agriculture. Plants (Basel, 521 Switzerland) 10, doi:10.3390/plants10091914 (2021). 522 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted January 11, 2024. ; https://doi.org/10.1101/2024.01.11.575089doi: bioRxiv preprint 13 Zhao, J. H. & Guo, H. S. RNA silencing: From discovery and elucidation to application 523 and perspectives. Journal of integrative plant biology 64, 476-498, 524 doi:10.1111/jipb.13213 (2022). 525 14 Jin, L., Chen, M., Xiang, M. & Guo, Z. RNAi-Based Antiviral Innate Immunity in Plants. 526 Viruses 14, doi:10.3390/v14020432 (2022). 527 15 Metje-Sprink, J., Menz, J., Modrzejewski, D. & Sprink, T. DNA-Free Genome Editing: 528 Past, Present and Future. Frontiers in plant science 9, 1957, 529 doi:10.3389/fpls.2018.01957 (2018). 530 16 Pandey, P. K. et al. Versatile and multifaceted CRISPR/Cas gene editing tool for plant 531 research. Seminars in cell & developmental biology 96, 107-114, 532 doi:10.1016/j.semcdb.2019.04.012 (2019). 533 17 Adli, M. The CRISPR tool kit for genome editing and beyond. Nature communications 9, 534 1911, doi:10.1038/s41467-018-04252-2 (2018). 535 18 Lin, Q. et al. Genome editing in plants with MAD7 nuclease. Journal of genetics and 536 genomics = Yi chuan xue bao 48, 444-451, doi:10.1016/j.jgg.2021.04.003 (2021). 537 19 Smith, M. C. M. Phage-encoded Serine Integrases and Other Large Serine 538 Recombinases. Microbiology spectrum 3, doi:10.1128/microbiolspec.MDNA3-0059-539 2014 (2015). 540 20 Yang, L. et al. Permanent genetic memory with >1-byte capacity. Nature methods 11, 541 1261-1266, doi:10.1038/nmeth.3147 (2014). 542 21 Olorunniji, F. J. et al. Multipart DNA Assembly Using Site-Specific Recombinases from 543 the Large Serine Integrase Family. Methods in molecular biology (Clifton, N.J.) 1642, 544 303-323, doi:10.1007/978-1-4939-7169-5_19 (2017). 545 22 Wang, X. et al. Bxb1 integrase serves as a highly efficient DNA recombinase in rapid 546 metabolite pathway assembly. Acta biochimica et biophysica Sinica 49, 44-50, 547 doi:10.1093/abbs/gmw115 (2017). 548 23 Andres, J., Blomeier, T. & Zurbriggen, M. D. Synthetic Switches and Regulatory Circuits 549 in Plants. Plant physiology 179, 862-884, doi:10.1104/pp.18.01362 (2019). 550 24 Kassaw, T. K., Donayre-Torres, A. J., Antunes, M. S., Morey, K. J. & Medford, J. I. 551 Engineering synthetic regulatory circuits in plants. Plant science : an international 552 journal of experimental plant biology 273, 13-22, doi:10.1016/j.plantsci.2018.04.005 553 (2018). 554 25 Gomide, M. S. et al. Genetic switches designed for eukaryotic cells and controlled by 555 serine integrases. Communications Biology 3, 255, doi:10.1038/s42003-020-0971-8 556 (2020). 557 26 Stark, W. M. Making serine integrases work for us. Current opinion in microbiology 38, 558 130-136, doi:10.1016/j.mib.2017.04.006 (2017). 559 27 Lloyd, J. P. B. et al. Synthetic memory circuits for stable cell reprogramming in plants. 560 Nature biotechnology 40, 1862-1872, doi:10.1038/s41587-022-01383-2 (2022). 561 28 Bonnet, J., Subsoontorn, P. & Endy, D. Rewritable digital data storage in live cells via 562 engineered control of recombination directionality. Proceedings of the National 563 Academy of Sciences of the United States of America 109, 8884-8889, 564 doi:10.1073/pnas.1202344109 (2012). 565 29 Bernabé-Orts, J. M. et al. A memory switch for plant synthetic biology based on the 566 phage ϕC31 integration system. Nucleic acids research 48, 3379-3394, 567 doi:10.1093/nar/gkaa104 (2020). 568 30 Merrick, C. A., Zhao, J. & Rosser, S. J. Serine Integrases: Advancing Synthetic Biology. 569 ACS synthetic biology 7, 299-310, doi:10.1021/acssynbio.7b00308 (2018). 570 31 Thomson, J. G. et al. The Bxb1 recombination system demonstrates heritable 571 transmission of site-specific excision in Arabidopsis. BMC biotechnology 12, 9, 572 doi:10.1186/1472-6750-12-9 (2012). 573 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted January 11, 2024. ; https://doi.org/10.1101/2024.01.11.575089doi: bioRxiv preprint 32 Guiziou, S., Maranas, C. J., Chu, J. C. & Nemhauser, J. L. An integrase toolbox to record 574 gene-expression during plant development. Nature communications 14, 1844, 575 doi:10.1038/s41467-023-37607-5 (2023). 576 33 Chen, W., Kaur, G., Hou, L., Li, R. & Ow, D. W. Replacement of stacked transgenes in 577 planta. Plant biotechnology journal 17, 2029-2031, doi:10.1111/pbi.13172 (2019). 578 34 Li, Y. et al. Recombinase-mediated gene stacking in cotton. Plant physiology 188, 1852-579 1865, doi:10.1093/plphys/kiac005 (2022). 580 35 Deich, C. et al. T7Max transcription system. Journal of Biological Engineering 17, 4, 581 doi:10.1186/s13036-023-00323-1 (2023). 582 36 Franco, R. A. L. et al. Signal Amplification for Cell-Free Biosensors, an Analog-to-Digital 583 Converter. ACS synthetic biology 12, 2819-2826, doi:10.1021/acssynbio.3c00227 584 (2023). 585 37 Chiu, W. et al. Engineered GFP as a vital reporter in plants. Current biology : CB 6, 325-586 330, doi:10.1016/s0960-9822(02)00483-9 (1996). 587 38 Müller, K. et al. A red light-controlled synthetic gene expression switch for plant 588 systems. Molecular bioSystems 10, 1679-1688, doi:10.1039/c3mb70579j (2014). 589 39 Thomson, J. G., Chan, R., Thilmony, R., Yau, Y. Y. & Ow, D. W. PhiC31 recombination 590 system demonstrates heritable germinal transmission of site-specific excision from the 591 Arabidopsis genome. BMC biotechnology 10, 17, doi:10.1186/1472-6750-10-17 (2010). 592 40 Durrant, M. G. et al. Systematic discovery of recombinases for efficient integration of 593 large DNA sequences into the human genome. Nature biotechnology 41, 488-499, 594 doi:10.1038/s41587-022-01494-w (2023). 595 41 Rech, E. L., Vianna, G. R. & Aragão, F. J. High-efficiency transformation by biolistics of 596 soybean, common bean and cotton transgenic plants. Nature protocols 3, 410-418, 597 doi:10.1038/nprot.2008.9 (2008). 598 42 Wick, R. R., Judd, L. M. & Holt, K. E. Performance of neural network basecalling tools 599 for Oxford Nanopore sequencing. Genome biology 20, 129, doi:10.1186/s13059-019-600 1727-y (2019). 601 43 Li, H. Minimap2: pairwise alignment for nucleotide sequences. Bioinformatics (Oxford, 602 England) 34, 3094-3100, doi:10.1093/bioinformatics/bty191 (2018). 603 604 Legend of Figures 605 606 Figure 1. Schematic overview of the IntPlex@ memory genetic switch architecture and 607 Boolean-based logic gates. (A) The diagram describes the architecture of the bimodular 608 genetic switch, which consists of a gfp sequence in reverse orientation (gray) 609 surrounded by four attB and attP sites (triangles) of four serine integrases: BxB1 610 (yellow), phiC31 (lilac), Int9 (blue) and Int13 (red). For in vitro assays, the switch 611 contains the degfp gene and the switch integrated into the plant genome contains the 612 mgfp gene. (B) Int9/Int13 -based XOR logic gate for gfp orientation flipping (turn gfp 613 on). Int9/Int 13 based OR logic gate for turning gfp on and off. BxB1/phiC31 -based 614 NOR logic gate for gfp excision. In the presence of Int9 or Int13, the gfp is inverted to 615 its coding sequence ( ON state). Moreover, in the presence of Int9 and Int 13, the gfp is 616 modulated and oscillates between the on and off states. In the presence of BxB1 or 617 phiC31 input, the gfp is excised. The system output is edited DNA: plasmid or plant 618 genome. (C) The diagram shows IntPlex@ memory genetic switch operation driven by 619 BxB1. The resulting edited DNA with the BxB1 attL scar (left) and the excised circular 620 DNA molecule with BxB1 attR (right). (D) The diagram shows IntPlex@ memory 621 genetic switch operation driven by phiC31. The resulting edited DNA with the phiC31 622 attL scar (left) and the excised circular DNA molecule with phiC31 attR (right). (E) The 623 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted January 11, 2024. ; https://doi.org/10.1101/2024.01.11.575089doi: bioRxiv preprint diagram shows IntPlex@ memory genetic switch operation driven by Int9. The 624 sequence flanked by the Int9 attB and attP sites and, consequently, the inversion of the 625 gfp to its coding sequence (GFP output). The Int13 att sites are also inverted and 626 oriented towards their reverse complement. (F) The diagram shows IntPlex@ memory 627 genetic switch operation driven by Int13. The gfp flanked by the Int13 attB and attP 628 sites is flipped to its coding sequence (GFP output). (G and H) The diagram shows 629 IntPlex@ memory genetic switch operation driven by Int9 and Int13. The gfp is flipped 630 in two steps according to which integrase is added first. (G) In the first step, Int9 631 recognizes its attB/P sites and catalyzes the inversion of gfp from its OFF state to ON 632 state, resulting in the formation of the intermediate sequence with attL/Int9 and 633 attR/Int9 sites and the output GFP. Instantaneously, this intermediate sequence is the 634 substrate for the action of Int13 that recognizes its attB/P sites and catalyzes the 635 inversion of gfp from its ON state to OFF state, resulting in the formation of the final 636 plasmid with both attL/R Int9 and attL/R Int13 sites. (H) In the first step, Int 13 637 recognizes its attB/P sites and catalyzes the inversion of gfp from its OFF state to ON 638 state, resulting in the formation of the intermediate sequence with attL/Int13 and 639 attR/Int13 sites and the output GFP. Instantaneously, this intermediate sequence is the 640 substrate for the action of Int 9 that recognizes its attB/P sites and catalyzes the 641 inversion of gfp from its ON state to OFF state, resulting in the formation of the final 642 plasmid with both attL/R Int9 and attL/R Int13 sites. The corresponding colors and 643 shapes indicate the genetic components of the switch. 644 645 Figure 2. In vitro function of Int -Plex@ memory switch driven by serine integrases in 646 cell-free TX-TL system. (A) Schematic overview of the IntPlex @ memory bimodular 647 genetic switch architect ure, which consists of a RBS sequence in reverse orientation 648 (black) and a degfp sequence in reverse orientation (gray) surrounded by four attB and 649 attP sites (triangles) of four serine integrases: BxB1 (yellow), phiC31 (lilac), Int9 (blue) 650 and Int13 (red). (B) Agarose 1% gel of PCR reaction for each primer combination used 651 to confirm the presence of edited DNA. (C) Agarose 1% gel of PCR reaction for each 652 primer combination used to confirm the presence of the excised DNA with attR scar. (D 653 and E) Sanger sequencing chromatograms of excised circular DNA with attR scar. The 654 DNA fragments were obtained through PCR using pairs of primers indicated in Table 1. 655 The position of the primers is indicated by black arrows. All sequences corresponded to 656 the in silico predicted genome -edited sequences. The corresponding colors and shapes 657 indicate the genetic components of the switch. First lane is 1 kb plus ladder (Life 658 Technologies, USA). Reverse complementary sequences are indicated by a black 659 asterisk. No off-target effects were detected in the analyzed sequences. 660 661 Figure 3. In vitro function of Int -Plex@ memory switch driven by serine integrases in 662 cell-free TX-TL system. (A) The diagram shows the flipping of the degfp in two steps. 663 In the first step, Int 9 recognizes its attB/P sites and catalyzes the inversion of degfp 664 from its OFF state to ON state , resulting in the formation of the intermediate plasmid 665 with attL/Int9 and attR/Int9 sites and the output deGFP. Instantaneously, this 666 intermediate plasmid is the substrate for the action of In t13 that recognizes its attB/P 667 sites and catalyzes the inversion of degfp from its ON state to OFF state, resulting in the 668 formation of the final plasmid with both attL/R Int9 and sites attL/R Int13. (B) In vitro 669 reactions after 24h post incubation of Int -Plex@ cassette plasmid with Int 9 and Int 13 670 plasmids in Arbor Biosciences ™ myTXTL system. Furthermore, it is noteworthy that 671 since both enzymes were added simultaneously (pART27+ Int9+Int13), the ON state 672 varies according to the first integrase to catalyze the plasmid inversion. Thus, the 673 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted January 11, 2024. ; https://doi.org/10.1101/2024.01.11.575089doi: bioRxiv preprint system presents the deGFP for several days according to the ON state mRNA molecules 674 and protein half-lives. (C) Agarose 1% gel of PCR reaction for each primer combination 675 used to confirm the presence of the inverted DNA with attL/R sites. (D and E) Sanger 676 sequencing chromatograms of regions flanking the degfp obtained after simultaneous 677 treatment with Integrases 9 and 13. The DNA fragments were obtained through PCR 678 using pairs of primers indicated in Table 1. The position of the primers is indicated by 679 black arrows. All sequences corresponded to the in silico predicted genome -edited 680 sequences. The corresponding colors and shapes indicate the genetic components of the 681 switch. First lane is 1 kb plus ladder (Life Technologies, USA). Reverse complementary 682 sequences are indicated by a black asterisk. No off-target effects were detected in the 683 analyzed sequences. 684 685 Figure 4. IntPlex@ memory genetic switch for DNA excision driven by Bx B1. (A) 686 Sanger sequencing chromatogram of edited plant genome with attL/BxB1 scar. (B and 687 C) Sanger sequencing chromatogram s of excised circular DNA with attR scar. The 688 DNA fragments were obtained through PCR using pairs of primers indicated in Table 1. 689 The position of the primers is indicated by black arrows. All sequences corresponded to 690 the in silico predicted genome -edited sequences. The corresponding colors and shapes 691 indicate the genetic components of the switch. 692 693 Figure 5. IntPlex@ memory genetic switch for DNA excision driven by phiC31. Sanger 694 sequencing chromatogram of edited plant genome with phiC31 attL. The DNA 695 fragments were obtained through PCR using pairs of primers indicated in Table 1. The 696 position of the primers is indicated by black arrows. The sequence corresponded to the 697 in silico predicted genome -edited sequence. The corresponding colors and shapes 698 indicate the genetic components of the switch. 699 700 Figure 6. Nanopore sequencing evidence of genetic switch operation driven by Bx B1. 701 (A) The diagram illustrates the nanopore raw reads mapped to the predicted excised 702 circular DNA molecule with Bx B1 attR. The corresponding colors and shapes indicate 703 the genetic components of the switch. (B) A detailed view of the excision region, 704 showing the resulting BxB1 attR and the flanking phiC31 attP/attB sites. 705 706 Figure 7. Basecalled reads length vs reads quality score. The color gradient indicates 707 the density of reads, with warmer colors representing higher densities. 708 709 Figure 8. IntPlex@ memory genetic switch for gene modulation (turn gene on ) driven 710 by Int9 or Int13 . (A) Sanger sequencing chromatogram of edited plant genome with 711 Int9 attL and attR post recombination sites. The sequence shows the flipping of the 712 DNA flanked by the Int9 attB and attP sites and, consequently, the inversion of the mgfp 713 to its coding sequence (MGFP output). The Int13 att sites are also inverted and oriented 714 towards their reverse complement. (B) Sanger sequencing chromatogram of edited plant 715 genome with Int13 attL and attR post recombination sites and the inversion of the mgfp 716 to its coding sequence (MGFP output). The fragments were obtained through PCR using 717 pairs of primers indicated in table 1. The position of the primers is indicated by black 718 arrows. A pair of primers was used to amplify the complete attL site (black upper 719 arrow) and a second pair of primers to amplify the attR site (black lower arrow). The 720 sequences corresponded to the in silico predicted genome -edited sequence s. The 721 corresponding colors and shapes indicate the genetic components of the switch. 722 723 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted January 11, 2024. ; https://doi.org/10.1101/2024.01.11.575089doi: bioRxiv preprint .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted January 11, 2024. ; https://doi.org/10.1101/2024.01.11.575089doi: bioRxiv preprint .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted January 11, 2024. ; https://doi.org/10.1101/2024.01.11.575089doi: bioRxiv preprint .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted January 11, 2024. ; https://doi.org/10.1101/2024.01.11.575089doi: bioRxiv preprint .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted January 11, 2024. ; https://doi.org/10.1101/2024.01.11.575089doi: bioRxiv preprint .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted January 11, 2024. ; https://doi.org/10.1101/2024.01.11.575089doi: bioRxiv preprint .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted January 11, 2024. ; https://doi.org/10.1101/2024.01.11.575089doi: bioRxiv preprint .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted January 11, 2024. ; https://doi.org/10.1101/2024.01.11.575089doi: bioRxiv preprint .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted January 11, 2024. ; https://doi.org/10.1101/2024.01.11.575089doi: bioRxiv preprint