To demonstrate that these synthetic elements can coordinate the expression of multiple genes in a tissue-specific manner, we expressed the synthetic activator under an endosperm-specific promoter, At2S3, stacked with three reporters driven by unique synthetic promoters . As the reporters can only be turned on by the synthetic activator, we observed the expected expression of all three exclusively in seed tissue . Additionally, examination of the vegetative tissue arising from the seeds with active reporters showed no reporter activity, with expression cycling from seed to seedling and from flowering plant to seed. Our findings demonstrate the use of synthetic promoters for the tissue-specific regulation of multiple genes simultaneously. Coordinating the expression of multiple genes in an inducible or environmentally responsive manner may provide a solution to many challenges in plant engineering. To highlight how our synthetic promoters may address some of these issues, we designed a similar circuit with the synthetic activator driven by a phosphate responsive promoter, AtPht1.1, which is induced under low external phosphate concentrations. Stable Arabidopsis transformants display the expression of all reporter genes in response to phosphate deprivation in the medium, and as expected, reporters were not observed when phosphate was supplemented . These results display how our synthetic system may enable control over stacked genes in applications that require engineering integrated with environmental cues . Importantly, these data also demonstrate our system can be transferred between different species while retaining its functional properties. Additionally,grow trays 4×4 no phenotypic or growth disturbances were observed demonstrating the orthogonality of the parts.
This strategy can be further expanded in future studies by linking this cue to a series of transcriptional events by expressing a second trans-element under a synthetic promoter to generate multiplexed transcriptional cascades originating from a single endogenous signal. Although Gal4-based synthetic TFs have previously been developed for plant systems, there are clear applications that would benefit from the expansion and development of other synthetic TFs. After validating our promoter design strategy, we investigated how conducive other TFs from different protein families would behave in our system, specifically MADS , Homeobox , GATA , and bZIP type TFs. We tested how these TFs behave when truncated to the predicted DNA-binding domain, and their potential to be functionally reconstructed through the fusion of transcriptional regulatory domains . Thus, in an effort to expand our parts library and explore the versatility of our approach, we designed and characterized additional sets of synthetic TFs along with corresponding promoter libraries. Cis– elements for Yap1 and Gat1 were designed using experimentally determined sequence motifs based on position-specific affinity matrices with randomly chosen bases for ambiguous nucleotides in each motif . Cis-elements for MCM1, Mata1, and Matα2 were generated with sequences from previously characterized yeast promoters and were designed to bind all three TF types . Complete synthetic promoter libraries for each set were then assembled through addition of the WUS minimal promoter, chosen for its high output in the initial library. Our orthogonal TF design strategy introduces protein modifications to alter expression strength while expanding the parts library. We tested two known transactivation domains , C1 from Zea mays and VP16 from herpes simplex virus type 1, to examine their effect when appended to the tested TFs. Our trans-element library was generated by fusing these TADs to either the full-length TF or the truncated TF consisting of solely the predicted DNA-binding domain . Additionally, these predicted DNA-binding domains were tested without a TAD fusion.
We selected two predicted domains for MCM1, Gat1, and Yap1 , and a single domain for Mata1 and Matα2 . The prospect of designing a minimal and modular DNA-binding protein on which various activation or repression domains can be appended could drastically expand the space for synthetic tool development. All modified trans elements also include an SV40 nuclear localization sequence to ensure proper import into the nucleus150. Using this strategy, we designed and synthesized various activating trans-elements for each TF and compared their expression output across a subset of their corresponding promoter library. We designed our screen to evaluate the efficacy of transcriptional activation using truncated TFs as a minimal DNA-binding scaffold. Comparing truncated TFs to their full-length versions revealed that minimal DNA-binding scaffolds could be directly fused to TADs to enhance gene expression, and in most cases outperform the full-length TF . The addition of a TAD to the truncated Gat1 TF chassis resulted in increased expression over the full-length TF with or without an activator, showing behavior congruent to our expectations . An interesting result was observed when analyzing how the DNA-binding domains of MCM1 and Yap1 behave without the addition of a TAD. Often times these trans-elements consisting of solely the DNA-binding domain resulted in substantial increases to expression strength. This would imply that the predicted DNA-binding domain of these constructs has inherent activation properties that may be interrupted when a TAD is fused. While the molecular basis of these observations has not been elucidated in this study, this phenomenon reveals the potential of limiting the genomic footprint of synthetic elements by utilizing minimal units with desirable transcriptional regulation properties in synthetic circuit designs. For both homeobox TFs tested , the trans-elements generated by the addition of the C1 or VP16 activation domain to the full-length TF generally resulted in higher expression levels than those built on the minimal DNA binding scaffold. This may be due to the removal of the flexible C-terminal tail that provides stability to the TF/DNA/protein transcriptional complex. While this was the observed trend, there were exceptions as demonstrated in Figure 10e.
Overall, the modifications we made to the native TF increased expression output, except in limited cases, as shown for Matα2 with promoter pMAlpha_9 where the full-length TF yielded higher expression than the altered versions . Importantly, we also demonstrate how both the promoter and the trans-element used to drive its expression alter the overall behavior. An additional level of control can be designed into our system with the introduction of repressive regulators, permitting logic principles in genetic circuits. To explore this potential, we designed repressive trans-elements that bind our synthetic promoters to repress transcription. As a proof of concept, we fused the SRDX repression domain to the Gal4 DNA-binding domain. When synthetic repressors were co-infiltrated into tobacco leaves with synthetic promoters driving GFP, GFP fluorescence decreased, indicative of a repression in gene expression . We then generated additional repressive trans-elements by fusing the SRDX domain to the other orthogonal TFs we tested . Addition of the SRDX domain to other TF types engendered repressive trans-elements capable of limiting the basal expression of their corresponding promoters. Though, repression functionality was often dependent on the location of the terminal fusion, these trends varied from family to family. Plant engineers can utilize these synthetic promoters intandem with repressor logic to achieve tissue specific gene repression, by driving the repressing trans-element with a tissue-specific promoter or build complex gene circuits that can be both activated and repressed. Surprisingly, the addition of the SRDX domain to some of the trans elements resulted in no change in expression, or even the increase in expression in some case. It is important to note,horticulture products modifications made to a TF from one family may be consistent in behavior with the modifications made to another, as context dependence plays a role in determining functional protein fusions. Often times the addition of a given regulatory domain alters behavior inverse to the expected, highlighting the need to empirically test various modification schemes when designing parts for optimal behavior in planta. Nonetheless, our findings summarize a set of trans-elements that can be used as effective repressors for future plant engineering efforts. To expand the dynamic potential of our system, we developed synthetic promoters with ciselements designed to bind multiple TF types; Mata1, Matɑ2, and MCM1. This was inspired by the native yeast mating system that determines haploid cell compatibility, and regulates the switch from haploid-specific gene expression to diploid-specific gene expression after mating. These promoters are regulated by MCM1, a-specific TFs , and ɑ- specific TFs at the cis-regulatory region, with output determined by the combination bound at a given time. This requires these promoters to have binding sites for multiple TFs allowing for the regulation of a single output with a multi-input parameter, allowing for more complex logic principles to be introduced. With this in mind, we designed two sets of the synthetic promoters with hybrid cis-elements composed of binding motifs for MCM1, Mata1, and Matɑ2 . Each trans-element from these families can interact with the hybrid promoters increasing the number of potential combinations and output range. For pMa_8 specifically, the background expression can be repressed with M2-D-SRDX, while MC-D1-C1 and M1-VP16 increase promoter output at varying levels.
This demonstrates expression strength modulation of a single promoter by altering the trans-element used to drive its expression. Although the development of our synthetic promoters was intended for protein accumulation, we used this hybrid promoter system to examine the correlation between protein and transcript abundance at two days and five days post-transformation. These data demonstrate a strong correlation confirming our method of quantifying output at the protein level is sound for observing changes at the transcriptional level . Potentially, the most powerful application of these hybrid promoters is their capacity to introduce multi-gated logic principles into genetic circuits. A simple ‘or’ gate can be generated by utilizing two activating trans-elements from MCM1, Mata1, or Matɑ2, with the promoter in the on state when either activator is present. Additionally, it is possible to develop a reciprocal ‘nor’ gate by utilizing two repressor constructs in concert with a synthetic promoter with high levels of background expression. The promoter remains in the off state when either repressor is bound, only activating when both are absent. Another interesting application of our repressor constructs is the potential to generate a genetic kill-switch that will shut down promoter activity even in the presence of an activating trans-element, as demonstrated by the combination of M2-D-SRDX and MC-D1-C1 with pMa_8 . Utilizing our multi-binding site promoter system, we lay the foundation for the fabrication of more complex and elegant genetic circuits in plants.A major challenge to circuit engineering has been the unravelling and elucidation of genetic determinants underpinning gene expression in eukaryotes, especially in plants. A core tenet of synthetic biology is the ability to understand the fundamental and reductionist rules that govern natural systems in order to reconstruct and engineer artificial molecular components of life. Our findings demonstrate a strategy to investigate contributions of various cis-elements, minimal promoters, and trans-elements that dictate gene expression, providing the foundation for future studies to rationally design transcriptional regulation systems with predicted expression strengths a priori. Although there are many nuances to transcriptional regulation in plants that have not been elucidated in this study, our results take one step towards the coarse dissection of the contributing effect of specific genetic elements in controlling gene expression. These findings may provide the foundation for the future identification of design principles that will enable the construction of more refined and targeted transgene expression systems in plants. It is important to note that the orthogonality of our system may vary in different plant species as imported cis elements may already be present in their genome; however prior work with the Gal4 system in Arabidopsis has suggested that there are no major pleiotropic effects from the introduction of new synthetic TFs. This will have to be confirmed on a case-by-case basis in future studies utilizing these parts in diverse plant species. Nonetheless, there is potential to expand our conceptual approach for promoter and TF design to other eukaryotes by fusing native core promoters and orthogonal cis-elements to build up DNA parts in less developed systems. A major obstacle that has often thwarted plant engineering efforts is the natural phenomenon of transgene silencing. Plants have evolved robust defense mechanisms that may perceive multiple transgenes driven by the same promoter as a threat, resulting in gene silencing at the transcriptional and post-transcriptional level. Thus, although many engineering efforts require the coordinated expression of multiple genes, it has long been observed that stacking the same promoter multiple times may also dramatically increase the chance of gene silencing. Inherent to our library design of synthetic elements is the avoidance of identical sequences, as addressed by the randomization of varying CCE combinations and minimal promoters to avoid the use of homologous sequences. This strategy permits the stacking of multiple genes with distinct promoters and may circumvent potential gene silencing.