Design of orthogonal genetic switches based on a crosstalk map of σs, anti-σs, and promoters

Abstract

Cells react to their environment through gene regulatory networks. Network integrity requires minimization of undesired crosstalk between their biomolecules. Similar constraints also limit the use of regulators when building synthetic circuits for engineering applications. Here, we mapped the promoter specificities of extracytoplasmic function (ECF) σs as well as the specificity of their interaction with anti-σs. DNA synthesis was used to build 86 ECF σs (two from every subgroup), their promoters, and 62 anti-σs identified from the genomes of diverse bacteria. A subset of 20 σs and promoters were found to be highly orthogonal to each other. This set can be increased by combining the -35 and -10 binding domains from different subgroups to build chimeras that target sequences unrepresented in any subgroup. The orthogonal σs, anti-σs, and promoters were used to build synthetic genetic switches in Escherichia coli. This represents a genome-scale resource of the properties of ECF σs and a resource for synthetic biology, where this set of well-characterized regulatory parts will enable the construction of sophisticated gene expression programs.

Figure 1

The strategy for the genomic mining of ECF σs, anti-σs, and promoters is shown. (A) σs recruit core RNAP to promoters; a function that is inhibited by the anti-σ. σs have a two-domain structure that binds to the −10 and −35 regions of the target promoter. (B) The complete libraries of 86 synthesized σs (top row) and their 62 cognate anti-σs (bottom row) are shown organized as a phylogenetic tree. Asterisks indicate active σs (>5-fold activation) or anti-σs (>2-fold repression). Carets indicate σs or anti-σs that appear in the final orthogonal sets. All σs in the library are named ECFXX_YYYY, where ‘XX’ denotes the ECF subgroup, and ‘YYYY’ denotes the unique σ ID given by Staroń et al (2009). The anti-σs were named ASXX_YYYY, where ‘XX’ and ‘YYYY’ denote the ECF subgroup and unique ID of the cognate σ. Consequently, cognate σ/anti-σ pairs have the same numbering (e.g., ECF11_987 and AS11_987). (C) For each σ, target promoters are identified through a process of computational search, selection, and design. The first step involves the organization of the ECF operons according to the subgroups defined by Mascher and co-workers (Staroń et al, 2009).

Figure 2

Promoter models are shown for 29 ECF σ subgroups. The models contain a sequence logo illustrating the −35/−10 motifs and intervening spacer sequence. The exact −35 and −10 sequences identified by BioProspector (Liu et al, 2001) are underlined underneath each sequence logo. The bar chart histograms illustrate the number of promoters with different length distances between underlined the −35 and −10 motifs. The promoters were organized vertically to cluster similar −35 and −10 motifs, as determined by eye. The bottom three promoter models (ECF5*, ECF14, and ECF27) represent promoters that were not found to be active (>5-fold activation) in our tests. Promoter model ECF5* represents the model for subgroups 5–10.

Figure 3

The activity and orthogonality of ECF σs are shown. (A) ECF σs are induced by IPTG via a T7 expression system, and σ-dependent promoter activity was measured by gfp expression and flow cytometry. Plasmid pN565 (incW ori) encodes the IPTG-inducible T7* expression system (Temme et al, 2012); plasmid series pVRa (pBR322 ori) and pVRb (pSC101 ori) encode the ECF σ library and test promoter library, respectively. The specific example shown (ECF11_987 and P_{11_3726}) is highlighted in the following subfigures. (B) Activities of active ECF σ library members titrated against their target promoters. The gray lines show levels of GFP expression for one active ECF σ:promoter pair in each subgroup induced with 0, 10, 20, 50, and 100 μM IPTG. The averaged activity of σ ECF11_987 against its promoter P_{11_3726} is highlighted in black. Data are shown from three independent assays and error bars represent one standard deviation. Plots of the other σ:promoter pairs are shown in more detail in Supplementary Figure S4. (C) The liquid culture growth curves (OD₆₀₀) are shown for each σ under high induction (100 μM IPTG). The growth curve of σ ECF11_987 averaged from three independent growth assays is highlighted in black and the error bars represent one standard deviation. Background growth curves show data from one growth assay. The growth curves of two negative controls are shown in dark gray. Note that 64 out of the 86 σs show no growth impact as compared with the control. (D) The activity of one promoter (P_{11_3726}) is shown for the complete library of active σs expressed with 100 μM IPTG. Each bar represents the average promoter activity from at least two independent assays and error bars represent one standard deviation. The two σs from subgroup 11 that were expected to activate the promoter are bracketed. (E) All cross reactions are shown for the 20 most orthogonal σ:promoter pairs. Each σ is induced with 100 μM IPTG, and the fold induction is measured as the fluorescence with σ induction divided by the basal activity of the promoter in the absence of any σ. Each square represents the average fold induction from at least two independent assays of a unique σ:promoter combination. All promoters were named using the convention P_{XX_YYYY}, where ‘XX’ and ‘YYYY’ denote the subgroup and unique ID of the downstream parent σ gene (e.g., P_{02_2817} is the promoter upstream of σ ECF02_2817). Promoters containing synthetic UP elements were renamed to P_{XX_UPYYYY} (e.g., P_{15_UP436}). The σ:promoter pairs were ordered by the absolute amount of off-target activity caused by/affecting the pair, with the lowest off-target activity in the upper left and the highest in the lower right. (F) Promoter scores, as calculated from PWMs, are compared with the experimental measurements in (E). The promoter scores are calculated using the ECF promoter models (UP+PWM₋₃₅+PWM₋₁₀+spacer penalty) for the −60 to +20 promoter fragment including 30 nt flanking vector sequence. The ECF11_987:P_{11_3726} activity is highlighted in red. (G) ECF02_2817 and ECF11_3276 were recombined in their flexible linker region between domains 2 and 4 to create chimeric σs ECF02-11 and ECF11-02. The promoters activated by the two parental σs were similarly recombined between the −10 and −35 regions to create chimeric promoters. (H) The activity and orthogonality of the σs and chimeric σs are shown against their cognate promoters. All of the σs are induced with 10 μM IPTG and the fold induction is as defined previously. Each square represents the average fold induction from three independent assays.

Figure 4

Overexpression of ECF σs has minimal off-target effects on the host genome. ECF σs from groups 03, 16, 20, and 38 are induced with 20 μM IPTG via the T7 expression system detailed above, and genome-wide transcription is measured by RNA-seq. Transcription of sfgfp from cognate ECF promoters encoded on the appropriate pVRb-series plasmids is also measured. As a negative control lacking exogenous σs, a strain carrying pET21a in place of a pVRa plasmid, as well as pVRb03_up1198, is used. Transcription of sfgfp is induced 10–60 fold in all samples relative to the negative control strain, while changes in host genome transcription are minimal. Sequencing read counts are quantified as reads per kilobase per million reads (rpkM), which is adjusted for CDS length and total sequencing reads.

Figure 5

Anti-σs can be used to create orthogonal threshold-gated switches. (A) In addition to the expression and reporter systems shown in Figure 3A, cells contain the plasmid series pVRc (p15A ori), which allows HSL-inducible independent expression of anti-σs to bind and sequester σs. (B) Repression of ECF11_987 activity on promoter P_{11_3726} by different anti-σs. Each bar represents average fold repression, as defined by normalizing the fluorescence of cells containing the promoter with both induced σ (induced with 10 μM IPTG) and induced anti-σ (induced with 50 nM HSL) against the fluorescence of cells containing just the promoter and induced σ. Bar heights represent the average from at least two independent assays and error bars represent one standard deviation. (C) The crossreactivity of 12 anti-σs on the set of 12 orthogonal σs targeted by the anti-σs. The activity of each σ paired with its cognate promoter was measured in the absence and presence of different anti-σs. σs were partially induced (10 μM IPTG) and anti-σs maximally induced (50 nM HSL). Colors indicate fold activity repression by the anti-σ, defined as the activity of the promoter in the absence of the anti-σ divided by the activity in its presence. The anti-σ:σ pairs were arranged in the same order as the σs in Figure 3E. Each square represents the average of at least two independent assays. (D) The influence of the expression of the anti-σ is shown for a series of switches. The plots show expression of GFP from each σ-dependent promoter for differing expression levels of cognate anti-σ induced by the inducer HSL: blue, no anti-σ plasmid; green, 10 nM HSL; red, 50 nM HSL. The data represent the average of three independent assays and error bars show standard deviations. Representative cytometry distributions are shown in Supplementary Figure S15.