Team:Wageningen UR/Results/Phage Repression

Xylencer

subproject hexbadge

Phage Repression

divider

The goal of this project was to gain control of the proliferation of the Xylencer bacteriophage in our phage delivery bacteria (PDB) by constructing a genetic switch mechanism. The requirement of the switch is to be tightly controllable and quick to switch from the OFF-state, meaning full phage inhibition, to the ON-state, meaning phage proliferation initiation. This way the genetic circuit ensures the coordinated and rapid production of Xylencer phages upon Xylella fastidiosa presence. We achieved this, by using dead versions of Cas9 (dCas9) and Cas12a (dCas12), which were shown to inhibit protein translation by 95-99%. Further, we showed reversion of inhibition through expression of Anti-CRISPR (Acr) molecules. By combining Acr and dCas9 we constructed a genetic circuit enabling gene repression and expression restoration. Functionality of the circuit was proven by controlling mrfp and gfp expression through targeting the early transcript regulatory region of phage Lambda. Although dCas9 and dCas12a have previously been shown to significantly inhibit gene expression in a variety of organisms, we were here unable to inhibit bacteriophage proliferation.

Introduction

CRISPR interference (CRISPRi) makes use of catalytically inactive variants of Cas9 (dCas9) or Cas12a (dCas12a) proteins to suppress gene expression [1]. Identical to their active counterparts, the co-expression of guide RNAs directs the ribonuclease protein (RNP) to its specific DNA target sequence. However, introduction of mutations in the RuvC1 and HNH nuclease domains of Cas9, and the RuvC I and RuvC II domains of Cas12a, cause the Cas protein to lose endonuclease activity, without impeding the DNA binding [2; 3]. This enables the reversible transcriptional inhibition by tightly DNA-bound dCas proteins, contrary to irreversible cleavage by active Cas9 or Cas12a. One way to reverse the effect of dCas-mediated gene repression is through their natural inhibitors, known as Anti-CRISPR (Acr) proteins. Acrs are small, phage-derived proteins blocking the natural CRISPR immune system of bacteria [4]. In most cases, they directly interfere with Cas nucleases, blocking binding or cleavage of the target DNA [5]. Therefore, Acrs may represent a powerful tool for the optimization of CRISPR/Cas-based genome editing approaches or the construction of synthetic circuits [6].

The genomes of the X. fastidiosa phages Sano & Salvo are organized in four different operons [7]. In a similar fashion, the phage Lambda genome is segmented into various partly overlapping operons. These phage operons summarize genes by function in the phage’s biology and/or time point of expression during the phage’s lifecycle [8]. In Xylencer, we developed a tunable dCas9/dCas12a-Acr synthetic gene circuit, controlling the immediate-early operon promoters PL and PR of the E. coli phage Lambda. PL and PR control the expression of “early transcripts” involved in the organization of phage DNA replication and phage particle formation. By repression and controlled repression relief of these early transcripts operons of phage Lambda, we aim to gain control of phage proliferation in host bacteria.

In Xylencer, we first assessed the function of the gene circuit on repressing and restoring GFP fluorescence using single-targeting dCas9. Second, repression of the promoters PL and PR controlling gfp and mrfp expression, respectively, was tested with a multitargeting dCas12a and Acr-dCas9 gene circuit. Third, repression of phage Lambda proliferation using dCas12a was examined in plaque and in kinetic plate reader assays.

Schematic overview of the regulation of phage proliferation by Acr-dCas9 gene circuit

Gene Circuit

  • Experimental Approach arrow_downward

    The AcrIIA4-dCas9 gene circuit mainly consists out of three parts; the AcrIIA4, the dCas9, and the sgRNA expression module. The Acr expression is under the control of the L-rhamnose inducible promoter (Prha and shares a bi-directional terminator with the dCas9 gene (Part:BBa_K3286009). The dCas9 is being expressed via the constitutive tet promoter but regulated via the IPTG-inducible lacI/lac operator (Ptet/lac)(Part:BBa_K3286008). The sgRNA (spacer and scaffold) are expressed by the strong constitutive J23119 promoter (Part:BBa_K3286003). The sequences for terminators, promoters, dCas9 and the sgRNA scaffold were provided by the Laboratory of Microbiology. The sequence of the acrIIA4 gene was retrieved from NCBI and ordered as a gene block via Integrated DNA Technologies. The circuit was inserted into the pACYC184 vector with ori p15a and chloramphenicol resistance using High Fidelity Assembly.

    Figure A1: Sequence Map of the AcrIIA4-dCas9 gene circuit. Promoters are displayed in green, cds in purple, terminators in orange, the lac operator in black, the non-coding sgRNA in blue. D10A and H840A refer to the two point mutations converting an enzymatically active Cas9 into the inactive dCas9.

    Assays were conducted in an E. coli DH10β strain carrying a genomically inserted lacUV5_gfp expression cassette. spacers were designed to either target the –10 element of the lacUV5 promoter or within the first 100 nucleotides of the gfp coding sequence (cds).

    Figure A2: Sequence Map of the lac UV5 promoter and first 100bp of gfp. The Lac UV5 promoter is displayed in white, the gfp cds in green, spacers in grey and PAM sequences in purple.

Fluorescent Loss Assay

divider

dCas9 and Acr-dCas9 constructs with three different targeting spacers each where tested in a plate reader experiment for the loss of GFP fluorescence and restoration of GFP signal upon Acr induction. Figure 1 shows reduction of relative GFP fluorescence by 90 – 99% for constructs dCas9 spacer 1 – 3. A trend is visible pointing towards spacer 1 being most effective, spacer 2 less, and spacer 3 least. The effectiveness of gfp repression correlates with the distance of the spacers from the lacUV5 promoter region (see approach). The results therefore confirm the findings of Larson et al., 2013 [16], who stated that targeting the promoter region is most effective as compared to targeting the 5’ UTR or cds further downstream. For constructs Acr-dCas spacer 1 – 3 the effect of the Acr is clearly visible as 0,2 and 1% L-rhamnose induction results in up to 90% of GFP signal restoration. However, even if not induced via L-rhamnose, leaky Acr expression restores fluorescence to an extend of 50 - 80%. Again, the efficiency of repression decreases from spacer 1 to spacer 3. The strong influence of Acr promoter leakiness does not meet the requirements of a tight genetic switch mechanism.

Figure 1:(AcrIIA4-)dCas9 targeting GFP. dCas9 targeting gfp expression with three different spacers, either with or without AcrIIA4. Different shades represent different amounts of L-rhamnose added inducing acrIIA4 expression levels. E. coli DH10ß served as a negative control and was subtracted from samples. Fluorescence is relative to (Acr-)dCas9 with non-targeting spacer. Data represent the averages of three biological replicates.

RBS randomization

divider

In order to lower the effect of basal acrIIA4 expression, the ribosomal binding site (RBS) of acrIIA4 was mutagenized. Via PCR, using a primer allowing two different nucleotides at four positions of the RBS core region, 16 (24 = 16) different RBSs were created (Figure 2). The resulting PCR product containing the mutagenized RBS was cloned upstream of acrIIA4 into the Acr-dCas Spacer 1 vector. As Spacer 1 was identified as most efficient regarding gfp repression, experiments with Spacer 2 and 3 were discontinued. After transformation, 43 colonies were picked and screened in a fluorescence loss assay for both low basal acrIIA4 expression when not induced, and high acrIIA4 expression, when induced by L-Rhamnose. The results of ten picked colonies shown in Figure 3 represent the full range of lowest to highest basal AcrIIA4 effect on gfp repression. The randomization of the acrIIA4 RBS resulted in different fluorescence repression levels, reaching from almost 0% expression to the original expression level of 28% (spacer 1). Constructs presented in Figure 3 were sequenced to link the point mutations in the RBS core region to the effect of AcrIIA4 mediated dCas9 repression (Table 1). Sequence identity between various RBS reduced the number of the randomized RBS library to seven different sites. Translation initiation rates of RBS were predicted using the De Novo DNA RBS Library Calculator [9; 10]. Predicted translation initiation rates of acrIIA4 correlate to Acr-mediated GFP signal restoration to some degree (Table 1). Also, the data indicate a correlation between levels of GFP signal restoration upon 0,2% L-rhamnose induction and gfp repression levels without induction. As RBS C11 showed lowest basal acrIIA4 expression and highest difference in relative GFP signal between the two treatments (data not shown) it was chosen for follow-up experiments.

Figure 2: RBS Randomization via PCR. Primer for RBS randomization allowing two different amino acids at four positions of the six bp RBS core sequence. S(trong) = Adenine or Thymine, W(eak) = Cytosine or Guanine. The RBS is marked in grey, randomization primer site in yellow, the Start Codon and cds of AcrIIA4 in green and purple, respectively.

Figure 3: AcrIIA4-dCas9_randomized-RBS targeting gfp. Acr-dCas9 spacer 1 targeting gfp expression with either the original RBS (spacer 1) or randomized RBS. Different shades of green represent different amounts of L-rhamnose added inducing acrIIA4 expression. Samples were ordered by lowest to highest basal acrIIA4 expression, except for spacer 1. dCas9 lacking AcrIIA4 served as a negative control and was subtracted from samples to see differences solely caused by AcrIIA4 presence. Fluorescence is relative to dCas9 with a non-targeting spacer. Data represent the averages of two biological replicates.

  • RBS Randomization Library arrow_downward
    Table 1: RBS randomization library. Nucleotides 2 – 5 of the acrIIA4 associated RBS core region were mutagenized and sequenced. Translation Initiation Rates of RBS were predicted using the De Novo DNA RBS Library Calculator. Bold letters represent point mutations as compared to the randomization template spacer 1 (Sp1).
    RBS no. Core Sequence Predicted Translation Initiation Rate (au)
    C11 ACGTGA 3.84
    B10 (C8) AGCTGA 6.03
    D2 AGGTCA 6.76
    C7 (D8) ACGAGA 12.01
    A12 AGGTGA 37.75
    B6 (C2) AGGACA 50.26
    Sp1 (A9) AGGAGA 129.33

RBS C11 was cloned in the control plasmids to repeat the initial fluorescence loss assay of Figure 1 and compare the effect of AcrIIA4 mediated GFP signal restoration among the different conditions (Figure 4). Relative fluorescence of dCas9 was subtracted from Acr-dCas and Acr-dCas_RBS-C11 to see the effect of AcrIIA4 presence on the circuit. Results show that by changes in the RBS, the basal expression of acrIIA4 under no L-rhamnose induction is lowered by ca. 30%. Restoration of fluorescence via acrIIA4 expression is lower in Acr-dCas_C11 as compared to Acr-dCas (Figure 4). Even though this was to be expected with regard to the data of Figure 3, the effect is stronger than anticipated.

Figure 4: dCas9 targeting gfp expression with different acrIIA4 associated RBSs. Different shades of green represent different amounts of L-rhamnose added inducing acrIIA4 expression levels. E. coli DH10ß served as a negative control and was subtracted from samples. dCas9 lacking AcrIIA4 served as a negative control and was subtracted from samples to see differences solely caused by AcrIIA4 presence. Fluorescence is relative to dCas9 with a non-targeting spacer. Data represent the averages of three biological replicates.

Discussion

divider

Presented data show that dCas9 and AcrIIA4 can be combined to construct a switch for control of gene expression. However, the system is leaky and does not guarantee a full inhibition or activation. By fine-tuning the basal expression levels of acrIIA4, the switch can be regulated to have a tighter ON- or OFF-function. Higher basal expression leads to a more sophisticated ON-switch, restoring fluorescence by about 90% (Figure 1) when induced. In contrast, lower basal expression of acrIIA4 allows a tighter OFF-state, limiting GFP signal to below 5% when not induced (Figure 4). However, our data indicate a trade-off between the ON- or the OFF-state, allowing for the tight control of either one of the states. It was shown that reduction of basal acrIIA4 expression can be achieved by varying RBSs, reducing the translation rate of the associated gene. Furthermore, other mechanisms of tight expression control are worth exploring, such as riboswitches or the usage of tighter regulated promoters.

In Xylencer, we constructed a genetic switch with a strong OFF-function to prevent uncontrolled proliferation and spread of Xylencer phages. Obtained data show that tight control of the circuit can be achieved via fine-tuning of the acrIIA4 expression. We described a variety of RBSs allowing different acrIIA4 expression levels. Among these, we identified RBS C11 serving our needs for achieving a tight control of the OFF-state. In Xylencer, the expression of acrIIA4 is controlled by a riboswitch linked to a DSF receptor sensing the presence of X. fastidiosa. Linking the circuit to a riboswitch allows for even stronger control of the gene circuit and consequently phage proliferation.

Phage Repression

  • Experimental Approach arrow_downward

    One of the first discovered and categorized bacteriophages was Enterobacteria phage λ (Lambda) by Esther Lederberg in 1953 [11]. In the past six decades, bacteriophage Lambda has extensively been researched by multiple groups and therefore established itself as a model organism for bacteriophages [12; 13].

    When bacteriophage Lambda infects a bacterial cell, essential products for phage proliferation are transcribed from the very early and early operon [8; 12]. The initiation of transcription of the very early and early operon starts at Lambda promoters PL and PR, respectively (Figure A3). Upstream and partly overlapping those two promoters are two operator regions, OL and OR, which each have three binding sites for Lambda’s regulatory proteins CI and Cro [14]. The OL and OR regions within the Lambda genome are essential for Lambda's genetic network by transcriptional control of the PL, PRM and PR promoters and subsequently their downstream products (Figure A4). Therefore, we will from now refer to the very early and the early operon initiation region of bacteriophage lambda, containing the three promoters and two operator regions, the Early Operon Regulatory Region (EORR).

    Figure A3. Bacteriophage Lambda genome, with its regions and most important promoters. The Operon Regulatory Region (EORR) is composed of three promoters and two operator regions OL and OR. The OL and OR operator regions are each composed of three binding domains that facilitate the binding of the regulatory proteins CI and Cro. The OL and OR regions are adjacent and partly overlapping with the PL, PRM and the PR promoters, enabling bacteriophage lambda to control its transcriptional products by promoter activation and/or activation.

    The bacteriophage Lambda promoter regions, with operators (OL and OR) and the 5’ UTR’s (Part:BBa_K3286046) were synthetically produced as a gBlock. The gBlocks were amplified with an overhang PCR and cloned into a mrfp and gfp containing PuA66 backbone plasmid. This plasmid contains a left anti-sense mrfp cds (Part:BBa_K3286045) and a sense gfp cds, conform bacteriophage Lambda PL, PRM and PR promoter directions (Figure A4).

    Figure A4. The sequence map of mrfp-gfp downstream of Lambda’s Early Operon Regulatory Region (EORR). The promoters are displayed in purple, the operators in white, 5’UTR elements (shine-Dalgano (SD) sequence, NutL and RNase III sensitive hairpin loop) are gray. The parts (Part:BBa_K3286045), (Part:BBa_K3286046) and (Part:BBa_E0040) were used to compose this construct.

    The Cas12a spacer array plasmid was amplified by overhang PCR, generating a linear sequence with a repeat and additional restriction sites at either flank. An insert was generated by annealing of two DNA oligo’s and ligated into a KpnI and BbsI digested linear backbone. The insert contained a Cas12a repeat region and a non-targeting spacer sequence with two BbsI digestion sites. Utilizing this plasmid, we were able to produce multiple spacers, simply by BbsI digestion and consecutive ligation of an annealed oligo or gBlock. This resulted in (Part:BBa_K3286041), a modular system for simple integration and production of new Cas12a (Cpf1) spacer arrays. The two plasmids containing dCas12a and the Cas12a spacer array originated from [15].

    Figure A5. The sequence map of dCas12a (Part:BBa_K3286040). The promoter is displayed in purple, the dCas12a protein in orange and the terminators in blue.
    Figure A6. The sequence map of the dCas12a (dCpf1) single- and multi-targeting spacer array (Part:BBa_K3286041). The promoters are displayed in purple, the repeat regions in orange, the spacer in gray and the terminators in blue. A. contains a single spacer for single targeting, which in the case of Part:BBa_K3286041 harbors two BbsI sites in the spacer sequence (dotted boxes). B. Contains multiple repeats and spacers, that have been used for multi-targeting (multiplexing), like PL and PR.

    Assays were conducted in E. coli DH5α strains, which were co-transformed with plasmids containing dCas12a (Figure A4) and the mrfp and gfp downstream Lambda’s (EORR) (Figure A3). Cells were made competent again, to ensure a standardized starting culture for the transforming of the spacer arrays and subsequently the inhibition assays. The spacers were designed to target the PL, PRM and the PR promoters within their -35 and -10 region, due to significant inhibition of downstream products shown in previous research [1; 2; 16].

Test Plasmid System

divider
Novel Bacteriophage control circuit inhibition by dCas12a

dCas12a and six different spacer arrays were tested in a plate reader for their potential to inhibit bacteriophage Lambda’s EORR and consequently the expression of mRFP and GFP. Figure 5 shows a relative reduction of fluorescence for targeted promoter regions of 95–98%. In addition, a 30-32% inhibition of mRFP can be observed in samples PR and PRM+PR, while the upstream promoter is not targeted by dCas12a. This could be due to the fact that either the dCas12 protein is in a too close proximity while targeting adjacent promoters, resulting in steric hinderance between RNA Polymerase (RNAP) and the dCas12a protein. Another explanation could be the flexibility of spacer-target binding, by tolerating small amounts of mismatches and therefore binding non-specifically on the relative similar PL promoter in the case of PR promoter targeting [17]. Overall, we observe a significant inhibition of transcriptional products by targeting the PL and PR promoter. Therefore, we will further asses this array for the potential to regain fluorescence upon induction of AcrIIA4 and finally the inhibition of bacteriophages.

Figure 5. The relative fluorescence of mRFP and GFP downstream bacteriophage Lambda’s EORR, targeted by six different spacer arrays. Samples were normalized to a control sample containing dCas12a, mrfp and gfp downstream Lambda’s EORR, but lacking a spacer array. Six spacer arrays were used, a non-targeted (NT) and different combinations of PL, PRM and PR targeting spacers. Data presented is an average of three biological replicates.

Novel Bacteriophage control circuit inhibition by dCas9

Figure 6 shows PL and PR to be the most promising targets for controlling gene expression of the EORR. Therefore, two dCas9 spacers for PL and PR were designed to assess the functionality of the Acr-dCas9 gene circuit in both regulating expression of the EORR and multitargeted repression. For the Acr-dCas plasmids, RBS C11 was used to reduce basal AcrIIA4 expression levels. Repression of GFP and mRFP fluorescence via dCas plasmids show strong inhibitory effects. A direct comparison between repression strength of the single- and multi-targeting approach cannot be made. spacers and targets used for single-targeting assays in Figure 1 and 5 differ compared to spacers used for Figure 6 which target the -10 or -35 element of the PL or PR promoter. Regarding the Acr-dCas plasmids, observed fluorescence restoration levels are comparable to the single-targeting approach in Figure 4. Basal expression of acrIIA4, when not induced, results in partial restoration of GFP (25–30%) and mRFP (5–10%) fluorescence. Upon L-Rhamnose induction, AcrIIA4 is restoring GFP and mRFP fluorescence to 85–100% and 70–100%, respectively. Similar to the single-targeting approach, a trade-off between the ON- or the OFF-state, allowing for the tight control of either one of the states, is visible. The differences in repression and restoration between mRFP and GFP signal could be explained by the PR promoter, which controls gfp expression, being stronger than the PL promoter of mrfp [17].

Figure 6: (Acr-)dCas9 double targeting gfp & mrfp. dCas9 targeting gfp and mrfp expression with three different spacer combination, either with or without AcrIIA4. Different shades of green and orange represent different amounts of L-rhamnose added inducing AcrIIA4 expression levels. E. coli DH5α served as a negative control and was subtracted from samples. Fluorescence is relative to (Acr-)dCas with non-targeting spacer. Data represent the averages of three biological replicates.

Test Phage

divider
Bacteriophage inhibition by dCas12a

To assess the potential of bacteriophage Lambda inhibition by dCas12a, we performed a bacteriophage spot assay. Infected E. coli DH5α cells harboring the dCas12a and the PL+PR CRISPR array are visualized in Figure 7. The bacteriophages were dripped on the plate in a serial dilution from 10-1 to 10-8. The bacteriophages have previously been shown to have 5,04 x 1010 Plaque Forming Units (PFU) ml-1 for a control sample without an CRISPR array.

The spots in serial dilution showed high number of plaque formation, throughout the range of dilutions. This indicated that the bacteriophages were able to proliferate, although the bacterial cells contained a CRISPR-Cas system. Potential explanations for the evasion are that either a 95% inhibition of transcript products is not sufficient for bacteriophage proliferation inhibition (see the prior results). There is competition between DNA binding of the regulatory proteins (CI and Cro) of the bacteriophage and the Cas12a protein in the operator regions [12]. Or, the bacteriophages evade the CRISPR-Cas system by proto-spacer-associated motif (PAM) mutations [18].

Figure 7. Bacteriophage spot assay in serial dilution, in triplicate, for E. coli DH5α cells harboring dCas12a (Part:BBa_K3286040) and a spacers array targeting the Lambda PL and PR promoter (a derivative of Part:BBa_K3286041). 2μl of serial diluted phage stock solution was dripped on top of soft agar layer and incubated overnight at 37 C.
Growth curve analyses of bacteriophage infected E. coli cells,

Performed bacteriophage spot assay to show inhibition of phage proliferation resulted in a lack of inhibition of bacteriophage proliferation. Therefore, we performed a kinetic growth curve analysis by infecting E. coli DH5α cells harboring the dCas12a and the PL+PR CRISPR array constructs. Cells were infected with 5 different multiplicity of infections (MOI), ranging from 10 to 0,001 bacteriophages per bacterial cell. As a control, we included a sample with only the dCas12a construct and without the CRISPR array. Cells were grown overnight, diluted to an OD600 of 0.02, incubated to recover to an OD600 of 0.10 and infected with phages. Figure 8 and 9 show the growth curve of the control and the sample harboring the CRISPR-Cas system. When we compare the results of the control (Figure 8) to the results of the CRISPR-Cas harboring samples, we cannot observe a significant difference and therefore we must conclude that we are unable to inhibit bacteriophage proliferation with dCas12a targeting the PL and PR promoters.

Figure 8. BActeriophage infected E. coli DH5α cells harboring singularly the dCas12a construct (Part:BBa_K3286040) and no spacer array. Cells were infected with bacteriophages at an OD600 of 0.10 and measured (with intervals of 4 minutes) over a period of 15 hours. Data presented is an average of three biological replicates.
Figure 9. BActeriophage infected E. coli DH5α cells harboring dCas12a (Part:BBa_K3286040) and a spacers array targeting the Lambda PL and PR promoter (a derivative of Part:BBa_K3286041). Cells were infected with bacteriophages at an OD600 of 0.10 and measured (with intervals of 4 minutes) over a period of 15 hours. Data presented is an average of three biological replicates.

Discussion

divider

Data presented in Figures 5 and 6 demonstrate that the EORR can be inhibited by dCas9 and dCas12a by more than 95%. When solely the PL promoter is targeted (Figure 5), we observe a significant increase of GFP compared to the non-targeted (NT) spacer array. PL and PR are strong promoters [17][21], that can produce high quantities of transcript products. Therefore, we hypothesize that upon silencing of the PL promoter additional resources are available for products of other promoters.

We could show strong inhibition of the EORR with either dCas9 and dCas12a. By applying our Acr-dCas9 gene circuit, we were able to control gene expression from the EORR. Although, we gained control over Lambda’s EORR, we did not observe this trend with bacteriophages. Even when we dilute the bacteriophages to a low MOI, we cannot observe a significant change to the control. We hypothesize that although 95% inhibition is a significant number, low amounts of transcript products from the EORR can be produced and therefore result in effective proliferation of the bacteriophage. Due to the fact that we target promoters in Lambda’s EORR, which includes multiple overlapping operator regions with the dCas12a targets, we hypothesize that the dCas12a, CI and Cro proteins are actively competing for the binding to target regions [19; 20]. In addition, due to the fact that bacteriophages exhibit high mutation rates, evasion of the CRISPR-Cas system can be realized [18].

Finally, additional experiments conducted by other team members hinted to the potential of a contaminated bacteriophage stock (see Alternative Phage Engineering), but due to time limitations we were unable to perform new experiments with a verified and standardized stock.

  • References arrow_downward
    1. Qi, L. S., Larson, M. H., Gilbert, L. A., Doudna, J. A., Weissman, J. S., Arkin, A. P., & Lim, W. A. (2013). Repurposing CRISPR as an RNA-γuided platform for sequence-specific control of gene expression. Cell, 152(5), 1173–1183. https://doi.org/10.1016/j.cell.2013.02.022
    2. Bikard, D., Jiang, W., Samai, P., Hochschild, A., Zhang, F., & Marraffini, L. A. (2013). Programmable repression and activation of bacterial gene expression using an engineered CRISPR-Cas system. Nucleic Acids Research, 41(15), 7429–7437. https://doi.org/10.1093/nar/gkt520
    3. Zetsche, B., Gootenberg, J. S., Abudayyeh, O. O., Slaymaker, I. M., Makarova, K. S., Essletzbichler, P., … Zhang, F. (2015). Cpf1 Is a Single RNA-Guided Endonuclease of a Class 2 CRISPR-Cas System. Cell, 163(3), 759–771. https://doi.org/10.1016/j.cell.2015.09.038
    4. Bondy-Denomy, J. et al. (2013) ‘Bacteriophage genes that inactivate the CRISPR/Cas bacterial immune system.’, Nature. England, 493(7432), pp. 429–432. doi: 10.1038/nature11723.
    5. Trasanidou, D., Gerós, A. S., Mohanraju, P., Nieuwenweg, A. C., Nobrega, F. L., & Staals, R. H. J. (2019). Keeping CRISPR in check: diverse mechanisms of phage-encoded anti-crisprs. FEMS Microbiology Letters, 366(9), 98. https://doi.org/10.1093/femsle/fnz098
    6. Nakamura, M. et al. (2019) ‘Anti-CRISPR-mediated control of gene editing and synthetic circuits in eukaryotic cells’, Nature Communications, 10(1), p. 194. doi: 10.1038/s41467-018-08158-x.
    7. Ahern, S. J., Das, M., Bhowmick, T. S., Young, R., & Gonzalez, C. F. (2014). Characterization of novel virulent broad-host-range phages of Xylella fastidiosa and Xanthomonas. Journal of bacteriology, 196(2), 459-471.
    8. Casjens, S. R., & Hendrix, R. W. (2015, May). Bacteriophage lambda: Early pioneer and still relevant. Virology, Vol. 479–480, pp. 310–330. https://doi.org/10.1016/j.virol.2015.02.010
    9. Farasat, I., Kushwaha, M., Collens, J., Easterbrook, M., Guido, M., & Salis, H. M. (2014). Efficient search, mapping, and optimization of multi‐protein genetic systems in diverse bacteria. Molecular systems biology, 10(6).
    10. Ng, C. Y., Farasat, I., Maranas, C. D., & Salis, H. M. (2015). Rational design of a synthetic Entner–Doudoroff pathway for improved and controllable NADPH regeneration. Metabolic engineering, 29, 86-96.
    11. Lederberg, E. M., & Lederberg, J. (1953). Genetic Studies of Lysogenicity in Escherichia Coli. Genetics, 38(1), 51–64. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/17247421
    12. Ptashne, M. (2004). A genetic switch: phage lambda revisited (3rd ed.). New York: Cold Spring Harbor Laboratory Press.
    13. Kutter, E., & Sulakvelidze, A. (2004). Bacteriophages (1st Editio; E. Kutter & A. Sulakvelidze, Eds.). https://doi.org/10.1201/9780203491751
    14. Kobiler, O., Rokney, A., Friedman, N., Court, D. L., Stavans, J., & Oppenheim, A. B. (2005). Quantitative kinetic analysis of the bacteriophage genetic network. Proceedings of the National Academy of Sciences, 102(12), 4470–4475. https://doi.org/10.1073/pnas.0500670102
    15. Leenay, R. T., Maksimchuk, K. R., Slotkowski, R. A., Agrawal, R. N., Gomaa, A. A., Briner, A. E., … Beisel, C. L. (2016). Identifying and Visualizing Functional PAM Diversity across CRISPR-Cas Systems. Molecular Cell, 62(1), 137–147. https://doi.org/10.1016/j.molcel.2016.02.031
    16. Larson, M. H., Gilbert, L. A., Wang, X., Lim, W. A., Weissman, J. S., & Qi, L. S. (2013). CRISPR interference (CRISPRi) for sequence-specific control of gene expression. Nature Protocols, 8(11), 2180–2196. https://doi.org/10.1038/nprot.2013.132
    17. Kincade, J. M., & DeHaseth, P. L. (1991). Bacteriophage lambda promoters pL and pR sequence determinants of in vivo activity and of sensitivity to the DNA gyrase inhibitor, coumermycin. Gene, 97(1), 7–12. https://doi.org/10.1016/0378-1119(91)90003-T
    18. Deveau, H., Barrangou, R., Garneau, J. E., Labonté, J., Fremaux, C., Boyaval, P., … Moineau, S. (2008). Phage response to CRISPR-encoded resistance in Streptococcus thermophilus. Journal of Bacteriology, 190(4), 1390–1400. https://doi.org/10.1128/JB.01412-07
    19. Oppenheim, A. B., Kobiler, O., Stavans, J., Court, D. L., & Adhya, S. (2005). Switches in Bacteriophage Lambda Development. Annual Review of Genetics, 39(1), 409–429. https://doi.org/10.1146/annurev.genet.39.073003.113656
    20. Gottesman, M. E., & Weisberg, R. A. (2004). Little Lambda, Who Made Thee? Microbiology and Molecular Biology Reviews, 68(4), 796–813. https://doi.org/10.1128/mmbr.68.4.796-813.2004
    21. Wilson, H. R., Kameyama, L., Zhou, J. G., Guarneros, G., & Court, D. L. (1997). Translational repression by a transcriptional elongation factor. Genes and Development, 11(17), 2204–2213. https://doi.org/10.1101/gad.11.17.2204