Fig. 1. Design-Build-Test-Learn Cycle for ADP1 Genetic Engineering.

Fig. 2. Acinetobacter baylyi ADP1's Natural Competence Allows for Uptake of DNA, such as PCR product or Golden Gate Assembly DNA. Adapted from [1].

Here, we describe the phases of our engineering process, using ADP1’s homologous recombination mechanism to our advantage (Figure 3). In each phase, we use standard molecular cloning techniques, including Polymerase Chain Reaction (PCR) and Golden Gate Assembly (GGA), to amplify and construct our DNA sequences. We demonstrate simple confirmation methods using the tdk/kan cassette with selective antibiotic agar plates for bacterial growth. Lastly, we explain strategies used to troubleshoot problems we ran into while following our Design-Build-Test-Learn cycles.

Phase 1: ampD Engineering

We chose to knock out ampD, an ADP1 gene that contributes to intrinsic β-lactam antibiotic resistance, to improve ADP1’s usability as a synthetic biology chassis. ampD is involved in breaking down peptidoglycan proteins [3]. Although we failed to knock out ampD, we applied what we learned to our future ADP1 genetic engineering phases.

Design

We designed primers to amplify ~1 kb homology 5’ and 3’ flanks upstream and downstream of the endogenous ADP1 target gene (Figure 4). Each set of primers contains a “Golden” primer that attaches BsaI and BsmBI restriction sites to the end of the primer, containing the correct GGA Type Overhang (BsaI) and 4 bp "rescue" complementary scar (BsmBI). See the Contribution Page for more details on how we utilize flexible GGA Type Overhangs to build our system. These amplified homology flanks, once assembled, are used to modify the ADP1 genome. In the first transformation step, the upstream and downstream flanks are ligated with the tdk/kan cassette using GGA via BsaI digestion. In the second transformation step, the upstream and downstream flanks are ligated to each other using GGA via BsmBI digestion.

Build

We synthesized homology flanks using PCR using our designed ampD primers and ADP1-ISx genomic DNA. We used gel electrophoresis to verify 1 kb upstream and downstream flank fragment sizes (Figure 5A, B) and the Zymo DNA Clean and Concentrate Kit to purify PCR products.

Next, we used GGA to ligate the upstream and downstream homology flanks with the tdk/kan cassette from the pBTK622 plasmid via BsaI digestion. Homology flanks were ligated together via BsmBI digestion to create our “rescue" cassette. We confirmed GGA ligation fragment sizes, the 3.7 kb target band consisting of a 1.7 kb tdk/kan fragment and 1 kb upstream and downstream flanks, with gel electrophoresis (Figure 5C).

Test

We added 15 μl of the first GGA reaction to a culture tube containing 500 μl of LB and 35 μl of overnight ADP1 culture. After overnight incubation (30°C, 200 rpm), we plated 100x diluted transformations onto LB-Kanamycin (Kan) plates to confirm the insertion of the tdk/kan cassette (Figure 6).

Figure 6 shows that the tdk/kan cassette transformation did not work, and repeated attempts continued to fail.

Learn

We were unable to transform the tdk/kan cassette into the ampD gene location. Repeated transformations on LB-Kan plates continued to fail. Looking back on our primer design, we found that ampD could form an operon with the essential gene murJ, and the tdk/kan cassette insertion could be cutting off the promoter for murJ.

In future primer designs, we checked nearby genes to ensure that knocking out our target gene did not affect ADP1-ISx viability. In Phase 2, we targeted pbpG and acrB, other genes that contribute to intrinsic β-lactam antibiotic resistance.

Furthermore, we found that using a gradient of annealing temperatures, ranging ± 5°C from the lowest Tm temperature in the set of primers (from NEB Tm calculator), helped improve PCR amplification and purification concentrations. Using a temperature gradient also helped us optimize annealing temperatures for each set of primers, given the intensity of the gel electrophoresis band.

Phase 2: pbpG and acrB Engineering

Next, we chose to knock out pbpG and acrB, ADP1 genes that also contribute to intrinsic β-lactam antibiotic resistance. pbpG encodes a penicillin-binding protein involved in peptidoglycan synthesis, and acrB encodes proteins involved in efflux pumps [3]. Creating these knockout ADP1 strains demonstrate that these ADP1 genetic engineering protocols are effective. Further data on the Results Page show that we successfully made ADP1 more susceptible to Carbenicillin.

Design

We designed primers to amplify ~1 kb homologous 5’ and 3’ flanks upstream and downstream of pbpG and acrB. These primers follow the same design as shown in Figure 4. Each set of primers contains a “Golden” primer that attaches BsaI and BsmBI restriction sites to the end of the primer, containing the correct GGA Type Overhang (BsaI) and 4 bp "rescue" scar (BsmBI). These amplified homology flanks, once assembled, are used to modify the ADP1 genome. In the first transformation step, the upstream and downstream flanks are ligated with the tdk/kan cassette using GGA via BsaI digestion. In the second transformation step, the upstream and downstream flanks are ligated to each other using GGA via BsmBI digestion.

Build

We synthesized homology flanks using PCR using our designed pbpG and acrB primers and ADP1-ISx genomic DNA. We used gel electrophoresis to verify ~1 kb upstream and downstream flank fragment sizes (Figure 7A) and the Zymo DNA Clean and Concentrate Kit to purify PCR products.

Next, we used GGA to ligate the upstream and downstream homology flanks with the tdk/kan cassette from the pBTK622 plasmid via BsaI digestion. Homology flanks were ligated together via BsmBI digestion to create our “rescue” cassette. We confirmed GGA ligation fragment sizes, the 3.7 kb target band consisting of a 1.7 kb tdk/kan fragment and 1 kb upstream and downstream flanks, with gel electrophoresis (Figure 7B).

Test

We use the tdk/kan cassette to select for successful ADP1 transformants. The kanR gene allows us to select on kanamycin (Kan), and the tdk gene allows us to select on azidothymidine (AZT).

We added 20 μl of the first GGA reaction to a culture tube containing 500 μl of LB and 35 μl of overnight ADP1 culture. After an overnight incubation (30°C, 200 rpm), we plated 100x diluted transformations onto LB-Kan plates to confirm the insertion of the tdk/kan cassette (Figure 8).

After inoculating a couple of colonies from the +DNA LB-Kan plate, we performed the same transformation process, adding 20 μl of the second GGA reaction to a culture tube containing 500 μl of LB and 35 μl of the overnight modified ADP1 culture, containing the new tdk/kan cassette. Next, we plate on LB-AZT instead of LB-Kan, to confirm the deletions of pbpG and acrB. This transformation inserted our “rescue” cassette, which contained the two synthesized pbpG or acrB upstream and downstream homology flanks ligated together. Plating on LB-AZT allows us to select for transformants that replaced the tdk/kan cassette with the "rescue" cassette since tdk makes ADP1 susceptible to AZT.

We further confirmed the deletions of pbpG and acrB from ADP1 via PCR by comparing ADP1 strains to the tdk/kan insertion strains and the gene deletion strains (Figure 10).

Learn

We decided to add the entire unpurified GGA reaction (20 μl) instead of saving 5 μl for gel electrophoresis in the transformation step. Confirmation of the GGA reactions via gel electrophoresis was tedious and time-consuming. We believed using the full mixture would help increase transformation frequency.

Phase 3: recJ and ACIAD2049 Engineering

The third phase shows how we knocked out recJ and ACIAD2049 in ADP1. recJ encodes a single-strand DNA exonuclease that degrades short cytoplasmic DNA [4]. Knocking out recJ could improve transformation frequency for smaller donor DNA fragments (< 500 bp) [5]. Knocking out ACIAD2049, a nonessential ADP1 gene, allows us to integrate other DNA sequences of interest into ADP1.

Design

We designed primers to amplify ~1 kb homologous 5’ and 3’ flanks upstream and downstream of recJ and ACIAD2049. These primers follow the same design as shown in Figure 4. Each set of primers contains a “Golden” primer that attaches BsaI and BsmBI restriction sites to the end of the primer, containing the correct GGA Type Overhang (BsaI) and 4 bp "rescue" scar (BsmBI). These amplified homology flanks, once assembled, are used to modify the ADP1 genome. In the first transformation step, the upstream and downstream flanks are ligated with the tdk/kan cassette using GGA via BsaI digestion. In the second transformation step, the upstream and downstream flanks are ligated to each other using GGA via BsmBI digestion.

Build

We synthesized homology flanks using PCR using our designed recJ and ACIAD2049 primers and ADP1-ISx genomic DNA. We used gel electrophoresis to verify ~1 kb upstream and downstream flank fragment sizes (Figure 11) and the Zymo DNA Clean and Concentrate Kit to purify PCR products.

Next, we used GGA to ligate the upstream and downstream homology flanks with the tdk/kan cassette from the pBTK622 plasmid via BsaI digestion. We ligated the homology flanks together via BsmBI digestion to create our “rescue” cassette.

Test

We use the tdk/kan cassette to select for successful ADP1 transformants. The kanR gene allows us to select on kanamycin (Kan), and the tdk gene allows us to select on azidothymidine (AZT).

We added 20 μl of the first GGA reaction to a culture tube containing 500 μl of LB and 35 μl of overnight ADP1 culture. After overnight incubation (30°C, 200 rpm), we plated 100x diluted transformations onto LB-Kan plates to confirm the insertion of the tdk/kan cassette (Figure 12).

After inoculating a couple of colonies from the +DNA LB-Kan plate, we performed the same transformation process, adding 20 μl of the second GGA reaction to a culture tube containing 500 μl of LB and 35 μl of the overnight modified ADP1 culture, containing the new tdk/kan cassette. Next, we plate on LB-AZT instead of LB-Kan, to confirm the deletions of recJ (Figure 13) and ACIAD2049. This transformation inserted our “rescue” cassette, which contained the synthesized recJ upstream and downstream homology flanks ligated together. Plating on LB-AZT allows us to select for transformants that replaced the tdk/kan cassette with the "rescue" cassette since tdk makes ADP1 susceptible to AZT.

Learn

Using this two-step ADP1 genetic engineering protocol, we were able to reliably modify ADP1’s genome. Selecting on LB-Kan and LB-AZT plates demonstrates how we use the tdk/kan cassette to confirm our transformations and genome modifications. We learned that we could easily design different “rescue” DNA sequences to replace the tdk/kan cassette, inserted in the place of each endogenous ADP1 gene. We followed these engineering phases to construct our ADP1 biosensors for detecting Antibiotic Resistance Genes and White Nose Syndrome.

References