iGEM is an engineering competition about developing functional biological building blocks to create a database for the general public. The process of creating such building blocks is described by the iGEM engineering cycle. From our point of view every individual team must create at some point their own engineering cycle which is fitting best to one's project. One team maybe needs more learning sessions whereas other teams need several rounds of testing. We made the experience that our project required a complete new perspective which helped us to come closer to the final results. Our engineering cycle is demonstrated in figure 1 where we developed from a complete cycle to a design and testing session in between and then finally design, build and test. Maybe this can be also a solution for other iGEM teams to develop their own engineering cycle to reach their achievements.
Figure 1: Our personalized circle about the engineering success for the natural transformation in A. baylyi ADP1. Created with BioRender.com
Our overall aim was to establish a protocol for natural transformation and build up a workflow for a stable cloning in Acinetobacter baylyi ADP1. In the following part our workflow is presented in the way it is shown in the engineering cycle (see Fig. 1). All data that is not included on this page can be seen on the cloning with natural transformation chapter.
Design overlap extension PCR
We contacted the authors of the most promising papers to ask for advice on our project. We expected to require linear plasmids containing the full information more than once called multicopy plasmids to successfully apply natural transformation. This was the output from the expert interview with Xinglin Jiang in February. His original idea was to design a plasmid and use rolling circle amplification with the phi29 polymerase. We decided against this technique because we found a well described protocol to create multicopy and linear plasmids by overlap extension PCR [1]. For this we designed the complex overlapping primers for the Lvl0_8_Amp/ColE1 plasmid.
Build multicopy and linear plasmid
The first step was to cut our plasmid with the restriction enzyme BsaI to receive the insert and vector as linear products. This step was successfully performed and we ran PCR with this purified mix of both insert and vector separately with both primer pairs. After that we controlled the PCR products in the agarose gel and they appeared as expected. But when controlling the overhangs by sequencing, it turned out that the overhangs that should be 20 bp long, were missing 8 bp on both ends. In the second PCR without additional primers these overhangs should function as primers. But with less than 20 bp they were too small to use.
Test multicopy and linear plasmids and natural transformation
We controlled the second PCR but there were no PCR products visible in the agarose gel. We changed the polymerase as a small troubleshooting but the results were the same. Additionally, we already tried the natural transformation according to the protocol in [1] with three different types of plasmids: circular (Lvl0_8_Amp/ColE1 and pET15b-mCherry), linear (Lvl0_8_Amp/ColE1 vector) and our overlap extension PCR product after the second PCR. None of these three types or transformant DNA was working for natural transformation.
Learn about natural transformation
We reviewed the experiments and as already said we changed the polymerase from taq to pfu but that did not work out. Whereas for the natural transformation the ColE1 origin of replication (ori) can not be used for A. baylyi ADP1, instead we have to use pBAV1k ori BBa_K3963004 [2]. The ColE1 can be used as a negative control which was congruent to our results.
Design a new protocol for natural transformation
We checked again the literature and decided due to time limiting problems to not clone the new ori pBAV1k in one of our backbones but order the two plasmids pBWB162 (Plasmid #140634) and lacI-PT5-gusA-pBAV1k (Plasmid #30501) with the suitable ori from Addgene. We created a new registry page for the pBWB162 plasmid (BBa_K3963000).
Test for successful natural transformation
We skipped the building part and extracted the new plasmids from purchased E. coli DH5ɑ strains to use them for A. baylyi ADP1 natural transformation. Now we received successful clones on our selective agar plates for both plasmids. Also the mCherry expression was visibly detectable for the successful clones with the pBWB162 plasmid. We compared the influence of the DNA amount on the transformation efficiency and compared two natural transformation protocols (data shown on the registry). Now we would like to clone the AvPAL gene of interest in the natural transformation plasmid.
Design primers with new restriction side overhangs
We tried to clone AvPAL into the natural transformation plasmid but first we wanted to control whether our technique to use primer with new restriction side overhangs is working. We designed to primer pairs one for EGFP to clone into pUC19 from pET15b and the other one is for the AvPAL to clone from the pET15b into the pBWB162.
Table 1: Primer for new restriction sites at the end (ordered by IDT)
Primer
Sequence
EGFP Forw
AAAAAAGGATCCTATGAAGACCCTGTCCCAGG
EGFP Rev
AAAAACATATGGTTAGTGATGGTGGTGGTGAT
AvPAL Forw
AAAACCATGGTGATGAAGACCCTGTCCCAGGC
AvPAL Rev
TTTTCGATCGTTAGTGATGGTGGTGGTGAT
Build new plasmid pUC19-EGFP and pBWB162-AvPAL
We cut the EGFP out of the pET15b backbone with a T7 promoter and cloned it into the pUC19 backbone by the restriction enzymes BamHI and NdeI. Normally, also pUC19 has these restriction enzymes in the multiple cloning site (MCS) but the other way around. With the successful cloning the EGFP should also be expressed in E. coli DH5ɑ because of the lac operon without T7 promoter. The AvPAL is cloned by the restriction enzymes NcoI and PvuI into the pBWB162 backbone by cutting out mCherry.
Test EGFP expression and AvPAL appearance
The EGFP expression could be easily detected by the control over UV-light table (see Fig. 2 A) whereas the AvPAL positive ligation should be visible by the white colonies in comparison against the red colonies. We detected a positive band in the ligation gel control (data not shown) and a lot of white and pink colonies on the ligation agar plates. We picked in total 27 white colonies but detected no correct plasmid with the band size of 6028 bp. However, the colony PCR showed positive results in several picked colonies. We transformed some plasmids again into E. coli DH5ɑ and sequenced them (see Fig. 2 C). All four colonies also showed positive ligation with AvPAL as an insert in the pBWB162 vector backbone. We called the plasmid pBAV1k-lacI-Trc-AvPAL (BBa_K3963002).
Additionally, we ordered β-agarase (BBa_K2094002) without PvuI and NcoI restriction sites inside of the enzyme sequence instead with PvuI and NcoI restriction sides at the ends. We ligated the β-agarase into the pBWB162 vector backbone and transformed them into the E. coli DH5ɑ (see Fig. 2 B). We called the plasmid pBAV1k-lacI-Trc-beta-agarase YM01-3 (BBa_K3963001). We detected pit formation for E. coli DH5ɑ after heatshock and A. baylyi ADP1 after natural transformation (see Fig. 2 E).
Figure 2: Engineering success results. (A) shows the successful test to use overhang restriction side primers to bring an enzyme in a new vector backbone. On the agar plate EGFP was cloned into the pUC19 vector backbone. The photo was taken with a smartphone therefore the quality and section is not optimal. (B) shows the successful ligated β-agarase into the pBWB162 vector backbone transformed into E. coli DH5ɑ. (C) represent the sequencing results from the AvPAL ligated enzyme into the pBWB162 vector backbone. The plasmids were prepared from transformed E. coli DH5ɑ. (D) shows the unsuccessful natural transformation with the pBAV1k-lacI-Trc-AvPAL plasmid. (E) shows successful natural transformation with A. baylyi ADP1 with the pBAV1k-lacI-Trc-beta-agarase YM01-3 plasmid.
Learn about new restriction side overhangs and review results
If we had had more time we would have repeated the natural transformation with the pBAV1k-lacI-Trc-AvPAL plasmid. We assume that the ligation of EGFP in pUC19 and AvPAL in the pBAV1k-lacI-Trc vector backbone (BBa_K3963003) works (see Fig. 2 A and 2 B). Only the acceptance of plasmids for the natural transformation machinery is not clear (see Fig. 2 D). Our designed overhang primers can be used to change the restriction sides except if the restriction side is included in the enzyme’s sequence.
Furthermore, the cloned pBAV1k-lacI-Trc-beta-agarase YM01-3 successfully transformed into A. baylyi show pit forming colonies (see Fig. 2 E).
Test pBAV1k-lacI-Trc-beta-agarase YM01-3 activity in vivo
We repeated the β-agarase DNS-method (BBa_K2094002) with the transformed A. baylyi colonies that show visually agarose degradation. We measured the reducing sugars and compared them to the standard sugar curve (see Fig. 3 B).
Figure 3: β-agarase "in vivo" DNS-method in A. baylyi. (A) shows the reaction tube with the DNS reaction solution. The graph in (B) represents the measured solutions at 540 nm regarding the standard curve. We compared negative control A. baylyi without a β-agarase in comparison to cloned A. baylyi with the pBAV1k-lacI-Trc-beta-agarase YM01-3 plasmid.
In figure 3 B we detected the efficiency of β-agarase with the DNS-method. The results show a difference between negative control and A. baylyi with the pBAV1k-lacI-Trc-beta-agarase YM01-3 plasmid. The difference is not significant which does not indicate β-agarase activity. We assume that the β-agarase is active due to the pit formation of the colonies (see Fig. 2 E). The protocol was slightly changed due to time limitations we picked colonies directly from the plate for the in vivo assay and shaked them at 75 rpm which might degrade growth.
Summary of our results
With our engineering success we gained a new functional plasmid that was able to be cloned by natural transformation into our model bacteria A. baylyi. Moreover, A. baylyi was able to express the new protein β-agarase. We also wanted to clone AvPAL into A. baylyi as well which did not work due to unknown problems. The cloning with overlap restriction site primers was working well for EGFP in the pUC19 backbone and probably as well for the AvPAL into the pBAV1k-lacI-Trc backbone but the natural transformation machinery of A. baylyi was interrupted somehow.
To sum up, we were able to establish a stable natural transformation protocol which was adaptable for newly designed plasmid (BBa_K2094002) containing the functional backbone (BBa_K2094003) of the pBWB162 plasmid (see Fig. 4).
Figure 4: Final protocol for natural transformation of A. baylyi. We combined the original natural transformation protocol with the design of the suitable plasmid which contains the pBAV1k origin of replication, an antibiotic resistance and the gene of interest in the cloning restriction side between NcoI and PvuI. The expression of the gene of interest is controlled by a trc promoter and can be induced through IPTG. Created with BioRender.com
References
[1] You, C., Zhang, X. Z., & Zhang, Y. H. (2012). Simple cloning via direct transformation of PCR product (DNA Multimer) to Escherichia coli and Bacillus subtilis. Applied and environmental microbiology, 78(5), 1593–1595. https://doi.org/10.1128/AEM.07105-11
