The designed plasmids were as follows:

We ran these 4 PCR reactions as follows using the NEB Q5 PCR protocol:

A 0.8% agarose gel was used for the extraction as this offered good separation around 1kb and 5kb. This gel was run and the bands extracted according to our QIAquick Gel Extraction Protocol using NEB purple loading dye and 2-log purple ladder and QIAquick extraction kit. Gel obtained:

As we later found out these NanoDrop readings are false since the QG buffer in the QIAGEN gel extraction kit interferes with the UV/Vis readings. Following extraction of our PCR products from the gel we used the NEB Gibson Assembly protocol to run an 8hr reaction over night. We ran an ‘A’ reaction which will insert the KanR gene into the pME6010 plasmid with the native promoter and a ‘B’ reaction that will insert the KanR gene with a pamp promoter and optimised RBS.

To run equimolar amounts of insert to vector the following table was calculated:

These gibson assemblies were then transformed into E.coli cells using electroporation with the voltage at 1.8kV and the exposure time ranging from 4.8ms to 5.2ms. The cells were then recovered at 37ºC for 1.5 hrs before 100 µl was plated onto KanR plates. These were left overnight at 37ºC.

Unfortunately, there were no colonies grown and as we have no product from our gibson assemblies remaining we will have to repeat the process from Week 1 in order to trouble-shoot the method and get the desired plasmid. It is possible that we accidentally swapped the primers or PCR products around at some point and thus we had 1A with 2B and 1B with 2A in the Gibson assembly. These combinations are not complementary and thus the Gibson would fail meaning that none of our bacteria received intact Kanamycin resistance in the transformation. However there are many other reasons why the process may have failed therefore we went through the entire method trying to cover all instances where we may have made an error.

First we confirmed that the primer sequences on the delivery tubes matched what we had ordered. We ran the PCR as before; we re-calculated the annealing temperatures and extension times with the same result as last week.

In week one, we did the DpnI digest after the Gibson assembly and not directly after the PCR. This time we did the digest immediately after the PCR to ensure that all template DNA was destroyed before running on the gel, ensuring cleaner bands on our gel and less contamination from non-PCR products.

We loaded 20 μl of each PCR product onto a 0.8% agarose gel as before. The resulting bands were visualized after staining as shown below:

The bands are not as bright as in the first attempt probably due to less time spent in the ethidium bromide staining tank. The bands correspond with the correct sized fragments as before. This suggests there was no error in the PCR reaction with the exception of again mixing up the primers (the PCR would be successful with any combination of the primers for a 1 and 2 reaction). The bands are cleaner allowing us to try an alternative to the gel extraction in case that caused our assembly and transformation to fail.

Next we excised and extracted our bands from the gel with the same QIAquick Gel Extraction Protocol protocol as before but with two changes. First we allowed additional drying time after the step involving ethanol supplemented buffer in case ethanol contamination inhibited the enzymes used in the Gibson assembly later in the process. We also replaced the 20 μl EB buffer with 20μl of MilliQ water. We did our second elution with the kit EB buffer. We then measured the concentration of DNA in each eluted sample using the nanodrop. This still gives us a reasonable idea (despite the interference from the remaining QG buffer residue) as to the amount of DNA in each sample from which to calculate our dilutions for the Gibson assembly reaction.

We used the remaining 10μl of our PCR products to do a Promega PCR clean-up and eluted using water. We gathered the following Nanodrop data:

Next we ran the following Gibson assembly reactions:

Next we transformed the Gibson assembly products into chemically competent E.coli cells. We decided to run a positive control for each kind of cell used using the pUC19 plasmid provided with the cells. Unfortunately we were limited with the amount of cells available and had to use two different cell strains; this resulted in some variation in the length of the first incubation with the DNA on ice as we began the transformation in the morning but then decided to pause the process so that the cells would be plated out later in the day in order to not have too much growth overnight. The NEB 5-alpha cells were also on ice slightly longer before the DNA was added.

The cells were plated out after 1.5 hour incubation. It was noted that one of the sample appeared to have leaked from the Eppendorf into the petri dish containing all the samples in the incubator. We think it was the 5-alpha control as this was the only sample that appeared to have lost volume. Samples 1-8 were plated on Kanamycin plates (i) that we made up during the final incubation period. The remaining sample was re-suspended in ~100μl of SOC to create a concentrated solution of cells that were also plated out on kanamycin plates (ii). The two controls were spread on Ampicillin plates made up by Nick. All plates were placed in the 37⁰c incubator overnight. The results in the morning were as follows:

Apart from the concentrated sample 1ii, none of the samples prepared using the gel extraction protocol grew. Sample 1ii also had limited growth compared to the other plates with colonies:

Samples 1i and 5i differed only in their clean-up method; 5i has lots of growth and 1i has none. This comparison holds true for Samples 2+6, 3+7 and 4+8 with the same result:

This suggests that we should use the Promega PCR clean-up kit and not the QIAquick Gel extraction. We will still run our samples on a gel to check the quality of our PCR reactions. We may investigate the Promega Gel extraction kit as this will enable us to pick our desired band out of a PCR that has also generated non-specific fragments which we cannot do with the Promega PCR clean-up protocol.

The samples that were treated with the master-mix and ligase supplied by Ciaran had more colonies than those that were treated with the NEB supplied master-mix as shown by the comparison between samples 5i and 7i:

This suggests that in future we should use the master-mix supplied by Ciaran.

All the controls grew showing that the lack of growth of samples 1- 4 was not due to inefficient transformation:

There was no observable difference between sample 8 and sample 7 which were treated the same but which used different chemically competent cells. Thus we can continue to use either the NEB 5-alpha or DH5-alpha chemically competent cells.

There was no observable difference between the samples that correspond to A Gibson assembly reactions (template for PCR primers and therefore fragment = pCM66) and those that correspond to B Gibson assembly reactions (PCR primers and therefore fragment include Pamp).

We picked colonies from the successful plates to grow up cultures overnight:

We treated the resulting Gibson assembly products with a PCR clean-up (labelled resulting samples 6i and 8i respectively) and gel extraction (labelled resulting samples 6ii and 8ii respectively). We then transformed these samples into chemically competent NEB alpha-4 cells. Having allowed these to grow up overnight on kanamycin plates we had colonies on all four plates. This confirms that the transformation worked but does not confirm that our Gibson assembly reaction was successful. We will test this next week so the plates have been placed in the cold room to prevent too much growth.

We grew up overnight liquid cultures of our Gibson assemblies. We extracted the plasmids from these samples (in order to get as much DNA as possible we used all the cells available resulting in several centrifugation steps to achieve a bacterial pellet before lysis). We eluted the each sample in only 30μl of elution buffer to concentrate the sample and the following NanoDrop data was acquired:

To test this theory we did a double restriction enzyme digest using HindIII and PstI using the following amounts and incubating at 37⁰c for an hour:

This further confirms our conclusion that our Gibson assembly reaction has failed and the vector backbone re-ligated without the insertion of the fragments. To ensure that none of our colonies had the complete construct we selected 12 more colonies from plates 6i and 6ii and grew them up in liquid culture with kanamycin overnight (labelled 1B-12B). We then extracted the plasmids from the cultures and repeated the same double restriction enzyme digest with HindIII and PstI. We however found the same result as before with only a single band running at ~5.5kb:

Sequencing data for samples 13-24 which potentially contain the pSRK-pdcmA-sfGFP construct came back as shown below:

Thus we will have to repeat the Gibson assembly for both our constructs. To this end we have used PCR to amplify the pME6010 Kan and pSRK Gm vector backbones, essentially repeating PCR reactions #6 and #8:

We will run a 2 hour DpnI digest on the resulting PCR products to ensure that there is none of the original template left. We will also run them very slowly on a gel to ensure maximum separation before excising the fragments and extracting them from the gel.

In addition to the above we made stocks of chemically competent MG155 E.coli cells to transform our plasmids into when we have succeeded in constructing them. We will make competent Pseudomonas stocks when we have our constructs completed as these need to be made only shortly before the actual transformation.

To confirm that none of the colonies on our plates had the correct construct we did a rough colony PCR on 48 colonies. We did not use specially designed primers but they should have produced a defined PCR product if our inserts were in any of the colonies. None of the colonies showed a defined product as shown below:

PCR reactions #14 and #15 were prepared as follows: TABLE OF PCR REACTION CONDITIONS We ran the PCR products for approximately 90 minutes at 70V on a 0.7% agarose gel, shown below:

The only band in lane 3 corresponds to the size of the pME6010KanR backbone at ~5.5kb. The band of size ~4.5kb in lane 5 corresponds to the pSRK Gm backbone. Due to the possibility that the indistinct separation of the large band in lane 3 was in fact two bands, when excising the fragment from the gel this band was cut in half. Thus three gel extractions were undertaken; #14TOP, #14BOTTOM and #15. The samples were eluted in 20μl of water. Following gel extraction the concentrations of DNA were determined using the NanoDrop:

As the concentrations form the second elution are so much higher, we will use those samples in our Gibson assembly reactions.

We set up the following Gibson reactions. Gibson #2, #4 and #5 have equimolar amounts of vector to inserts. Reactions #1 and #3 are not equimolar. All these reactions were catalysed by the Mastermix and ligase provided by Ciaran’s lab.

The products of these Gibson reactions were then transformed into various types of cells (as we had a limited amount of the NEB alpha-5 cells). This will allow us to compare the transformation efficiency of the various types of cells available to us.

The results of the transformations are shown below. All the transformations grown on Kanamycin plates but transformations h., h.ii. i., i.ii, l., m. and n. should have been on gentamycin plates; thus these transformations were repeated as shown in the second table below.

Plated out new transformations:

Key: See Gibson reactions for construct components NEB = NEB 5-alpha cells CK TSB = DH5-α provided by Ciaran Glen = DH5-α cells made competent by Glen

This suggests we have four colonies that potentially contain our pME6010 KanR construct. This is supported by the fact that the positive control for transformation c.ii has growth and the negative control, whilst it potentially has a single colony has less growth than the c.ii plate and this is also possibly contamination. There is also potentially another colony on plate b.ii. The colonies on plate h.ii are most likely contamination as none of our construct should have grown on the kanamycin – however as there were quite a few colonies on plate m. which also shouldn’t have had any growth this potential resistance may be something to bear in mind.

We made up the following overnight liquid cultures (in 5ml LB):

Cultures c.ii. 1-4 were mini-prepped and eluted in 30μl of EB buffer. The following concentrations were determined by Nanodrop:

The growth on plates p.- r. suggests that the small colonies might contain our pSRK Gm construct . After 16 hours in the incubator the colonies were very small, so the plate p.ii. was left on the bench and liquid cultures made a further 8 hours later. These liquid cultures had no growth the next morning.

We have left p.ii (after a short while in the cold room) and the liquid cultures in incubator to see if there is any growth over a longer period of time.

Plates e. f. and g. provide a comparison of transformation efficiency between the three types of competent cells. Plates e. and g. had full lawns of colonies. Plate f. had fewer colonies but still quite a lot. Thus Glen’s cells seem to be as good as the NEB cells. Ciaran’s do not have as high transformation efficiency.