Team:Edinburgh/Lab Notebook

Note: Protocols were generally similar to those found on Open Wetware's [http://www.openwetware.org/wiki/French_Lab pages] dedicated to Chris French's Lab.

June 1: Began work to add the Plac-lacZα sequence to the pSB1C3 vector. PCR was used to introduce the Bgl2 restriction site (so the final thing becomes E-X-B-Plac-lacZα-S-P). This gives us a vector that we like. For the principles behind the use of lacZα see [http://utminers.utep.edu/rwebb/html/alpha_complementation.html here]. It will enable us to determine whether inserts are present using blue/white screening. In the future, vectors in which the desired part successfully replaces lacZα will be white.

Having a Bgl2 site in the vector at the end of the BioBrick prefix is useful as it means any DNA sequence can easily have the sites needed for incorporation into the vector added (by PCR) at both ends, whereas in the standard RFC10 system the two most obvious sites (XbaI and SpeI, which come immediately before and after the BioBrick itself) cannot actually be used this way, because they generate compatible sticky ends, meaning the product would circularise.

June 2: PCR seemed successful (though 2 bands were seen on the gel, 1 was the right size). We already have pSB1C3-RFP (Red Fluorescent Protein) so we cut out the RFP (using restriction enzymes XbaI and PstI) and inserted the new sequence.

June 3: After transformation of competent cells we plated onto an [http://en.wikipedia.org/wiki/X-gal Xgal]- and [http://en.wikipedia.org/wiki/IPTG IPTG]- containing media.

At this point we also prepared about 15 plates of agar, with chloramphenicol (which the pSB1C3 vector provides resistance to) as well as Xgal and IPTG for later blue/white screening.

June 6: Colonies that have successfully lost RFP and gained lacZα should be blue (on this media). We found several such colonies (though many others were red and some were white). Six blue colonies were selected, replated and grown for 6 hours on the media mentioned above. Some cells were then transferred to bottles of media for further growth.

June 7: Plasmid DNA from the bottled cultures was purified (by the "miniprep" protocol). We digested it with EcoRI to linearise it, and then ran it on a gel. All six cultures showed a band of the correct size (about 2600 b.p.) but we will send some DNA for sequencing.

June 8: Clones 1 and 2 of pSB1C3-lacZα were selected for sequencing. Four tubes were sent to the university's sequencing service:

1. Clone 1, Forward
2. Clone 1, Reverse
3. Clone 2, Forward
4. Clone 2, Reverse

We expect results on the 10th (Friday).

We also began PCR to add the Bgl2 restriction site to some RFP and YFP (Yellow Fluorescent Protein) coding sequences (with no promoters). Bands on a gel showed that PCR apparently worked for the YFP but not the RFP. Nevertheless, we purified both and placed them in the pink box in the fridge. For the record, the YFP source is [http://partsregistry.org/wiki/index.php?title=Part:BBa_E0030 BBa_E0030].

June 9: We started to insert YFP into our new vector. We digested both pSB1C3-lacZα and our YFP product with Bgl2 and SpeI. We then started a ligation reaction and left it overnight. There was some concern about contamination as the reaction mixture picked up a yellowish colour, possibly from dirty (unused but old) pipette tips.

If successful, some plasmids should become: E-X-B-YFP-S-P

Since there's no promoter and no lacZ gene, these should give white colonies on Xgal/IPTG media. We expect to do the transformations and plating tomorrow.

June 10: We transformed and plated our pSB1C3 + YFP (+ PlaclacZα, which we are trying to excise) ligation onto Xgal/IPTG plates. There are 4 plates; we decided to investigate whether the heat shock part of the transformation protocol has any effect, thus 2 were not heat shocked.

We also got back sequences of our vector (see June 8). Eugene says the sequence from clone 2 appears better than that of clone 1. [http://www.geospiza.com/Products/finchtv.shtml FinchTV] is freely available software we can use. Allan has added his best guess for the exact sequence of clone 2 to the sequences page (the chromatogram is a mess in a few places, but those bits can be filled in from the expected sequence).

We should probably use our pSB1C3-Plac-lacZα clone 2 as our vector of choice in the future. (16 June: the others were discarded)

June 13: Our transformed plates showed a few colonies which were white, and yellow under the blue light. Some others were white and not yellow under the light, and most were blue. We are working on the assumption that the yellow ones are having YFP transcribed and translated even without a promoter or obvious RBS in the biobrick. Four colonies have been streaked onto a new Cml/IPTG/Xgal plate.

However, Chris mentioned the possibility that the original PCR template plasmid could have made it all the way through the process and be causing this. But: did the original plasmid confer Cml resistance? If not, this isn't something to worry about. The current iGEM distribution has BBa_E0030 on pSB1AK3. Was this the source?? Need to check this. [Update: no, the source was also on pSB1C3. But because it was quite large, a gel of a miniprep will show us if we have the correct thing.]

June 14: We started bottled cultures of our pSB1C3-YFP for minipreps tomorrow.

We also did a PCR of a whole pSB1C3 backbone. This is mostly for Chris's BioSandwich idea. It looked OK on a gel.

[Update: the resulting sequence is shown here. By design, there is a bunch of stuff between the BglII site and the BioBrick suffix.]

June 15: Lee and Sylvia made 4 minipreps of (what we hope is) our pSB1C3-YFP plasmid.

They also started on Chris's BioSandwich test, doing a digest with:

• lacZα - digest with Bgl2/SpeI
• YFP - digest with Bgl2/SpeI
• Vector - digest with Bgl2/XbaI

Each was purified on 5 uL glass beads and eluted to 10 uL elution buffer.

Afterwards, these three fragments were ligated to their respective spacers, and left overnight at 16 C.

Meanwhile, Allan tested the size of the (putative) pSB1C3-YFP by cutting it with EcoRI and running it on a gel. The size should be about 2700? It seemed about right or maybe slightly larger. Clone 1 was barely visible, but clones 2, 3, or 4 seem OK.

June 16: Allan sent clones 2 and 3 of the putative pSB1C3-YFP off to be sequenced. Results can be expected on Monday 20th. Note that the labels given to the sequencing people do not match our own internal names. Specifically:

• "1F" and "1R" are our clone 2
• "2F" and "2R" are our clone 3

Lee and Sylvia purified the ligations from yesterday.

We then tried two ways of joining everything up (see BioSandwich page for more).

1. Ligation independent cloning
2. Overlap Extension PCR

The PCR failed; Chris later discovered we used the wrong primers. The LIC transformants were plated and left incubating overnight.

June 17: A few colonies were seen on the LIC plates, but all were white [Update: or not, see June 20].

Sylvia and Lee started on a fresh attempt of the OEPCR [Update: it didn't work].

June 20: Sequencing results came back from our putative pSB1C3-YFP. The sequences show that the fluorescent colonies we made in fact have part [http://partsregistry.org/wiki/index.php?title=Part:BBa_K216011 K216011], which includes a promoter. It is thus not what we were trying to amplify. Rather, it's the original PCR template that made it all the way through the process, from PCR to transformations.

It is likely that colonies on the plate that are white but don't fluoresce are what we actually wanted.

Over the weekend, Chris discovered that there was in fact a colony from the Ligation Independent Cloning attempt that was blue and fluoresced yellow. This is being miniprepped.

Some of the LacZα cloning vector (clone 2) is being grown up in LB + cml in the shaking incubator (the rightmost one). This will be maxiprepped tomorrow.

June 21: Allan did the maxiprep of pSB1C3-LacZα. This seemed to go well and the product has been placed in the box. It has not yet been tested on a gel however [Update: See June 27].

Lee and Sylvia did a miniprep of one blue, three colourless, and two yellow colonies from the BioSandwich Ligation Independent Cloning. After digesting to linearise, a band for the blue colony on a gel was very faint, almost invisible.

The whole LIC transformation process was repeated today.

June 22: Lee and Sylvia sent some minipreps from the LIC attempt to be sequenced.

June 23: Sylvia and Lee made some competent E. coli cells since we were running low.

June 24: The sequencing results came back for the Ligation Independent Cloning attempt of the BioSandwich protocol. One colony has successfully incorporated LacZ and YFP; also spacers are seen; all indications are it worked perfectly.

June 27: Worst gel ever. This has been "enchanced" in GIMP.

June 27: Allan did a digest (SpeI) on the LacZα maxiprep and ran it on a gel. A band of the correct size was sort of visible if you used your imagination, but it was very faint. The maxiprep DNA may be quite dilute. [Update: Chris later said he successfully used this as a vector.]

June 29: At Chris's suggestion, Allan designed primers for E. coli malS and bglX genes.

July 3: Chris ordered primers for the malS and bglX genes. They should arrive on Wednesday (6th). Phage [http://www.neb.com/nebecomm/products/productn4018.asp M13mp18] has also been ordered.

July 4: We could do with a source for Ice Nucleation Protein. Gary Loake in Plant Science informs us they have Pseudomonas syringae pv. tomato DC3000 and Pseudomonas syringae pv. phaseolicola. Both these strains have problems. DC3000 seems to just lack the ice nucleation protein gene. Phaseolicola has a slightly different gene, inaZ ([http://www.genome.jp/dbget-bin/www_bget?refseq:NC_005773 see here]) and the published sequence has a transposable element in it.

The gene that seems to be most commonly used (inaK) was first gotten from the Korean Collection for Type Cultures strain 1832.

July 6: PCR primers for malS and bglX arrived. Eugene showed us how to make stock solutions of PCR primers, and then how to make working solutions.

July 7: Allan has attempted to design BioSandwich-compatible primers for many of the genes involved in the project. See the Assembly page.

Lee and Sylvia performed PCR to try and isolate malS and bglX genes from E. coli. The desired products are:

1. malS with RBS, start and stop codons; normal RFC10 part.
2. malS from start codon, no stop codon, intended for fusions.

1. bglX with RBS, start and stop codons; normal RFC10 part.
2. bglX from start codon, no stop codon, intended for fusions.

According to the gel, the first three of these worked successfully (though bglX #1 is faint). The fourth will be repeated later, maybe the third also.

July 8: Allan and Yassen continued searching for genes and designing primers.

Sylvia and Lee redid PCR on the bglX gene (which seems to have worked this time, due to a lower annealing temperature of 55 C), as well as starting the process of inserting malS into our vector, starting with a purification of the malS PCR product.

July 11: Lee and Sylvia started the procedure for digesting and ligating malS and bglX into pSB1C3.

July 12: Lee transformed malS and bglX into cells on chloramphenicol.

We have also acquired the pG8SAET vector for display of proteins on pVIII; this might be useful as a fallback system. Lee has transformed this into cells on ampicillin.

July 13: The pG8SAET transformation has worked and the plate has been placed in the cold room. Any colony on the plate should have the plasmid.

New cml/Xgal/IPTG plates have been made and put in the cold room. We also seem to have acquired a bunch of Ampicillin plates.

The two malS and the two bglX transformations seem to have worked (all gave a mix of white and blue colonies). Four white colonies from each have been streaked out onto plates and placed in the 37 C incubator.

Overnight cultures of 2 colonies from each transformation have been started (8 colonies total). The source plates are still in the incubator and must be removed tomorrow...

FedEx sent Allan a tracking thingy for BBa_K265008 - it seems to be getting treated as high priority and is supposed to arrive by July 15, 2011, 6:00 PM. It should be coming on plasmid pSB1AK3, so when it arrives it needs to be plated onto ampicillin.

July 14: Lee did minipreps of the eight colonies. Afterwards, samples were cut with EcoRI and run on a gel:

1. Ladder
2. malS f1/r1 colony 1
3. malS f2/r2 colony 1
4. bglX f1/r1 colony 1
5. bglX f2/r2 colony 1
6. malS f1/r1 colony 2
7. malS f2/r2 colony 2
8. Ladder

Allan prepared sequencing reactions (minus primers) for all the above "colony 1" minipreps. Depending on Lee's gel results stuff will be sequenced...

July 15: Eugene has seen the gel and says it looks OK, so Allan sent forward and reverse sequencing reactions off for all the "colony 1" samples.

In other news, the INP part <partinfo>BBa_K265008</partinfo> arrived from Boston; some bacteria were streaked from the agar onto an Amp50/Kan10 plate that Eugene happened to have; it was then placed in the incubator.

July 18: Over the weekend Eugene has grown up some <partinfo>BBa_K265008</partinfo> in liquid culture and replated onto Ampicillin. Chris wants to miniprep this and get it sequenced. Eugene started overnight cultures.

Eugene informs us there's a failure in one of the University's 3730 sequencers, which will likely delay sequencing of the samples sent on the 15th.

July 19: Sequencing results came in on time. All four sequences look good. So all our "colony number 1" minipreps for bglX and malS are OK.

Eugene has miniprepped the INP (<partinfo>BBa_K265008</partinfo>) part, still in vector pSB1AK3. He digested two different minipreps with EcoRI and ran both on a gel. They gave identical results, though the bands look too high up. We expect about 4.1 kb (part + plasmid) yet the bands suggest over 5 kb. We'll sequence it anyway.

July 20: Allan sent INP miniprep 2 (in pSB1AK3) off to be sequenced. The samples are labelled simply F and R.

July 22: The INP sequence is a spot on match to what's expected. The part came in [http://dspace.mit.edu/bitstream/handle/1721.1/45139/BBFRFC12.txt?sequence=1 RFC12] format and so has a non-standard prefix and suffix. The actual suffix contains an extra base that it shouldn't. This information was entered into the relevant [http://partsregistry.org/Part:BBa_K265008:Experience experience page] on Partsregistry.org.

July 25: Attending the UK meetup in Norwich. About this time, a whole bunch of primers were ordered, for pVIII, cellulases, and xylose isomerase.

July 27: Lee and Sylvia did PCR for:

• The pVIII leader sequence
• The pVIII main sequence
• The xylA (xylose isomerase) gene

July 28: The gel from yesterday showed bands of about the correct sizes for the two pVIII sequences. The xylose isomerase however is absent and is being redone at a lower annealing temperature of 50 C.

This worked.

Sylvia and Lee began work to add a Plac-RBS-LacZ construct (<partinfo>BBa_J33207</partinfo> [Update 9 Aug: it seems <partinfo>BBa_K523000</partinfo> was used]) upstream of the RBS-malS BioBrick, and a Plac-RBS-LacZ-RBS construct (<partinfo>BBa_J33207</partinfo>+<partinfo>BBa_J15001</partinfo>) upstream of the (no RBS) malS BioBrick, in both cases using [http://partsregistry.org/Assembly:Standard_assembly standard assembly].

Successful clones, having gained the promoter (and RBS in case 2) should thus be blue.

In other news, Chris informs us that the lac promoter is already extant in the lab as a BioSandwich part, which will save us the trouble of creating it.

July 29: We had some starch agar helpfully made by the 6th floor lab tech people. There's now a bunch of plates of the stuff (with cml + IPTG) in the cold room. The recipe calls for:

• 1% starch
• Half-strength nutrient agar

Also, the following PCR reactions were done:

• INP to add BglII and SpeI sites
• Exoglucanase to add BglII and SpeI sites
• β-glucosidase to add BglII and SpeI sites

And transformations were done from the previous day's assembly.

August 1: The gels for the cellulases were a mess (perhaps because of annealing temperatures) so PCR is being redone. The PCR reaction for the INP was done seperately and the gel looks good.

MABEL is being done on the Endoglucanase plasmid to remove the BglII site.

Four colonies of the Plac-RBS-LacZ-RBS-MalS[f1r1] successful transformations are being cultured overnight for minipreps. So are three colonies (numbers 2-4) of Plac-RBS-LacZ-RBS-MalS[f2r2], though this has no stop codon and so seems of dubious value.

We decided to streak the four [f1r1] colonies onto starch also, so we could assay for starch degradation tomorrow; our first real experiment...

August 2: Minipreps are being made of the Plac-malS cultures. The starch plate has growth on it but could probably do with a bit more time before the iodine-starch assay is carried out.

The MABEL'd endoglucanase plasmid has been transformed into cells.

Preparation is in progress for a real BioSandwich attempt to construct an INP-malS fusion.

First the parts will have to be digested with the appropriate enzymes, as described on the BioSandwich page. Then spacers will have to be annealed as follows:

• Vector
  • Upstream spoligos:
    • RBS-2 FS
    • RBS-2 RL
  • Downstream spoligos:
    • Start FS
    • Start RT
• Plac-LacZ
  • Upstream spoligos:
    • Start FS
    • Start RL
  • Downstream spoligos:
    • RBS-1 FS
    • RBS-1 RT
• INP
  • Upstream spoligos:
    • RBS-1 FS
    • RBS-1 RL
  • Downstream spoligos:
    • Linker FS
    • Linker RT
• malS [f2/r2]
  • Upstream spoligos:
    • Linker FS
    • Linker RL
  • Downstream spoligos:
    • RBS-2 FS
    • RBS-2 RT

Note that the RBS-2 spacer is not actually being used for its RBS, but just because its the only spacer we have left.

The final construct would be:

• Vector - (spacer) - PlacLacZ - (RBS) - INP - (Linker) - malS - (spacer)

August 3: Some Plac-LacZ-malS minipreps were sent for sequencing. [Update: these were Plac-LacZ-RBS-malS(f2r2) colonies 2 and 3; in neither case had standard assembly worked, see August 5]

The BioSandwich annealments outlined above were set up. Ligation proceeded overnight.

August 4: The 4 colonies of (putative, awaiting sequencing) <partinfo>BBa_K523006</partinfo> were tested for starch degradation by flooding their starch agar plate with iodine. None showed a zone of clearing.

The BioSandwich part/spoligo fragments have been purified. A gel was run to check for removal of excess oligos, but in fact everything had disappeared. [Update: in fact the quantity of DNA present is expected to be too low to see, so this gel is meaningless. In all probability the ligation and purification worked.]

August 5: The sequencing results for clones 2 and 3 of our attempt to make <partinfo>BBa_K523007</partinfo> show that assembly failed for those clones; the plasmids only contain Plac-RBS-LacZ-RBS (i.e. <partinfo>BBa_K523005</partinfo>).

We accidentally never got our attempts at Plac-RBS-LacZ-RBS-malS[f1r1] (i.e. <partinfo>BBa_K523006</partinfo>) sequenced. This is the one that actually matters for control purposes, since it contains the malS gene with a proper stop codon. All four minipreps of this have now been sent off for sequencing.

Lee and Sylvia continue to work on BioSandwich. PCR was done for spacer1-INP-linker1 and linker1-malS-spacer2.

Chris informs us that the Ligation Independent Cloning version of BioSandwich assembly has not been working well; in 4 attempts there has been one successful colony. However, the OEPCR version of the protocol seems to work with some reasonable success rate ("thousands of colonies") though a significant number of these are not the desired construct.

August 6: Sylvia and Lee transformed INP+pSB1C3 into cells. Colonies from cex+pSB1C3, cfbglu+pSB1C3 and xylA+pSB1C3 were replated into chloramphenicol plates.

Gel electrophoresis was performed to confirm the PCR product of spacer1-INP-linker1 and linker1-malS-spacer2. However, the expected bands were not seen for both samples.

August 7: Lee and Sylvia replated INP+pSB1C3 onto fresh chloramphenicol plates. We also made overnight culture for cex+pSB1C3, cfbglu+pSB1C3 and xylA+pSB1C3 as preparation for miniprep.

August 8: Sylvia and Lee continue with miniprep of cex+pSB1C3,cfbglu+pSB1C3 and xylA+pSB1C3 and sent them for sequencing.

The MABEL site-specific mutagenesis is causing problems; no colonies were seen after transformation. We realised this was because the plasmid encodes kanamycin and not chloramphenicol resistance.

The modified INP (with BglII site and terminal "gg" bases, <partinfo>BBa_K523008</partinfo>) has been successfully cloned into pSB1C3, as far as we know, and is being miniprepped.

In case BioSandwich doesn't work well enough, primers for a hypothetical straight-up Gibson assembly of the various INP constructs have been designed. INP and malS have been ligated to oligos and the both ligations are used as templates for PCR and the PCR products were purified. The purified PCR product will be checked using gel electrophoresis the following day.

We used cen A mut miniprep as template for PCR, ran a gel to confirm that the PCR product was of expected size and purified the PCR product.

August 9: All of the putative Plac-LacZ-RBS-malS[f1/r1] sequences appear to be simply <partinfo>BBa_K523000</partinfo>, our BglII-containing cloning vector. How is this possible? Presumably that was used as the Plac-LacZ source for the assembly. Anyway, it's failed.

August 10: Sequencing results came back for:

• our modified (BioSandwich ready) C. fimi β-glucosidase
• our modified (BioSandwich ready) C. fimi exoglucanase
• our modified (BioSandwich ready) xylose isomerase xylA

For the β-glucosidase, there seems to be a frameshift in both clones that were sequenced (at the same place in both). [Update: after some discussion we believe the 2nd ATG in the gene is the true start codon. But this sequence will still not be usable for fusion proteins at its N terminal.]

The sequences of clone 1 of the exoglucanase are bad. But clone 2 looks generally good but there are a couple of base changes, one of which is not synonymous (aspartic acid to glycine).

Both sequences of xylA are bad.

August 12: Sequencing came back for what was supposed to be a colony of our modified INP, but it was not present in the plasmid. Lee and Sylvia now plan to make minipreps of more white colonies from that transformation.

In case the worst comes to the worst and we need to start from scratch with INP, Allan recultured some colonies from the original pSB1AK3-K265008 plate and placed them in the incubator.

Chris reports that the Gibson method of assembling BioSandwich parts is working about as well as OEPCR - that is, many correct colonies but also many incorrect ones.

August 15: Allan's reculture of pSB1AK3 (from the cold room plate) produced only a single colony; hopefully it is the right thing. It was restreaked onto a new plate and placed in the incubator.

Lee and Sylvia continue the main lab work.

August 17: The MABEL on the endoglucanase has somehow given us a plasmid that contains only <partinfo>BBa_J33207</partinfo>.

We also received many sequencing results from BioSandwich tests... these colonies are supposed to code for LacZ and EYFP and so be blue but also fluoresce yellow.

Note that after spacer2 we expect to see some extra bases before we reach the RFC10 suffix. I shall call these extra bases the "stuffing". So, the expected sequence is:

start1 -- PlacLacZ -- spacer1 -- EYFP -- spacer2 -- stuffing

BioSandwich with Gibson:

• gby [blue, yellow] - expected scar is missing at approx base 680 of the forward chromatogram. Also, the spacer ends with ATG and the part starts with ATG, but only one ATG is seen. In total 9 bases are missing. There is homology of "tatgg" at both ends of the deleted fragment. There is a point mutation at approx base 500 of the forward chromatogram.
• gb [blue] - there is a deletion of about 80 bases at approx base 700 of the chromatogram. This has been caused by homology of a "gggcgagg" found at both ends of the deleted sequence. The 9 bases mentioned for gby are also missing.
• gwy [yellow] - seem correct except for a point mutation in the reverse chromatogram. The mutation is synonymous (ctg -> ttg).

BioSandwich with OEPCR:

• bby [blue, yellow] - mostly correct but the EYFP is followed not by spacer2 but by an RFC54 suffix, after which come a few mystery bases and the stuffing. [See update, below]
• bb [blue] - has a massive deletion (several hundred bases) in the EYFP gene after about 60 bases. There is partial homology between the start and end of the deletion: "tggtcgagctggac" vs "tggacgagctgtac". In addition to this, EYFP is not followed by spacer2 but instead by an RFC54 suffix (i.e. minus NotI). After this comes a few mystery bases, then the stuffing.
• bw [plain] - has start1 joined to spacer1, then ~60 bases of the start of EYFP, the same massive deletion as bb, ~20 bases of the end of EYFP, an RFC54 suffix (i.e. minus NotI), then the stuffing.

BioSandwich with Blunt-End LIC (expected to fail):

• bb1 [blue] - seems to be BBa_J33207 followed by part of E. coli ferrous iron transport gene A.
• bw2 [plain] - has start1 followed by the stuffing.
• bw3 [plain] - has RFC10 prefix followed by RFC10 suffix, as if XbaI and SpeI sites have joined in a normal vector.

[Update 18 August] Chris informs us that the primers used for the OEPCR were not the normal ones, and in fact the expected product lacks spacer2 and has an RFC54 suffix.

August 18: We received results from some BioSandwich with CPEC tests...

• cby [blue, yellow] - there is a single-base deletion near the start of the PlacLacZ part, with no obvious cause. It is too early to affect expression. The last few bases of spacer1 are missing, as well as the first few bases of the EYFP part (9 bases total). This is explained by homology of "tatgg". Otherwise seems correct.
• cb [blue] - Normal until the end of spacer1, which is followed immediately by spacer2 and the stuffing.
• cw [plain] - start1 is followed immediately by a second copy of start1, after which comes spacer2 and a second copy of spacer2. Then the stuffing.